View original document

The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.

U.S. DEPARTMENT OF COMMERCE
Charles Sawyer, Secretary
BUREAU OF T H E CENSUS

J. C. C a p t , Director

BUREAUOFTHECENSUSMANUAI, OF

TABULARPRESENTATION
An outline of theory a n d practice in t h e presentation
of statistical d a t a in tables for publication

P r e p a r e d by

BRUCE L. JENKINSON
Chief, S t a t i s t i c a l A b s t r a c t U n i t

U N I T E D ST ATES
GOVERNMENT P R I N T I N G O F F I C E
WASHINGTON : 1949

F o r sale by t h e S uperintendent of D o c u m e n t s , U. S, G o v e r n m e n t P r i n t i n g Office, Washington 25, D. C.
P r i c e $1.50 B u c k r a m


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

BUREAU OF THE CENSUS
J.

C.

CAPT,

Director

A. Ross Eckler, Deputy Director
Howard C. Grieves, Assistant Director
Morris H. Hansen, Assistant Director for Statistical Standards
Robert Y. Phillips, Executive Assistant to the Director

ACKNOWLEDGMENTS
This manual constitutes a special report of t h e Bureau of t h e Census. T h e
preparation of the volume was undertaken at t h e suggestion of Philip M. Hauser,
former Deputy Director, who established t h e basic objectives and reviewed t h e
initial detailed outline.
In the planning and preparation of this report, Mr. Jenkinson was assisted by
Dorothy M. Belzer. Her constructive suggestions, and her translation of general
concepts into practical reality, have left their imprint throughout the volume.
T h e final manuscript was reviewed by the Bureau of the Census Committee
on Presentation and Publication Practices. Leon E. Truesdell, Chief Demographer, and Calvert L. Dedrick, Coordinator, International Statistics, members
of t h e committee, gave t h e manuscript a particularly detailed examination.
Their many comments, suggestions, and criticisms have contributed greatly to
improve the content of the report. Dr. Truesdell also contributed heavily in
another respect in t h a t the author, during the course of the Sixteenth Decennial
Census, worked under his supervision and profited immeasurably from t h a t
experience.
Finally, the manuscript was reviewed by all subject-matter divisions of the
Bureau of the Census. Especially helpful were the suggestions of R a y Hurley,
Chief, Agriculture Division, and his professional staff. Morris B. Ullman provided invaluable detailed comments which were incorporated, in large part, in
the final report. Acknowledgment also should be made of the aid of many other
members of the professional and clerical staff of the Bureau of the Census who
contributed their suggestions and criticisms.
June 1949.

Brief citation
U. S. Bureau of the Census. Bureau of the Census Manual of Tabular
Presentation, by Bruce L. Jenkinson. U. S. Government Printing
Office, Washington, D. C., 1949.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

FOREWORD
The interest of the Bureau of the Census in the manner in which
statistical data should be arranged and described is a reflection of its
own day-to-day operations. Its publications, whether issued as
formal reports, monographs, bulletins, or releases, constitute official
sources of information in many subject fields. The primary purpose
of these publications is to provide reliable and needed information to
the users of statistics-a purpose that can best be served if the tabular materials are made easy to read and to understand. In itself,
this places upon the Bureau of the Census a serious responsibility for
the development and maintenance of sound principles and standards
in tabular presentation.
This manual is designed as a reference aid for use by analysts and
technicians of the Bureau of the Census in the continuing effort to
meet the obligations of the Bureau in this field. It is intended as an
operating tool, not as a book of regulations. Absolute uniformity in
presentation of the Bureau's statistics is not contemplated. However,
unnecessary variations can be most easily avoided by reference to a
commonly accepted norm. Within obvious limits, this manual is
intended to provide that norm.
In general, this volume is intended to lead the analyst or technician
toward the solution of his particular problem in table design and preparation. In some cases, it may be found that the manual answers
the specific question. In other cases, it may prove desirable to adapt
the manual example to the specific need, a course that is entirely appropriate as long as the basic principles involved are taken into account.
In the final analysis, there are only two rules in tabular presentation
that should_be applied rigidly: First, the use of common sense when
planning a table, and second, the viewing of the proposed table from
the standpoint of the user. The details of mechanical arrangement
must be governed by a single objective; that is, to make the statistical
table as easy to read and to understand as the nature of the material
will permit.
J. C. C A P T , Director,
Bureau of the Census.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

3221

.

"Statistics are far from being t h e barren array of figures
ingeniously a n d laboriously combined into columns a n d
tables, which many persons are a p t to suppose them. They
constitute rather t h e ledger of a nation, in which, like t h e
merchant in his books, t h e citizen can read, at one view, all
of t h e results of a year, or of a period of years, as compared
with other periods, and deduce t h e profit or the loss which
has been made, in morals, education, wealth or power."

- J . D. B. DEBow,
Superintendent of the Census, 1853-1854.
IV


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

CONTENTS
Page

Introduction_

lX
CHAPTER

1.

C L A S S E S OF T A B U L A R P R E S E N T A T I O N

Paragraph

1-B

General classes_
Leader work_
The text tabulation_The formal statistical table_
CHAPTER

2-A
2-B

2.

101-102
111-114
121-125
131-133

T H E FORMAL T A B L E : STRUCTURAL P A R T S AND T Y P E S

Structural parts of the table_
Structural types of tables __

201-208
211-226

3-B
3-C
3-D
3-E
3-F
3-G

C H A P T E R 3. T H E T A B L E
General_
Methods of title writing_
Universe segment__
Classification segment __
Area or space segment_
Time-reference segment_
Title description of percent data_-

4-A
4-B
4-C
4-D

General_Numbering standard-series tables_
Use of prefix letters_
Use of suffix letters_

5-A
5-C

C H A P T E R 5. T H E
General_
Specific use of the headnote_Wording of the headnote __

6-A
6-B
6-C
6-D
6-E
6-F
6-G

General_
The specific footnote __
T h e source note_
Reference symbols: Type, placement, and sequence __
Reference range of symbols_
Placement and arrangement of footnotes_
Wording of t h e footnote __

3-A

CHAPTER

4.

CHAPTER

CHAPTER

7-B

7.

301-309
311-314
321-325
331-336
341-352
361-368
371-374

THE TABLE NUMBER

6.

401-407
411-414
421-424
431-435
HEADNOTE

501-506
511-513
521-523

T H E FOOTNOTE

601-606
611-615
621-624
631-640
641-643
651-656
661-665

A L L O C A T I O N TO S T U B A N D B O X H E A D

GeneraL
Allocation principles and problems __


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

TITLE

701-703
711-713
V

CONTENTS
CHAPTER

8.

T H E STUB: G E N E R A L AND DEFINITIONS
Paragraph

8-A
8-B
8-C

GeneraL
Definitions of component parts_
Standard s t u b arrangements_
CHAPTER

9-A
9-B
9-C
9-D
9-E
9-F

9.

801-803
811-821
831-835

T H E STUB: H E A D S AND CAPTIONS

T h e stub head or box__
Center heads a n d subheads_
Colon and dash (read-in) lines_
The line caption: T h e data line_
T h e line caption: Total and subtotal lines_
T h e unit-indicator_
CHAPTER

10-A
10-B
10-C
10-D

10.

T H E S T U B : C O N S T R U C T I O N A I D S TO C L A R I T Y

S t u b indention_
Leaders_
Space breaks: Analytical and reader_
Use of boldface and italic type__

CHAPTER

11.

1001-1010
1011-1018
1021-1030
1031-1033

T H E S T U B : STANDARD LISTINGS AND I N T E R - S T U B CLARIFICATION

11-A
11-B
11-C
11-D

Listing a r e a s Listing time periods __
Quantitative and qualitative classifications_
Clarifying relationships between stubs of varying detail_-

12-A
12-B
12-C
12-D
12-E
12-F
12-G

General_

CHAPTER

12.

column head: Total a n d subtotal column_
dash (or read-in) head_
spanner h e a d _
banner h e a d _
undercut spanner_CHAPTER

13.

1101-1118
1121-1124
1131-1133
1141-1144

T H E BOXHEAD

T h e c o l u m n h e a d or c a p t i o n _

The
The
The
The
The

901-909
911-920
921-933
941-948
951-960
961-966

1201-1211
1221-1227
1231-1233
1241-1248
1251-1255
1261-1266
1271-1272

H O R I Z O N T A L AND V E R T I C A L R U L E S

13-A
13-B
13-C

General_
T h e horizontal rule_
T h e vertical rule_

14-A
14-B
14-C
14-D
14-E

General_
T h e celL
T h e unit-indicator_
T h e field spanner_Braces_-

1301-1302
1311-1316
1321-1334

CHAPTER

14.

T H E FIELD

1401-1402
1411-1414
1421-1426
1431-1434
1441-1448
INDEX
Page

Alphabetical index_
VI


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

259

CONTENTS
Figure

1.
2.
3.
4.
5.
6.

7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.

Page

FIGURES

2
10
13
14
15
17
18
20
22
34
57

Examples of leader work_T h e formal statistical table and its major p a r t s _
Bureau of the Census page sizes_
T h e fractional-measure table_
T h e narrow table_
T h e narrow-divide table_
Multipage narrow table with continuous boxhead_
The broad table_
T h e parallel table_Table titles: Capitalization a n d indent_Experiment in formal terminology for percent tables__
Numbering sequence for footnote reference symbols in boxheads
of statistical tables_
Range of reference of footnote symbolsLetterpress and typewriter-offset combinations for center heads in
stub
Colon and dash lines: Their use and misuse_
Indentions, with a n d without conflicts at various levels_Stubs for regions, divisions, and States_
Boxheads and their component p a r t s _
Use of abbreviations to decrease depth of boxhead_
Placement of bottom rule_
Allocation of white space to tables_


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

85
88
115
121
143
160
184
190
227
234
VII


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

.

.

INTRODUCTION
Purpose and nature of manual.-This manual is devoted largely to
an outline of theory and practice in the presentation of statistical
data in tables for publication, illustrated by specific examples throughout. The emphasis is placed upon principles, rather than rules, in
order to provide the necessary flexibility required by the varying needs
of the several subject divisions of the Bureau. The immediate objectives are as follows:
1. To provide an operating reference manual for the use of
analysts and technicians within the Bureau of the Census in connection with the planning and review of statistical tables which are
to appear in census publications.
2. To provide a general statement in this field, with specific
reference to the standards and practices of the Bureau of the Census,
for the convenience of other producers of statistics, of technicians,
and of students of the subject.
Organization of volume.-Designed as a reference volume, this
manual has been organized so as to make each major section as nearly
independent as possible. In consequence, a certain amount of repetition will be found, particularly in the case of definitions.
The basic unit of the volume is the numbered item or "paragraph."
These paragraph numbers are keyed to the chapter numbers. Thus,
paragraphs 101 to 133 comprise chapter 1, paragraphs 201 to 226
comprise chapter 2, etc. It follows that, while the numbering is not
continuous, no paragraph number is repeated. This organization and
reference principle has been carried also into the table of contents and
the alphabetical subject-index in that they have been keyed to paragraph numbers, rather than to page numbers.
Selection of materials.-The field of tabular presentation is characterized by diversity of practice, particularly in matters of detail.
The methods described here comprise a selection of those in common
use which have proved satisfactory in the experience of the Bureau of
the Census. This does not mean, however, that they constitute the
only correct methods, or even necessarily the "best" methods, since
viewpoints and needs differ broadly in this field. Detailed examples
have been labelled as "right," "wrong," etc., primarily in terms of
Bureau of the Census practice. More general application of these
labels must rest upon the reasons assigned and the merits of the
individual case.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

IX

INTRODUCTION

The material included represents, to a limited extent, a codification
of existing practices 1 common to all Divisions of the Bureau of the
Census, selection where present divisional practices diverge, and a
setting-down of general principles which all of the Bureau's statistical
publications should reflect. The practices recommended conform also
to the requirement that tabular presentation should be of the highest
quality consistent with the demands of mass production.
Applicability for general use.-Throughout this volume, the
emphasis is on the mass-production situation found in the Bureau of
the Census. Therefore, the problems selected for detailed discussion
will not always reflect the general experience outside the Bureau.
However, in most instances, the basic principles laid down will be
found applicable in general statistical practice.
The basic principles of tabular presentation are the same irrespective of subject matter. Hence, the frequent use of age, race,
sex, and marital status for illustrative purposes does not restrict
the applicability of t h e manual to those and related subjects. Where
problems peculiar to a given subject or subjects are discussed, the
subject content of the examples has been adapted accordingly. Considerations involved in the standardization of subject matter in the
examples, and in the selection of age, race, sex, and marital status
for the purpose, are as follows:
1. The purpose of the examples is to illustrate points in presentation. The shift from one subject to another would tend to focus
attention on the differences between subject material, example
to example, rather than on the presentation points involved.
2. The decision to illustrate the manual by examples on each
point, run within the text, rather than to use full-page tables illustrative of many points simultaneously, placed a premium on the
selection of subject material which could be compressed within
a small amount of space. Both the brevity of the individual terms
and the shortness of the complete lists of items (such as male and
female, under 45 years old and 45 and over) made the classifications
of age, race, sex, and marital status suitable to the purpose.
3. Since the manual will be used by analysts and technicians
in many subject fields, it seemed important that no user should
be required to have a specialized knowledge of a given subject
in order to understand the analytical relationship of one line or
column to another. Again, the everyday concepts of age, race,
sex, and marital status seemed appropriate.
1 An important departure from past Bureau practice provided in this manual is the use of lower case
(small letters) in stub boxes and in uppermost spanners of tabular boxheads (see par. 1210). Traditionally,
the Bureau of the Census has used either capital letters or "small caps" for such captions. T h e new style
was adopted February 4, 1948. T h e old style will be continued for some time in the Statistical Abstract
of the United States, which is printed, in large part, from standing type.

X


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

INTRODUCTION

General Problem of Tabular Presentation
Tabular presentation as an end-product.-Since it represents the
culmination of a long series of census or survey operations, tabular
presentation can only reflect the results of the previous stages. Even
though planned in advance, statistical tables cannot be completed
until after the census or survey planning, questionnaire designing;
enumerating or transcribing, coding, card punching, and tabulating
operations have been completed. However, the published tables,
and their accompanying text, tend to be the only results of the expenditure of effort, time, and money which the using public sees; hence
the user's evaluation of all operations tends to be based on the final
step, that is, on the statistics as presented. In itself, this makes
sound presentation vitally important, Good presentation transmits
the quality of the previous operations and, in turn, assists in making
clear the meaning of the material. Poor presentation not only fails
to clarify the data, it tends to impair or destroy the effect of the
work that has gone before.
Basic objective.-In the design and preparation of statistical tables
the basic objective is to arrange and present data in such a way that
their meaning and significance can readily be grasped by the user.
In this respect, the statistical table represents a compromise, rather
than a solution. It is an efficient, but by no means an ideal, way of
presenting data, and its usefulness increases with multiplicity of
classification. It must be remembered, however, that many potential, and even actual, users of statistical tables do not find it easy to
understand columns and rows of figures. This is the group about
whom the table designer should be most concerned, rather than
about his more experienced colleagues. It follows that table design
should be kept as simple as possible, taking into account the subject
matter and available space. This need not penalize users who are
statisticians since they find simplicity in presentation a decided
convenience. It is extremely important, however, to those who
may have an equal need for the data but have difficulty in reading
tables of any kind.
The dividing line between good and bad presentation may best
be determined by the answer to the question: "Which method is
simple, focuses the reader's attention on the data (rather than on
the form), and makes clear the meaning and significance of the information?" That which clarifies the meaning of the statistics and
concentrates the attention upon them is good presentation; that
which obscures the meaning of the data or distracts the attention
from them is poor presentation.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

XI

INTRODUCTION

Fundamental approach.-Tabular design must be approached from
the standpoint of the user rather than from that of the producer.
It is not enough that the table be clear to the designer, who necessarily
has an intimate knowledge of the statistics and their qualifications.
The imperative requirement is that the table be clear to the user;
that is, the table must be so designed that the user will be able to
obtain from it that which he wishes and needs to know, so far as the
desired information is present. Conformity with this requirement
is not always simple since it is difficult for the table designer to discount his own knowledge and review his own work through the eyes
of the using public. Yet that task must be performed if the purpose
of the table is to be achieved.
Nature and significance of style. Style in tabular presentation
consists of the distinctive or characteristic mode of presentation or
construction determined upon as satisfactory by the producer of the
statistics. In general, it refers to the aggregate of practices and methods; in particular, it refers to the precise way of doing things prescribed by, or agreed upon within, a given organization. The significance of style lies primarily in its influence in simplifying
the production of statistics and in facilitating their understanding
when produced.
In terms of production, style consists of the great body of usual
practices common to any large organization and essential to any
large-scale operation. In this respect it is merely another name for
custom.
The role of style as an aid to the understanding of the data is frequently overlooked. By promoting uniformity in the treatment of
many details, it builds up an attitude of "normal expectation" on
the part of the reader. This tends to remove from the reader's notice
many points of minor detail which may distract him from the content of the table if they are forced on his attention by unaccountable
variation.
This development of "normal expectation" can be capitalized
upon by the producer of statistics. If "differences" which have
no meaning are scrupulously avoided, the reader is far more likely
to note and respect those real differences which can only find expression in the change of a single word in a phrase, or in a shift in the
terminal point of a column rule.
Style is rigid only in that it should be followed unless there is a
sound analytical or mechanical reason to the contrary. lt is not
properly a barrier to useful development and change. It is, and
should be, a barrier against unnecessary, unwarranted, or capricious
variation.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

INTRODUCTION

Comparative roles of content and mechanics.-The purely mechanical aspects of tabular presentation, although important, can be overemphasized. Mechanics must be consistently thought of as a tool of
presentation, not as an end in itself. Proper attention to mechanical
style and form is essential, but forcing statistics into a rigid physical
pattern, or an overreliance upon mechanical tricks, can easily defeat
the whole purpose of presentation.
In the last analysis the only purpose of mechanics in this field is to
help in making clear to the reader the significance of the d a t a presented; that is, the content of the table. T h e following statements
may aid in clarifying the comparative roles of content and mechanics:
1. T h e point of departure for both content and mechanics is the
meaning and significance of data.
2. Content is t h e reason for the table's existence; mechanics
merely translates requirements of content within physical limits.
3. Content determines inclusion and relative placement of d a t a ;
mechanics assists in presentation of content by means of typing,
drafting, and printing aids.
4. Content poses problems of space; t h e function of mechanics is
to solve them in a fashion which aids in, rather than detracts from,
the presentation of content.
5. Content is limited by space only in terms of t h e amount to be
presented, never in terms of t h e meaning of t h e data. Mechanics
is limited by space in terms first of preserving or enhancing meaning, and second by rigid physical and operational requirements.
6. Content is restricted by mechanics only so far as mechanical
limitations add to, or detract from, the presentation of meaning.
Mechanics is completely bound by content since a perfect physical
pattern fitting the space is useless if meaning is impaired or destroyed.
Effect of different reproduction processes on table design.Although the difference between letterpress and offset printing is basic
in terms of reproduction technique, it has comparatively little effect
on table design as such. In planning a table to be typewritten for
offset reproduction, however, the designer must not rely on italic,
bold face, or differences in type size, as such, to bring out relationships. T h e typewriter "substitutes" for these letterpress techniques
· are discussed at appropriate points in the detailed materials.
In some instances, final materials for offset reproduction consist of
tabulating-machine sheets. In most such cases, these sheets will have
been planned accordingly. When designing the sheets, the normal
principles of tabular presentation should be followed as far as is prac1 An exception is the instance of the use of special typewriters which provide the opportunity to use bold
face and italic type, as well as type of different sizes. In general, the same principles should be followed
in planning tables for such typing situations as for type-set composition.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

XIII

INTRODUCTION

ticable. Beyond that point, the obvious mechanical limitations of
the printer-tabulator must be taken into account and the detailed
rules of presentation modified as necessary.

Request for comment and criticism.S u g g e s t i o n s and comments
leading to the improvement of this manual will be appreciated. This
invitation is extended not only to analysts and technicians within the
Bureau of the Census, but also to those persons outside the Bureau
who may have occasion to refer to this volume with respect to their
own problems, and to all others who are in any way interested in this
subject. All suggestions received will be studied carefully, acknowledged, and will be taken into account in the event of a revision of this
manual. Finally, while the Bureau of the Census is unable to assure
complete and detailed answers to all questions relating to problems
of tabular presentation, as such, any inquiries submitted on this
subject will be answered as completely as limitations in time and
personnel permit.

SUGGESTIONS, COMMENTS, AND
INQUIRIES
s h o u l d be s e n t to

The Director
Bureau of the Census
Washington 25, D. C.

XIV


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Chapter 1
CLASSES OF TABULAR PRESENTATION (101-133)
Sec. 1-A. General Classes (101-102)

101. Tabular presentation defined.-Tabular presentation is a
means of bringing together and presenting related material or other
information in columns or rows. I t s object is to present in a concise
and orderly fashion information that could not be presented so clearly
in any other way. Since the tabular arrangement facilitates reference, comparison, and interpretation of the data, it is particularly
useful in presenting large masses of related statistics.
102. Threefold classification.-In terms of construction, tabular
presentation is largely restricted to three general classes of materials:
( a ) Leader work, (b) text tabulations, and (c) formal statistical tables.
Of these, the first two are largely used to illustrate text although both
are occasionally used in footnotes.
These three general classes are defined below. In addition, specific
descriptions of leader work and of the text tabulation are provided
in this chapter. (See secs. 1-B and 1-C, respectively.) These two
forms of tabular presentation are not discussed further in this volume.
In contrast, the formal statistical table is discussed here (see sec. 1-D)
in broad terms only, without reference to construction detail, since
the remaining chapters of this manual are taken up with discussion
of the formal statistical table as such.
Sec. 1 - B . Leader Work (111-114)

111. Definition.-Leader work is a form of tabular work, usually a
simple listing without title, boxhead, or rules. Ordinarily, it presents
only one or two columns of figures or indicates the relationship between two sets of facts or categories.
For examples of leader work in materials of the Bureau of the
Census, see figure 1, page 2. For other examples and details of
treatment, see GPO Style Manual, pages 135-137.
See United States Government Printing Office, Style Manual (revised edition, January 1945), p. 123.
This threefold classification reflects operating practices of the Bureau of the Census and other mass
producers of Government statistics. From a theoretical point of view, the position of the text tabulation
as a special class, and even its acceptability as a sound form of tabular presentation, may be open to question.
1
1


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

1

FIG.

1

CLASSES OF TABULAR PRESENTATION

SEC.

1-B

F I G U R E 1 . - E X A M P L E S OF L E A D E R W O R K

Example A. Classification listing Example 0. Conversion listing
within text:
wi thin text:
T h e classes of duration of unemOperation expenses for some of t h e
trades are analyzed in t h e summary ployment t h a t are used, and their
equivalents in terms of weeks as retabulation by size groups as follows:
ported, are listed below:
Business-size group
$2,000,000 and over
$1,000,000-$1,999,999
$500,000-$999,999
$300,000-$499,999
$200,000-$299,999
$100,000-$199,999

Group number

IV
VI

Duration in months

Less than 1 month_-1 month_2 months-3 months_4 and 5 months_6 to 8 months_-9 to 11 months_-12 to 23 months_-24 to 59 months_-60 or more months_--

Reported duration in
weeks
Under 3 weeks
3 to 6 weeks
7 to 10 weeks
11 to 14 weeks
15 to 23 weeks
24 to 35 weeks
36 to 49 weeks
50 to 99 weeks
100 to 249 weeks
250 or more weeks

Example B. Classification listing Example D. R e p r o d u c t i o n of
as a footnote to text or to a table:
schedule inquiries:
a The following is a complete list of the contiguous
States for each State:

State
Alabama.
Arizona_-

T h e inquiries for t h e latest census
read as follows:

Contiguous States
If you own all or part of this farm. Fla., Ga., Miss., Tenn .
Was there any mortgage debt on
Calif., Colo., Nev., N. Mex.,
the land and buildings so owned
Utah.
on Apr. 1, 1940? _
Arkansas_La., Miss., Mo., Okla.,
(Yes or No)
Tenn., Tex.
Total amount of outstanding
California.
- Ariz., Nev., Oreg.
Colorado ..
mortgage debt on such land and
- Ariz., Kans., Nebr., N.
buildings__
Mex., Okla., Utah, Wyo.
. Mass., R. I . , N. Y.
(Omit cents)
Connecticut_Md., N. J., Pa .
What was the annual rate (conDelaware_tract rate) of interest on the first
Dist. of Columbia--- Md., Va.
Florida_mortgage debt? (Report frac- Ala., Ga.
Georgia__- Ala., Fla., N. C s. C
tions)_-(Percent)
Tenn.

Example E. Statement of relative accuracy:
T h e following percentages give a rough measure of the maximum difference
t h a t would ever be expected between a detailed figure of t h e specified magnitude
a n d t h e corresponding figure t h a t would have been obtained from a complete
enumeration:
250,000 persons _
100,000 persons _
50,000 persons_-25,000 persons. __
10,000 persons_..5,000 persons __

2

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

2% percent
_ 4 percent
5)e percent
_ 8 percent
_ 12 percent
- 1 7 percent

sE0.

1-B

GEN'ERAL CLASSES

PAR.

112

112. General features.-Since leader work does not carry a table
num her or title, it cannot stand alone; hence it is used largely for
illustrative purposes within text. It is introduced by a text statement ending in a colon. Less formal than a table, leader work is an
integral part of the text and therefore complicates the make-up
problem; that is, its position in a text column cannot be shifted
readily to avoid breaking the presentation and running the remainder
at the top of the next column or page.
113. Footnotes. Leader work in text may be footnoted in the same
manner as the formal table. If there are two or more separate presentations of leader work within a single text, footnote reference
symbols begin again with "1" for each presentation. They are not
included in the same numbering series as the footnotes to text. Footnotes to leader work are placed at the bottom of leader work, whereas
text footnotes are placed at the bottom of the page. For treatment
where both tabular and text footnotes must appear at the bottom of
the same column (or page) of text, see section 6-F, paragraph 654.
114. Indexes and tables of contents: Special cases.-Indexes and
tables of contents constitute special cases in leader work. They are
set in the same style as text; other forms of leader work are governed
by the same rules as other tabular work. Indexes and tables of
contents are not discussed in this manual. For their treatment,
see GPO Style Manual, page 159.
Sec. l - C . The Text Tabulation (121-125)
121. Definition.-A text tabulation is a simple ruled table without
a table number or title, which usually appears as a part of the text of

a report or volume.* The conditions of its insertion in the text are the
same as for leader work. Its form, however, is the same as that of the
formal table, except that the table number and title are omitted and,
in consequence, any headnotes are converted into general or specific
footnotes.
122. General features.-Since it lacks an identifying table number
or title, the text tabulation cannot stand alone; it must be introduced
by an explanatory statement. Because of the disadvantages indicated
in paragraph 124, its usefulness for presenting statistics of primary
significance tends to be limited. The formal table is more adequate
for such a purpose.
In its simplest form, the text tabulation is introduced by a text
statement ending in a colon. As an integral part of the text, it com4

United States Government Printing Office, Style Manual (revised edition, January 1945).
The form of the text tabulation is also occasionally used in a tabular footnote.

776554

49

2


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

3

PAR.

122

sEc. 1-C

CLASSES OF TABULAR PRESENTATION

plicates printing (or reproduction typing) make-up in the same way as
does leader work. Specifically, it cannot be shifted away from its
introductory statement to avoid breaking over to the next column or
page. Footnotes are treated in the same manner as in leader work.
(See par. 113.)

Example A. Text tabulation with introductory statement:
T h e independent stores in this size group are further analyzed in the detailed
tables herein
. into five smaller size groups as follows:
Stores

Sales

Size group
Number
Independents under $10,000-.
$5,000 to $9,999--$3,000 to $4,999___
$2,000 to $2,999 ___
$1,000 to $1,999-Under $1,000---

Percent

Amount
(add 000)

Percent

943, 533

100. 0

$3, 739, 732

100. 0

330, 545
200, 272
117,342
145,630
149, 744

35. 0
21. 2
12. 5
15.4
15. 9

2,379,165
788, 994
279, 972
214, 043
77, 558

63. 6
21. 1
7. 5
5. 7
2. 1

123. Tendency toward overuse and misuse.-The text tabulation
should be used cautiously since it lends itself readily to overuse and
misuse. Two factors contributing to this tendency should be guarded
against:
a. T h e belief that a small text table is necessarily "overburdened" if given a
formal title and number. T h e criteria for use of t h e title should be t h e significance and purpose of the table, rather t h a n its size.
b. T h e ease with which any formal table in text can be made a text tabulation
by the omission of t h e title and number and conversion of the headnote, if any,
into a general or specific footnote.

124. Disadvantages to the user.-The form of the text tabulation
may cause· inconvenience to the user. This is particularly likely
when it is substituted for the formal table; that is, when it is used to
present summary or other highly significant data.
The figures shown in text tables, including those converted to text
tabulations by omission of table title and number, are usually those
judged by the analyst as being of primary importance. As such, they
are selected from the mass of statistics shown in detailed tables and
brought forward into the •text where their importance is further
emphasized by the text discussion. Such figures tend to be referred
to frequently by the user, are commonly cited by him, and are often
lifted bodily for incorporation in his own reports.
& T h i s difficulty m a y be overcome by assigning a table n u m b e r without a title, thereby transforming t h e
text tabulation into a poorly identified variation of the formal table.

4

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEc. 1 - C

THE

TEXT TABULATION

PAR.

124

The lack of a table title or table number makes the text tabulationa. Hard to find again when desired.
b. Difficult to cite, or to refer to, in another publication or report, whether of the
producer or user.
c. Awkward to extract bodily for presentation elsewhere because the user is
forced either to supply his own title or to reproduce also the text statements which
introduce t h e data. This problem arises also where a table is numbered but not
titled.

125. Release tables without titles: A special c a s e . - I n those
Bureau of Census releases and advance reports which contain only one
table, the table may sometimes be run without a title. This represents
a special case of the formal table. It is not a text tabulation as such.
This practice is satisfactory only if all of the following conditions are
present:
a. Only one table appears in the release;
b. The table title would be practically identical with the release title;
C. The text is brief and serves merely to announce the release of t h e statistics
and to introduce the table;
d. The basic purpose of the release is to present the statistics shown in t h a t
table; and
e. The entire release, including the entire table, will appear on not more t h a n
two pages.

Sec. 1 - D . The Formal Statistical Table (131-133)

131. Definition.-The formal table is the most finished form of
tabular presentation. Mechanically, it is distinguished by the presence of an individual table title and number, a formal boxhead, and
by the use of tabular rules. Analytically, it is designed to be as selfsufficient as possible. Dependence upon text explanation and discussion is replaced to a maximum extent by use of headnotes, footnotes, careful handling of stub and boxhead phrasing, and precision in
tabular arrangement.
132. General purpose and special purpose t a b l e s . - I n terms of
function and purpose, formal tables are commonly classified into two
major groups: (a) General purpose or reference tables and ( b ) special
purpose or analytical tables. This classification rests primarily upon
theoretical concepts. Few Bureau of Census tables fall strictly into
either group. The difference in concept is useful, however, because
the central idea, which is function and purpose, is paramount in
tabular presentation.
6 For a more detailed discussion, see Mudgett, Bruce D., Statistical Tables and Graphs, Houghton Mifflin
Co., New York City, 1930, p. 30-39, and Walker, Helen M., and Walter N. Durost, Statistical Tables-Their
Structure and Use, Teachers College, Columbia University, New York City, 1936, pp. 1-3.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

5

PAR.

132

CLASSES OF TABULAR PRESENTATION

sEc.

1-D

a. General purpose or reference t a b l e s . - T a b l e s in this group are sometimes
called source or repository tables. Typically, their function is merely to record
t h e results of a census, survey, or tabulation, with the materials arranged for
ready reference. Such a table is not designed to bring out a n y particular point
but serves merely as a repository of information of general interest. Users m a y
extract and rearrange portions of this source material, and develop their own
derived statistics, in accordance with their own special purposes or interests.
Although " t h e census table" is usually cited as the primary example for this
group, Bureau of Census tables actually depart from t h e pure type of generalpurpose table in t h a t analytical selection is involved in determining which of t h e
enumerated material is to be tabulated; which of t h e tabulated material is to be
presented; and how t h e material presented is to be arranged within tables, between tables, and within and among the various published volumes or reports.
Furthermore, the many derived figures which are presented (percentages, medians, averages, etc.) reflect a selective process involving a critical analysis. It
is broadly true, however, t h a t Bureau of Census tables are intended as repositories of basic information, arranged for ready reference, and designed to provide
material for specific analysis by users in terms of their own interests and needs.
b. Special purpose or analytical t a b l e s . - T a b l e s in this group illustrate or
demonstrate a specific analytical point or answer a specific question. Selection
of material is limited strictly to t h a t which bears on t h e problem at hand. Arrangement of data emphasizes those relationships pertinent to the problem and
subordinates those which are not.
Few Bureau of Census tables fall into this category except those (especially
text tables) published in monographs or other special studies. Summary tables,
although analytically invaluable, are not necessarily special purpose or analytical
tables of t h e type described here. This is particularly true of Census summary
tables since their purpose usually is to bring together major totals of general
purpose d a t a , rather t h a n to present selected data arranged to demonstrate a
particular point.

133. Complex and simple tables.-A clear distinction should be
drawn between complexity and confusion in table design. A confusing presentation can rarely be justified on t h e grounds that the materials are necessarily complex.
a. Complexity of meaning versus complexity in construction.-Complexity in
meaning of d a t a and complexity in table mechanics are not necessarily associated.
Some tables are simple in construction but t h e conceptual aspects of the d a t a
m a y be comparatively hard to grasp.
Example: Figures on births and deaths, by place of occurrence and by place
of residence, for cities of a given size group, may be presented in a 4-column
table with a city stub. Few tables are simpler in construction, yet t h e concepts a n d relationships involved may be confusing to the general reader. Tables
of this type sometimes include a headnote explaining how they should be read.
(See sec. 5 - B , par. 512f.) Again, data on State of birth classified by State of
residence can be shown in a simply constructed table, but such a table requires
close attention when interpreting t h e statistics.
Other tables are complex in construction but t h e d a t a and their relationships
are comparatively simple in meaning. This is t h e type of "complex" table most
frequently encountered in t h e work of t h e Bureau of the Census. It exists
largely because of the need to present t h e maximum quantity of statistics on t h e

6

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEc. 1-D

T H E FORMAL STATISTICAL TABLE

PAR.

133

minimum number of pages. For this reason a large number of small tables,
simple in construction, are consolidated into one large table in which t h e mechanics may become quite complicated.
b. Nature of the complex t a b l e . - T h e nature of t h e complex table, and its
relationship to the simple table, may be summarized as follows:
(1) The complex table is usually complex only in terms of arrangement,
rarely in terms of meaning.
(2) It is merely a collection of simple related tables brought together under
a common title.
(3) The degree of its apparent complexity is In direct ratio to t h e disparity
of relationship between its parts.
(4) When designing a complex table t h e first requirement is to make clear
t h e meaning of the individual component parts (simple tables) ; t h e second is
to make clear the exact relationship between them. If these needs can be
satisfied in such a way t h a t the results are readily grasped by the reader, t h e
table ceases to be complex.
(5) Overreliance upon mechanical tricks, or carelessness in construction,
can make even a simple table too complex for the user to follow.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

7

_

Chapter 2

T H E FORMAL TABLE: STRUCTURAL PARTS AND
TYPES (201-226)
Sec. 2-A.

Structural Parts of the Table (201-208)

201. General.-The statistical table is made up of a limited number
of structural parts, most of which appear in all formal tables. From
table to table the differences between similar parts are mainly the
variations required by subject matter and meaning.
T h e accompanying illustration (fig. 2) shows some of the major
structural parts of a statistical table in relationship one to another.
This table is a "narrow" table. (See sec. 2 - B , par. 216.) The running
head and folio (page number) are a function of the page and not of
the table, a distinction upon which too much stress cannot be laid.
202. The heading.-The portion of the table appearing above the
body; that is, above the top rule of the table. It comprises the table
number, title, and headnote. It does not include the folio line which
is a function of the page, not of the table. "Heading" is a general
classification term. Further discussion is confined to its component
parts, as follows:
a. Table n u m b e r . - A n indicator of relative position of t h e table within a series.
Normally, it is omitted in a table standing alone; t h a t is, when only one table
appears in the given publication. (See ch. 4.)
b. T i t l e . - A brief statement indicating t h e nature, classification, and time
reference of t h e data presented, and t h e political division or physical area or plant
to which t h e d a t a refer. It serves both as a catalog of content and a guide for
ready reference. (See ch. 3.)
c. H e a d n o t e . - A statement in brackets appearing below t h e title. It qualifies, explains, or provides information relating to t h e table as a whole, to an easily
identified major portion of t h e table, or to a constantly recurring specific term,
abbreviation, or situation. (See ch. 5.)

203. The s t u b . - T h a t portion of the table, usually at the left, devoted
to a listing of line or row captions or descriptions, together with
needed classifying and qualifying centered heads and subheads.
For discussion purposes, the stubhead is considered a part of the stub
although physically it is a part of the boxhead. (See chs. 8 to 11,
inclusive.)

8

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEc. 2-A

STRUCTURAL PARTS

PAR.

203

a. Stubhead or b o x . - T h e column head or caption of t h e stub. It describes
t h e stub listing as a whole. (See par. 811 and sec. 9-A.)
b. Center head or s u b h e a d . - A classifying, descriptive, or qualifying statement applying to all subheads and line captions below it until the next center
head of coordinate or superior classification is reached. (See par. 812 and sec.
9-B.)

c. Line caption.-The basic unit of t h e stub. The descriptive title of t h e
d a t a appearing on the given line. Special cases are colon and dash (read-in)
Jines. A line caption ending in a colon (:) is a heading that describes, and one
ending in a dash ( - ) is an integral part of, its subentries. (See pars. 813-815
and secs. 9 - C to 9 - E , inclusive.)
d. B l o c k . - A distinctive segment of the s t u b consisting of a group of related
line captions with their attendant heads and subheads; usually a self-contained
unit. A stub, irrespective of length, may consist of a single block; a series of
repeated blocks which are identical except for changes in heads; or a series of
independent nonrepetitive blocks radically different in content and coverage.
"Block" is a term of convenience. As such, it is not discussed further in this
manual. However, t h e term is used, on occasion, in the above sense.

204. The boxhead.-The portion of the table in which are located
the individual column beads or captions describing the data in each
vertical row or column, together with needed classifying and qualifying spanner heads. (See ch. 12.) Although the stubhead is
physically a part of the box, it is classified with the stub for discussion
purposes.
a. Column head or caption.-The basic unit of t h e boxhead. The descriptive
title for all d a t a appearing in the given column at the top of which t h e bead
appears. It may or may not be qualified, supplemented, or described by one or
more spanner heads above it. (See par. 1202b and secs. 1 2 - B to 1 2 - D , inclusive.)
b. Spanner h e a d . - A classifying, descriptive, or qualifying caption spreading
across (and above) one or more column heads, or across one or more lower spanners, and applying in varying degree to all columns or subordinate spanners thus
covered. The spanner head is the boxhead counterpart of s t u b center heads
and subheads and non-data line captions. (See par. 1202c and sec. 12-E.)
c. P a n e l . - A distinctive segment of the boxhead consisting of a group of related
column heads with their attendant spanners; frequently a self-contained unit.
(See par. 1202g.) The panel is t h e boxhead counterpart of t h e stub block.

205. The field. The portion of the table extending from the
bottom rule of the boxhead to the bottom rule of the table, and to
the right of the stub. The depository of statistical and other information. (See ch. 14.)
a. C e l l . - T h e basic unit of tabular presentation. The intersection of a n y line
caption with a n y column head. (See par. 1402a and sec. 14-B.)
b. L i n e . - A horizontal row of cells with a common classification, extending
across from a descriptive entry, or line caption, in t h e stub. (See par. 1402b.)
c. C o l u m n . - A vertical row of cells with a common classification, extending
down from a descriptive entry, or column head, in t h e box. (See par. 1402c.)


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

9

FIG.

2

sEc.

T H E FORMAL T A B L E : STRUCTURE

2-A

F I G U R E 2 . - T H E FORMAL T A B L E AND I T S M A J O R P A R T S

[See pars. 201-206]
T A B L E 6 . - A G E OF ALL

PERSONS AND OF C I T I Z E N S

BY S E X , FOR T H E U N I T E D STATES, U R B A N AND

HEADING<

RuRaL:

1940

[Age classification based on completed years]
All persons

Citizens

1

Area and age
Total

Male

Female

Total

Male

Fe-

-BOXHEAD

male

UNITED STATES
All ages ...............................

769

Under 5 years.........................
I 5 to 14 years.......... . ...
{ 15 to 24 years.........
, 25 to 34 years....... . ..............
I 35 to 44 years...... ............ ....
1 45 and over.......... .................

26
115
139
178
205
106

1

21 and over.....

988

475

513

567

302

265

URBAN
All ages ...............................

453

Under 5 years.........................
5 to 14 years............................
15 to 24 years.........................
I 25 to 34 years......................
35 to 44 years............
.............
45 and over...........................

15
73
86
104
116
59

21 and over..............................

328

S T U B <
1

I

RURAL

1

All ages ..............................

316

Under 5 years.........................
5 to 14 years...........................
15 to 24 years.......................
25 to 34 years.......................
35 to 44 years.......
45 and over............................

11
42

21 and over..........................
Includes both native and naturalized.

10


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

74
89
47
239

« F I E L D

sEc. 2-A

STRUCTURAL PARTS

FIG.

F I G U R E 2 . - T H E F O R M A L T A B L E A N D I T S M A J O R P A R T S-

2

Con.

[See pars. 201-206]
T A B L E N O . - T I T L E OF T A B L E
_[Headnote]

PANEL

Spanner head

Spanner head

The
column

1

Stubhead
IColumn Column Column' {Column! Column Column
I head
head , head
head
head
head

Total

CENTER HEAD

I

Total line caption.......

Line
Line
Line
Line
: Line
Line
1

BLOCK

caption
.
caption............................
caption
.
caption...........................
caption
.
caption..........................

L I N E » L i n e caption.........................

Cell

Cell

Cell

Cell

769

Cell
Cell
Cell
Cell
Cell
Cell

26
115
139
178
205
106

Cell

Cell

Cell

567

CENTER HEAD
Total line caption..............
,

Cell

453

caption............................
caption.................
caption.........................
caption..
caption
.
caption.....................

Cell
Cell
Cell
Cell
Cell
Cell

15
73
86
104
116
59

Line caption............................

Cell

328

Cell

316

Cell
Cell
Cell
Cell
Cell
Cell

11
42
53
74
89
47

Cell

239

Line
Line
Line
Line
Line
Line

CENTER HEAD
Total line caption..........
Line
Line
Line
Line
Line
Line

caption.......................
caption.......................
caption..................
caption...................
.
caption...........
caption............................

Line caption.......................
1

Footnote.

- » 2 1 and over..............................

988

475

513

567

302

265

B R I E F I N D E X TO MAJOR DISCUSSION IN T E X T
Heading:
Table number __
Table title_
Headnote __
Boxhead_
Spanner head_Column head __
Panel_Stub __
Line caption:
Total line_
Other __

Ch.4
- Ch. 3
_ Ch. 5
_ Ch. 12
Sec. 1 2 - E
Sec. 12-B
and 12-C
Par. 204c
Ch. 8 to 11


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Sec. 9 - E
Sec.
9-C
and 9-D

Stub- C o n t i n u e d
Stub block_
Stubhead _
Center head_
Field_
Cell_
Footnote_Column_
Line_
Horizontal
rules

and

Par. 203d
Sec. 9 - A
Sec. 9 - B
Ch. 14
Sec. 14-B
_ Ch. 6
Par. 1402c
Sec. 9 - D
vertical
_ Ch. 13

11

PAR.

206

T H E FORMAL TABLE ; STRUCTURE

SEC.

2-A

206. The footnote.-A statement qualifying or explaining the information presented in, or omitted from, a specific cell, column, line,
or group of columns or lines. The use of the reference symbol makes
clear the point or points of reference. (See ch. 6.)
In the form of a general note (preceded by "Note:" instead of a
reference symbol) it may, like the headnote, qualify the table as a
whole. Because of the difference in location, greater length of statement is possible in the general footnote than in the headnote.
207. The source n o t e . - A n exact citation of the source of the data
presented in the table. It may appear as a headnote or in the form
of a general note at the bottom of the table. In either case, it is
preceded by the word "Source" as "Source: Department of Commerce, Bureau of the Census; Sixteenth Census Reports, Population,
vol. I, p. 385." As a matter of convenience, the source citation is
discussed in the chapter devoted to the footnote. (See sec. 6-C.)
208. The folio l i n e . - A page characteristic in a formal report.
Although not a part of the table, it is referred to here because it is a
normal adjunct of the statistical table in its most formal setting. The
folio line usually comprises ( a ) the running head and (b) the folio
proper (page number).
Normally, where a running head is used, the folio (page number)
appears in the outer corner at the top of the page; the inner corner
is left blank. Also, in the normal case, the left-hand running head
is the title of the book and the right-hand running head is the title
of the chapter, each in brief. Another common practice is to have
the left-hand running head provide the chapter title; and to have the
right-hand provide a description of the page content.
This manual is unusual in that it carries a running head at the top
of the page and the folio (page number) at the bottom. The outer
corner of the running-head line, where the folio would normally appear, is taken up by the paragraph number of the first paragraph
appearing on the page. Again, in this manual, the left-hand
running head provides the chapter title in brief; the right-hand
running head gives the section title in brief.
Sec. 2-B. Structural Types of Tables (211-226)
211. Definition.-"Structural type" refers to the classification of
tables according to the physical combination of their major structural
parts; particularly, the boxhead, stub, and field.
The descriptive terms used in the classification (by major structural
t y p e - s e e par. 212) emphasize its mechanical nature. Thus, the
term measure is a printing term referring to the width of the type page

12

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

SEC.

2-B

STRUCTURAL TYPES

P AR.

211

or text column. The term divide means that the boxhead is split
because of space requirements and is continued to the next page, or
to a lower deck on the same page.
Available space is the determining factor; so much so, that a table
which would be one structural type on a large page, such as "census
size," might well become another structural type on a small page, such
as "document size." (See fig. 3 for book sizes.)
F I G U R E 3 . - B U R E A U OF T H E C E N S U S P A G E S I Z E S
["Wide document" should not be used without specific prior approval of the Graphics Section. "Census
size" type page, formerly 43 x 58 picas, was changed to 47½ x 60½ picas, as shown here, effective May 1,
1949; trim size remained unchanged]
Type page
Size

In picas
Width

Census size_Document size:
Standard_Wide_--

In elite typewriter

Depth 1

Width

Depth 1

Spaces

Lines
121

47%

60)9

190

26)9

46
46

106
112

28

2

Books trim to-

9S x 11% inches.

92 5J4 x 9% inches.
92 574 x 9§ inches.

1 Page depth as given must include running head if one is used.
If bottom folio is used, it is placed in
the bottom margin of the page.
2 Assumes a 50-percent reduction before printing.

212. Major types.-The major types of tables, according to structure, are as follows: (a) Fractional-measure, or double-up, of which the
half-measure (doubled) is the prototype; (b) narrow, the most common
form, and its subtype, the narrow-divide; (c) broad, or broadside, and
the broad-divide; and (d) parallel, or double-page spread, and the
parallel-divide. With the exception of the broad-divide and paralleldivide, all these forms are used frequently in Bureau of Census publications. An additional table type, (e) table without stub, represents
a special case rarely used in Bureau of Census work. Finally, (f)
composite forms may reflect combinations of features of various major
types of tables. (See figs. 4 to 9 for illustrations.)
213. Classification not based on number of pages. Except for the
parallel table, structural type has no effect on the number of pages
occupied by the table. Thus, fractional-measure and narrow tables
may each occupy a portion of a page, a full page, or a number of pages.
The broad table is most presentable in one or more full-page units, but
full-page units are not mandatory.
The parallel table,1 as such, always appears in pairs of facing pages;
therefore, it must always start on the left-hand (even-numbered) page.
1 See

pars. 221, 222, and 225.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

13

4

FIG.

FORMAL TABLE ; STRUCTURE

THE

FIGURE 4 . - T H E FRACTIONAL-MEASURE

SEC.

2-B

TABLE

[Titles, beadnotes, and footnotes omitted}

Example 4 - A . Half-measure, doubled, with continuous stub.

Kind of wood

Consumption (cords of
128 cubic feet)
1944

Total pulpwood_

Softwoods, total.
Spruce and fir
Hemlock ___
Pine_-Western fir
Western spruce_-

Consumption (cords of
128 cubic feet)

Kind of wood

1943

16,757,400

15,644,500

13,931,075
3, 290, 466
2,246,833
7, 994, 650
260,643
138,483

13, 477, 074
3,633,239
2, 081, 609
7, 343, 615
243, 891
174, 740

1944

Hardwoods, totaL _
Aspen ___
Yellow poplar ___
Gum--

Chestnut.Birch.
Other woods_
Saw mill waste.. _

1943

1,550, 117
661,710
284,882
160,166
434,982
8,377
988, 873
287,335

1, 287, 983
687,486
149,724
66,617
384,156
641, 838
237,605

Example 4-B. Half-measure, doubled, with repeated s t u b blocks.
Area and age

Total

Male

Female

The State __

499,261

258,170

53,133
100,801
92,205
71,303
65,285
116,634

26, 823
50, 969
46,236
41, 843
33,786
58, 513

173,981

86,050

87,931

14,838
29, 974
32,243
29,582
24,753
42,591

7, 444
15,076
15,393
14,565
12,014
21,558

7, 394
14. 898
16,850
15,017
12, 739
21,043

Under 5 years_5 to 14 years_15 to 24 years ___
25 to 34 years_-35 to 44 years__45 years and over_-Urban __

Under 5 years_5 to 14 y e a r s 15 to 24 years_-25 to 34 years __
35 to 44 years_45 years and o v e r - -

Area and age

Total

Male

241,091

Rural-nonfarm.

211,258

111,631

99,627

26,310
49,832
45, 969
39,460
31,399
58, 121

Under 5 years __
5 to 14 years_15 to 24 years_-25 to 34 years_..
35 to 44 years_.45 years and over_--

23, 726
42,915
38,089
36, 708
27,975
41,845

11,974
21,710
19, 336
19, 393
15,203
24,015

11, 752
21,205
18, 753
17,315
12,772
17,830

Rural-farm_-

114,022

60,489

53,533

Under 5 years_5 to 14 years_-15 to 24 years_.25 to 34 years-35 to 44 years.45 years nnc1over ___

14, 569
27,912

7,405
14, 183

21,873

11,507

15, 013
12,457
22, 198

7,885
6,569
12,940

7, 164
13, 729
10, 366
7,128
5,888
9. 258

Female

Example 4 - C . Third-measure, tripled.
Cost
Quantity, all Total Averspecies {1,000 age
(cords) do!
per
lars) cord

Year

1940 __
1939-1938 __
1937-1935--

13,742,958 109,740 $7. 99
10,811l,466 84. 539 7.82
9, 193, 991 74. 433 8 . 1 0
10,393,800 82,885 7. 97
7, 628, 274 58,244 7. 64

Year

Cost
Quantity, all Total Aver
species {1,000 age
(cords) dolper
Jars) cord

6,581,674 48,508
19336, 722, 767 73,524
1931-7,645,011 100,054
1929-1 9 2 7 - --I6, 750, 935 95,452
1 9 2 5 - _[6, 093, 821 94,340

$7. 37
10.94
13. 09
14.14
15. 48

Year

1923.19211919-1909.1S99--

Cost
Quantity, all Total Averspecies (1,000 age
(cords) dolper
lars) cor<l
5,872,870
4,557, 179
5,477,832
4,001,607
1,986,310

95, 306 $16. 23
91, 589 20.10
87,386 15. 95
34,478 8.62
9,838 4. 95

Example 4 - D . Fourth-measure, quadrupled.
Year

Tons
(tbou-

Year

sands)
1934 __
1933 __
1932
19311930 __

5, 583
4, 908
4,385
6,354
8,222

19291928-1927 -1926-1925 __

14

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Tons
(thousands)
8,012
7,986
6,844
7,329
7,334

Year

1924-1923-1922-19211920--

Tons
(tbousands)
6,826
6,445
5,672

4,863
7,177

Year

19191918-1917
1916-1915 __

Tons
(tbousands)
6,626
6,467
5, 926
5,125
5, 324

sEc. 2-B

STRUCTURAL TYPES

5

FIG.

FIGURE 5 . - T H E NARROW TABLE
[Titles, headnotes, and footnotes omitted]

Example 5 - A . Simple form; nonadditive box and stub.
City·
New York ___
Chicago_
Philadelphia_Detroit_-Los Angeles_Cleveland_Baltimore_St. Louis_-Boston_Pittsburgh_-

1940

1930

7,454,995
3,396,808
1,931,334
1,623,452
1,504,277
878, 336
859,100
816,048
770,816
671, 659

1920

1910

1890

1900

1880

1870

6,930,446 5,620,048 4,766,883 3, 437, 202 2, 507, 414 1,911,698
3,376,438 2, 701, 405 2,185,283 1,698,575 1,099,850
503,185
l, 950,061 1, 823, 779 1,549,008 1,293,697 1,046,964
847,170
l, 568,662
993,678
465,766
285,704
205,876
116,340
319,198
1,238,048
576, 673
102, 479
50,395
11,183
796,841
560,663
900,429
381, 768
261, 353
160, 146
733, 826
558,485
508,957
434, 439
322,313
804,874
772,897
687,029
575,238
821, 960
451,770
350,518
781, 188
748,060
670,585
448,477
560,892
362, 839
669,817
588,343
533,905
451,512
343,904
235,071

1,478,103
298,977
674, 022
79,577
5,728
92, 829
267,354
310,864
250,526
139, 256

Example 5 - B . Simple form; additive box and stub.

Division

United States_New England_-Middle Atlantic ___
East North Central_West North Central.-South Atlantic __
East South Central_West South Central_.M o u n t a i n ___
Pacific __

Total
to
d e r 3 0 30 to 49 50 to 99 100 to 179 180
n u m b e r U nacres
259
acres
acres
acres
of farms
acres

260 to
499
acres

500 to
999
acres

1,000
acres
and
over

6, 096, 799 1,519,373 767,289 1, 291, 048 1, 309, 741 486, 336 458,787 163, 694 100.531
135,190
348,100
1, 006, 095
1,090,574
1,019,451
1,023,349
964,370
233,497
276,173

41,674
84, 573
164,996
100, 331
319,359
379,072
250,663
47, 152
131,553

14,662
34, 650
104, 780
62,800
175, 986
181,360
138,939
18,756
35,356

29,442
91,677
268, 033
155, 998
253,090
232, 461
196,166
28,833
35, 348

28, 698 10, 976
7, 836
92,440 28, 138 14, 424
295, 624 106,289 58, 601
323, 965 156, 666 195, 183
164, 778 53, 270 37, 384
148, 731 52, 755 28,388
194,101 64, 961 71, 448
34, 440 12,738 30,663
26, 964 10, 543 14,890

1, 574
1,846
6,895
67,887
11,367
7,752
28,618
27,557
10,198

328
352
877
27,744
4,217
2,830
19, 474
33,388
11,321

Example 5 - C . Continuous stub, with panels.
Active proprietors and
firm members

Employees (average
for year)

Total pay roll (1,000
dollars)

K i n d of contractor
Total

Special trade contractors_
187,373
Air conditioning_671
Carpentering_32,718
Concreting ___
4, 944
Electrical (other t h a n
power lines)
16, 121
Elevator (installation and
289
repair)--Excavating a n d foundation ___
446
Excavating a n d grading __ 1,992
Flooring (wood only) __
1,197
Flooring (wood and other)_
Flooring (surfacing and
resurfacing)_-Heating and plumbing
group_-Insulation __
Lathing__.
Plastering a n d lathing_Plastering_
Ornamental iron_-Painting__Other ___

$25,000 Under
and
$25,000
over

Total

11,586 175, 787 452, 413

$25,000 Under
and
over $25,000

Total

245,117 207,296 607, 207

$25,000 Under
and
over $25,000

400, 371

206,836

1,063 10,715
25, 428 32,872
9,297 19,467

9,534
9, 315
12,247

1,181
23, 557
7,220

17,192

68,749

49,828

18,921

6, 320

696

15,138

14, 211

927

1, 692
3,457
1,388

1,134
4, 486
1,662

2, 932
8,017
4, 094

2,042
4, 075
2, 297

890
3,942
1,797
681

187
479
365

484
32, 239
4,579

6,754
31,282
18, 106

5,691
5,854
8,809

1,213

14, 908

42, 390

25, 198

53

236

7,016

69
182
92

377
1,810
1,105

2,826
7,943
3,050

402

105

297

3,280

2,652

628

4,811

4,130

706

8

698

828

146

682

853

228

625

38, 138
508
608
811
6,045
607
11,023
70, 147

3,897
106
49
254
411
85
213
3,818

34, 241
402
559
557
5,634
522
10,810
66,328

108,173
6,970
3,253
7,754
17,413
2, 622
13, 733
169,020

62,604
5,673
1,838
6, 164
8,500
1,839
4, 116
93,176

45, 569 146, 622 101,189
1, 297 10, 025
8,707
1,415
5,276
3,831
1, 590 11,686
9, 940
8, 913 21,625 12,831
783
3, 926
3,051
6, 729
9, 617 16,916
75, 844 223,483 146, 186

45,433
1,318
1, 445
1,746
8, 794
875
10,187
77, 297


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

15

FIG.

5

THE FORMAL TABLE ; STRUCTURE

Figure

5 . - T H E NARROw

SEC.

2-B

TABLE-Con.

[Titles, headnotes, and footnotes omitted]

Example 5 - D . With panels and blocks.
a n d with narrow-divide.)

(Compare with half-measure, doubled,

Excluding armed forces overseas

Including armed forces overseas

Age and sex
Total

White

132, 563, 271

118,602,260

34, 561, 078
60,141,378
27,985,586
9, 875, 229

30. 184, 483
53, 473, 624
25, 728, 512
9,215,641

4, 376, 595
6,667,754
2,257,074
659,588

34, 561, 078
65,618,360
28,046,207
9, 875, 229

63, 540, 744

56,862, 777

6, 677, 967

17, 568, 925
27,094,168
14,124,085
4, 753, 566

15,382,872
24, 080, 889
12,975,061
4, 423, 955

2, 186, 053
3,013,279
1,149,024
329, 611

69,022,527

61,739, 483

7, 283, 044

16,992,153
33,047,210
13,861,501
5,121,663

14,801,611
29,392, 735
12,753,451
4,791,686

2,190,542
3, 654, 475
1,108,050
329, 977

Nonwhite

Total

White

Nonwhite

TOTAL
All ages_
U n d e r 15 years_15 to 44 years __
45 to 64 years __
65 years a n d over_-

13, 961,011 138, 100, 874

123, 790, 994

14, 309,880

30,184,483
58,605,540
25,785,330
9, 215, 641

4, 376, 595
7,012,820
2, 260, 877
659,588

69, 047, 321

62, 020, 485

7, 026, 836

17,568,925
32,540,723
14,184,107
4,753,566

15, 382, 872
29,182,378
13,031,280
4, 423, 955

2,186,053
3, 358, 345
1,152,827
329, 611

69,053,553

61, 770, 509

7, 283,044

16,992, 153
33,077,637
13,862.100
5,121,663

14,801,611
29,423,162
12,754,050
4,791,686

2, 190, 542
3,654,475
1,108,050
329,977

MALE
All ages_
U n d e r 15 years __
15 to 44 years __
45 to 64 years_65 years and over_FEMALE
All ages_
Under 15 years __
15 to 44 years_45 to 64 years_65 years and over_-

(Compare with half-measure, doubled, and

Example 5 - E . With field spanners.
with narrow-divide.)
T o t a l - assets class
( t h o u s a n d s of
dollars)

Number of
returns

Com
Total T o t a l
piled
comassets
net
or liapiled
profit
bilities receipts or loss

Net
income
or
deficit

Number of
returns

Manufacturing
Total_U n d e r 50-50 to 9 9 - 100 to 249250 to 499--500 to 999-1,000 to 4,999-5,000 and over_--

Total Total
assets
com
or liapiled
bilities receipts

Public utilities

78,645 70, 070.8 91,606, 2 10, 309. 8 10,300.0

18, 405 58, 472. 2 15, 739. 2 1,918.0 1,916. 3

34,822
686.4 2,009.6
14.0
14.0
11,939
858. 3 2,085.2
67. 9
67.8
13,291 2,122.9 4,698.5
246.2
245. 9
7,175 2,529.8 5,124.9
357. 5
357. 2
4,814 3,377.5 6,299.7
541.5
540.8
4,961 10,454. 7 16,225.9 1,796.1 1,794.0
1,643 50,041.2 55, 162. 4 7, 286. 7 7,280.2

10,322
170. 2
361.0
5. 5
5.5
2,305
163. 3
245. 0
11.0
11.0
2,187
345. 5
424. 7
28. 0
28.0
1,115
392. 3
407.2
39. 2
39.1
783
558.1
414. 9
39. 6
39. 5
944 2,077.3 1,005.4
128. 5
128.3
749 54, 765. 4 12,880.9 1, 666. 4 1,665.1

Trade
Total_Under 5 0 - 50 to 99-100 to 249-250 to 499-500 to 999-1,000 to 4,999-5,000 a n d over ___

Com
Net
piled
income
net
or
profit
or loss deficit

Service

123, 439 22,133.9 56,512.4 2,070.9 2, 069. 0
76,085 1,452.8 5, 665. 8
19,542 1, 385. 1 4,664.8
16, 312 2,542.5 7,897.7
6,068 2,111.5 5, 946. 9
2, 870 1, 995. 0 5,179.1
2,133 4, 147. 1 10,401. 7
429 8, 499. 9 16, 756. 4

16

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

45. 7
96.8
220. 7
208.4
211. 6
454. 2
833. 7

45.6
96. 8
220.5
208. 2
211. 3
453. 7
832. 9

33, 296 4,366. 1 4, 029. 3
23, 833
3,897
3, 122
1, 106
684

352. 3
274. 8
483. 5
382.1
474. 9
582 1, 197. 5
72 1,201.0

898.8
421. 4
571.3
372. 7
376.1
759. 9
629. 1

188, 7

188. 1

6.2
15. 2
25. 6
21. 0
21. 9
54. 7

6.2
15. 1
25. 5
21.0
21. 9
54. 5
43.9

SEC.

2-B

STRUCTURAL TYPES

Figure

6

FIG.

6 . - T H E NARROW-DIVIDE TABLE
[Titles, headnotes, and footnotes omitted]

Example 6 - A . Repeated s t u b a n d continuous box.
with panels and blocks.)

(Compare with narrow table

Expenditures from own sources

Intergovernmental transfers

Function
Total
TotaL

Federal

State

Federal
to State

Federal
to local

State to
local

24,276

13, 878

4, 836

5,562

750

136

1,220
6, 685
1, 674

439
6, 685
205

185

596

4

6

1,002

467

184

18

341

1, 432
2, 986

1,288
2,094

89
548

55
344

25
331

14
1

1
407

1. 974
2, 603
1,678
4, 227

1,007
169
1,098
893

961
913
117
1,021

6
1, 521

66
101

463
2,313

30

39

67

General control_-National defense_-Highways and streets__Agriculture and natural resources_
Welfare_-Old-age and unemployment
insurance_Schools and libraries ___
Interest_All other_-

Intergovernmental
transfers-Con.
Function

Local

Total_-

9
252

Expenditures for own functions

Total
and
Local to State
local to
State
Federal Amount Percent

Federal

State

Local

24, 276

100. 0

13,027

3, 904

5. 0
27.6
6.9

429
6, 685
3

189

602

4

1,220
6,685
1,674

849

822

12

1, 432
2,986

5. 9
12.4

1, 249
1, 762

113
484

70
740

1,974
2,603
1,678
4, 024

8.1
10. 7
6.9
16. 6

941
38
1,110

1, 027
278
110
854

6
2, 287
58
2,360

69

General controL __
National defense_-Highways and streets_Agriculture and natural resources_Welfare_-Old-age and unemployment
insurance_Schools and libraries ___
Interest_All other_-

1, 746

36

12

48

Example 6 - B . Repeated box and s t u b with two decks.
table with field spanners.)

810

7, 345

(Compare with narrow
Urban

United States
Country of birth
Total
England_-Scotland_-Wales __
Northern Ireland __
Irish Free State (Eire)_--

621, 975
279,321
35,360
106, 416
572, 031

Percent
5. 4

2. 4
0.3
0.9
5. 0

Male

Female

Total

310,299
139,019
18,692
47,827
244, 092

311,676
140,302
16, 668
58,589
327,939

489,496
230, 296
26, 741
90,504
513,926

Percent
5.4

2. 5
0. 3
1.0
5.6

Male

Female

240,040
113, 426
13,848
40, 166
217,023

24.9, 456
116,870
12,893
50,338
296, 903

Rural-farm

Rural-nonfarm
Country of birth
Total
England_-Scotland ___
Wales_Northern Ireland.-Irish Free State (Eire)__-

99,311
37,901
6,250
11,400
43,672


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Percent
7. 2
2.8
0. 5
0.8
3. 2

Male

Female

Total

51,455
19, 423
3, 423
5,248
19, 443

47,856
18,478
2,827
6,242
24, 229

33,168
11,124
2, 369
4,422
14,433

Percent
3.6
1.2
0.3
0. 5
1. 6

Male

Female

18,804
6,170
1,421
2,413
7, 626

17

14, 364
4, 954
948
2, 009
6, 807

FIG.

7

THE

FORMAL T A B L E : STRUCTURE

SEC.

2-B

FIGURE 7 . - M U L T I P A G E NARROW, CONTINUOUS BOXHEAD
[Titles, headnotes, and footnotes omitted]

Example 7. Common form, with repeated stub (page 1 of 2). (Compare with
narrow-divide and parallel tables. A divide table with full-page decks.)

Division and State

All
Industries

Agriculture,
forestry,
and
fishery

United States __

34,027,905

7,988,343

2, 131,784
204,215
124, 072
96, 163
1,050,678
179, 082
477, 574

156, 358
38, 75ti
15,559
30,283
40, 153
5, 461
26, 146

308
1,435
1, 449
177
581

7,059,570
3, 521, 163
1.120, 137
2, 418, 270

442,137
206, 354
47, 488
188,295

N e w England__

Mining

Construction

Transportation,
commu- Whole
Manu nication,
sale and
facturand
retail
ing
other
trade
puhlic
utilities

902, 061 2, 022, 032 8,250,690 2, 768,267
153,089
15, 062
7, 602
6, 281
84,611
10, 844
28,149

382,588
31, 473
18,639
12,264
209,298
33,131
77,783

235,385
8,614
3, 461
223,310

453, 940 2,210,034
235, 763
968,453
413,381
70, 994
838,200
140,183

700,217
363,343
111,847
225, 027

1,320,225
739,295
202,725
378,205

7,078,388 1,207,887
252, 735
1,792,154
901, 55
201,228
2,127,498
279, 669
1, 427, 459
210,543
829, 672
263, 712

112,270
32,102
13,315
48,715
15,679
2, 459

373,856 2,412,359
99,325
652, 102
49, 315
284, 027
637, 213
114,598
613, 849
71,809
225,168
38,767

593,701
156, 869
72, 676
223,327
87,731
53, 098

1,142,140
287,597
134, 705
389,814
207,873
122,151

3,550,388 1,419,016
722,505
275, 987
692,431
303,619
988,641
298,820
105,027
166, 741
167,497
96,382
345,629
159, 532
466, 944
179, 649

45,919
7,840
6,313
12,326
972
2, 840
593
15, 035

179,516
36,828
35,592
55,649
4, 040
6,125
16,749
24, 533

439,238
95,017
82,197
179, 997
4,272
7,965
25,024
44, 766

287,661
58,038
50, 522
87,961
9,129
7,775
29,721
44, 51o

575,023
122,114
107,078
172,295
21,119
21,940
o5, 158
75, 319

4,578,782 1,426,249
76,200
13,784
509,990
70,471
189, 587
558
722,026
219,856
424,405
77,211
898,296
374,670
466, 773
208, 957
806,694
348, 529
484, 811
112,213

151,829
3, 994
123
24,055
112,773
2,871
1,315
4, 031
2, 568

286, 251
6, 847
38,723
19,379
47,944
20, 455
46,460
22,142
41,127
43,165

958,159
22,801
141, 781
19,210
145, 772
77,479
225, 490
110,706
148,407
66, 513

316, 154
7,380
52,544
17,798
58,873
36,342
37,478
17,609
46, 533
41, 597

610,421
10,614
82, 339
36,776
86, 668
45, 975
100,089
49,045
98,110
100,805

2,662,419 1,270,839
698, 571
303,512
299, 548
728,203
682, 366
311, 540
553, 279
356, 239

106, 464
60, 239
14,32ti
30,024
1, 875

129,917
35, 140
42,312
28,478
23,987

390,876
77,144
128, 981
129,767
54,984

150, 733
49, 029
36,641
35,240
19, 823

286, 298
78,584
94, 506
68, 041
45,107

W e s t South Central_-

3,279,365 1,329,991
485,846
283,740
587, 901
232,061
530,123
212, 690
1, 675, 495
601, 500

112,944
5,848
14,474
32, 954
59,668

187,488
17,069
34,512
26,517
109,390

371,096
53, 339
87,388
45, 854
184,515

224,338
22, 899
46,259
28,875
126, 305

504,168
49,108
83,323
84,571
287, 164

Mountain __

Montana_Idaho_
Wyoming_Colorado_
New Mexico_-Arizona_Utah_-Nevada_--

1,014,363
153,300
132,748
72,703
271,025
112,500
117,427
120,109
34, b51

322, 444
58. 198
57, 781
25,007
72,097
43,901
30,824
28,451
6,185

79,747
13,407
6,673
6, 225
15,748
8, 793
12, 701
10,018
6,212

65,668
8,759
7,133
4,121
17, 628
8,305
8,714
7,795
3,033

90,880
12,934
11,688
4,322
30,812
5, 923
9,534
13,913
1, 754

95,386
13, 711
9, 364
8,503
25,864
8,633
10,493
13,911
4, 907

162,348
21, 623
19,319
9, 154
48,440
16,680
20,718
21, 296

Pacinc_Washington __
Oregon
California_-

2,672, 896
478,325
303, 554
1,891,017

413,422
84,434
70,411
258,577

52,998
5,352
2,902
44,744

206,513
37,005
20,445
149,063

544,010
120,019
74,184
349,807

246,988
46,023
27,687
173,278

526, 019
81,206
51,108
393, 705

Middie Atlantic_-

New York_-New Jersey ____
Pennsylvania_-East North CentraL

Ohio_Indiana_-Illiuois __
Michigan ___
Wisconsin_West North Central_-

Minnesota_-Iowa_Missouri_-North Dakota_-South Dakota_Nebraska_-Kansas_South Atlantic_-

Delaware_-Maryland_
District of Columbia_Virginia_West Virginia_North Carolina ___
South Carolina_-Georgia_-FloridaEast South CentraL

Kentucky_-Tennessee_Alabama_
Mississippi___
Arkansas __
Louisiana_Oklahoma_-Texas_

18

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

99

138,883
11,906
8,895
5, 535
67, 043
12, 838
32, 66%

5,509,228

833, 938
68,160
49,914
22,767
400, 369
78,960
213,768

Maine_-New Hampshire ___
Vermont_Massachusetts ___
Rhode I s l a n d - Connecticut-

4,505

5,118

SEC.

2-B

STRUCTURAL TYPES

FIG.

7

F I G U R E 7 . - M U L T I P A G E N A R R O W , C O N T I N U O U S B O X H E A D- C o n .
[Titles, headnotes, and footnotes omitted]

Example 7. Common form, with repeated stub (page 2 of 2).
Finance,
insur- Business
ance,
and
Personal
and
repair
services
real
services
estate

Division and State

United States_--

1,013,297

New England__

Maine_New Hampshire_Vermont.Massachusetts.Rhode Island_Connecticut__Middle Atlantic_-

787,377 1,133,555

Amusement,
recreation,
and
related
services

Professional
and
related
services

Govern- Industry
not
ment
reported

316,063 1,472,453 1,414,069

450,570

70,430
3,586
2,314
1, 657
39,847
4,790
18, 236

51,745
5, 179
3,170
2,424
26,533
3, 959
10,480

82, 137
6,504
4,469
2,629
42, G03
7,093
18, 839

17,748
1,457
853
519
9,898
1,520
3,501

108, 121
8,253
5, 562
3,748
59,702
8,062
22, 794

98,949
9, 295
4,440
4, 318
53, 413
10,217
17, 266

33,293
4, 029
2, 347
1, 763
15, 759
2, 030
7,365

341,574
216,106

290,986
172, 664
43, 953
74,369

74,541
47,221
9,678
17, 642

374, 017
212, 765
54,253
106, 999

309,017
182, 687
46, 224
80,284

123,832
65,807
23,037
34, 988

66,927

183, 586
101,091
28, 654
53,841

East North Central_Ohio_-Indiana_Illinois __
Michigan_Wisconsin . . .

195,377
46,642
19, 770
80, 213
32, 014
16,638

167,569
41, 835
19, 993
57, 900
29, 970
17, 781

195,903
49, 581
21, 717
71,875
34, 995
17,735

60,456
15, 573
G,728
21,036
11, 4-53
5,666

298,871
78, 500
39, 420
94,835
53, 100
33,016

236,011
57, 956
24. 947
84,193
43,543
24,372

81,938
21,337
12,672
24,010
14,900
9,019

West North Central__

89, 173
19, 096
14, 811
30, 751
2, 239
2,496
9,014
10, 766

87,145
17,958
16,418
24, 499
3,602
3,871
8,838
11,959

85,452
17,575
13, 733
31,725
2, 483
2,689
8, 016
11, 229

27,440
6, 233
4, 943
7, 581
972
1,317
2,741
3,653

150,722
32,477
27,772
42, 244
6,109
6, 996
14, 168
20, 956

118,578
25, 939
18,868
32, 143
4,997
5,283
12, 814
18, 534

43,505
7,403
10,565
12,650
1,778
1,818
3,261
6,030

92,445
1, 952
16, 252
9,558
13, 584
5,540
11,344
5,985
13,8ll
14,419

77.572
1,600
11, 489
4,539
11,287
6,440
12, 790
5,724
12,039
11,664

157,892
2,815
17,842
11,670
21,601
8, 598
24,348
12,280
26,097
32,641

34,211
569
5,204
2,066
4,078
2,486
4,784
2,074
4,382
8,568

158,579
2,812
22,639
13,997
23, 127
16,270
25,271
12,641
22,426
19,396

251,320
3,208
37, 316
51,677
56, 464
9, 764
21,634
14, 400
32, 868
23, 989

57,700
1, 719
9, 396
2,236
8,717
5,072
11,058
3,895
8, 334
7,273

East South CentraL

37,159
10,693
12,930
9,379
4,157

39,618
12, 554
12,620
8,488
5 956

67,573
16,341
22. 249
16,912
12,071

12,058
4,060
3,621
2, 666
1,711

76,820
21,733
24,025
17,812
13, 250

66, 207
21,642
18,386
16, 988
9, 191

27,857
7, 900
7, 998
7,031
4, 928

West South Central_-

67,772
5,062
11,056
11, 913
39,741

71,350
7,030
10,544
13, 369
40,407

109,652
10,355
19,673
15,842
63,781

24,778
2,393
4,684
4, 523
13, 178

121,597
13,483
20, 069
24,871
63, 174

116,056
9,608
18,072
20,284
68, 092

38, 137
5,912
5786
7,859
18,580

19,899
2, 362
1,927

504

26, 735
3,700
3,216
1, 692
8,351
2,803
3,000
3,138
835

29,828
3,375
2,976
1,841
9, 054
3,344
4,660
3,228
I, 350

9,663
1, 115
1,229
568
2,478
831
1,342
1, 171
929

47,788
5,649
5,387
2,849
14,368
5,537
6,101
6, 580
1, 317

50, 376
6,861
4,256
6, 7S2
14,653
4,591
5, 796
5, 630
1,807

13,601
1,606
1,799
739
3,913
1,801
1,483
1,660
900

99,567
13,737
7,493
78,338

82,057
12,644
8,392
61,021

112, 132
14,434
8,784
88,914

55,168
4,316
2,791
48,061

135,938
22,174
13,681
100,083

167,377
31,336
11,494
124, 547

30, 707
5,645
4,183
20,879

New Y o r k New Jersey __
Pennsylvania_-

Minnesota_Iowa_Missouri_North Dakota_-South Dakota_-Nebraska_-Kansas.South Atlantic_-

Delaware__Maryland_.District of Columbia_-

Virginia_.-

West Virginia_-North Carolina_-South Carolina_-Georgia__
Florida_-Kentucky ___
Tennessee_-Alabama_MississippL

Arkansas.-Louisiana_Oklahoma_-Texas_Mountain_-

Montana__Idaho_-Wyoming ___
Colorado_-New Mexico_Arizona ..
Utah-Nevada_--

7,619
1,388
2, 061
3,138

Pacific ..

Washington . . .
Oregon ___
California_-

776554
49

3


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

19

Figure 8 . - T H E BROAD TABLE
[Titles, headnotes, and footnotes omitted]

Example 8. Multipage broad table with continuous stub. (Need not start on left-hand page; need not appear on facing pages. Where
pages face, letterpress practice omits box on lower page if identical wtih box on upper page. Offset practice usually repeats box on every
page because of difficulty of maintaining typing line-up of columns on the two pages. Must be placed on page so title is at left when
book is in upright position.)

Industry

Abrasive wheels, stones, paper, cloth and related
products_
Agricultural machinery (except tractors)__Aircraft and parts, including aircraft engines _
Alloying; and rolling and drawing of nonferrous
metals, except aluminum_Aluminum products (including rolling and drawing
and extruding), not elsewhere classified_Aluminum ware, kitchen, hospital, and household
except electrical appliances_Ammunition_
Artificial leather and oilcloth_Artists' materials_Asbestos products (except steam packing and pipe
and boiler coverage)
Automobile stampings __
Automobile trailers (for attachment to passenger
cars)
Automotive electrical equipment_Baking powder, yeast, and other leavening compounds_Baskets for fruits and vegetables_Batteries, storage and primary (dry and wet)_Batting, padding, and wadding; upholstery filling_Beauty-shop and barber-shop equipment_Beehive coke_Beet sugar_Belts (apparel) regardless of material_
Biscuit, crackers, and pretzels_-

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Average
Number employed on 15th day of month or nearest representative day
number
employed
during January February March April May
June
July
August SeptemOctober Novcmyear
ber
ber

7,734
27,806
48,638

7,027
27, 900
34,433

7, 131
29,743
36,512

7,171
30,142
38,188

7,300
30,480
41,101

7, 291
29, 924
44, 567

7,287
28, 344
48,057

2,336
25,653
49, 832

7,405
25, 719
51, 535

7,881
26,033
52, 946

8,610
26,667
56,289

9,177
25,748
62,601

December

0

9, 186
27,314
67,589

38, 816

35, 418

35,503

35,617

35,452

35, 543

35, 687

35,812

36,803

40,028

45, 103

47, 449

47,378

17,249

14,076

14,699 15,629

16,031

16,320

16, 390

16, 798

17,204

17, 366

19, 269

21,239

21, 968

6,297

6,063

6, 194

6, 374

6,485

6,382

6, 301

6, 213

6,313

6, 214

6,163

6,305

6, 550

4, 264
3, 976
397

3, 679
3,869
397

3,862
3,950
402

3, 909
3, 907
396

3, 960
3,835
377

3, 924
3,772
367

3, 962
3,664
363

4,061
3, 791
366

4,236
4, 055
368

4, 252
4,173
403

5, 099
4,277
440

5, 381
4, 242
447

4, 844
4,178
443

9,979
8,597

9,234
8,993

9,371
8, 358

9, 397
8, 375

9, 743
8,094

9,538
7,582

9, 817
7, 495

9, 700
7,372

9, 952
7,727

10, 298
8,860

10,891
9, 922

10,937
10,076

10,875
10,312

1,426
17,495

923
17, 217

945 1,125
17,110 16,738

1,398
16,768

1,583
16,753

1,639
16,526

1,700
16,041

1, 575
16,797

1, 474
18,032

1,518
19, 210

1,589
18, 306

1,643
20,437

2,334
8,048
15, 034
4,505
1,986
685
10,410
4,222
29, 173

2,302
5,737
13,657
4,425
1,849
480
6,710
3,858
28, 551

2,362
2,337
2,333
2,306
2,334
2,366
6,505
7,431
9,003
8,786
7, 817
9, 693
13, 726 14,269 13,690 13, 616 13,818 14, 309
4, 476
4,377
4,077 4,077 3,833
4,232
1,785 1, 935
2,071 2,242 1,953
2, 146
475
475
395
431
412
102
3,636
3, 631
4, 022
4, 541
5, 136
4,368
4,281
4,351
4, 091
4,008
3,899
4,280
29, 407 29, 519 28,847 29,075 29, 261 28,803

2, 345
9, 591
15, 504
4,033
1,885
421
8,094
4, 555
28,716

2,333
9, 951
16,423
4, 783
1, 989
758
10,949
4, 544
29, 907

2, 333
9,160
17,827
5,303
2,108
1,434
27, 541
4, 547
29, 935

2,329
6,615
17, 359
5,160
1, 957
1,477
27, 195
4, 252
29, 440

2, 334
6, 291
16,207
5, 278
1, 915
1,359
19, 092
3,999
23,907

°

Blast-furnace products__Blended and prepared flour made from purchased
flour
Blowers; exhaust and ventilating fans_
Bluing_
Boat building and boat repairing __
Bolts, nuts, washers, and rivets-made in plants
not operated in connection with rolling mills-Bone black, carbon black, and lampblack_Bookbinding and related industries_-Books: Printing without publishing_Books: Publishing and printing
Books: Publishing without printing __
Boot and shoe cut stock and findings_Bread and other bakery products (except biscuit,
crackers, and pretzels)
Brick and hollow structural tile __
Brooms_Brushes_Buttons.Candles_
Candy and other confectionery products _
Cane sugar-except refineries_Cane-sugar refining _

18,845

19,317

201,533
29, 069
3,787
7,891
10,972
840
49, 740
4,217
14, 133

196, 123
22, 457
3, 745
7, 056
10,083
785
48, 321
2,150
13,531

98,022
15, 735
3,859
1,741

44,056
11,706
3, 026
1,698

43, 309
11,654
3,238
6,703

47, 117
12,012
3,259
1,728

57,133
11,288
3,850
1,733

3,189
25,590

3,003
24,711

3,018
25,342

3, 004
26,018

3, 137
3,137

2, 982
3, 206

3,148
3, 340

24, 523
2,732

21, 361
2, 413

12,447
16, 4S8
23,801
7, 458
5,009
60,268
2,627

12,504
15,538
18,441
6,951
4, 436
57,002
2,446

Canned and dried fruits and vegetables (including
canned soups)
Canned fish, crustacea, and mollusks_-Canvas products (except bags)_
Carbon paper and inked ribbons __
Carbon products for the electrical industry, and
manufactures of carbon or artificial graphite __
Carpets and rugs, wooL _
Carpets, rugs, and mats made from such materials
as paper fiber, grass, jute, flax, sisal, cotton, cocoa
fiber, and r a g s Carpet yarn, woolen and worsted_-Cars and car equipment-railroad, street, and
rapid-transit_-Cars and trucks, industrial_Caskets, coffins, burial cases, and other morticians'
goods_-Cast-iron pipe and fittings__Cement_Cereal preparations_-Cheese_
Chemicals not elsewhere classified_Chewing gum_-


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

19, 537

17,472

17, 609

17,863

17,614

16,478

17,735

18,142

18, 635

20,880

24, 007

24,099

706
3, 885
55
2,630

650
3,333

2, 449

676
3,402
52
2,621

673
3,538
57
2,090

678
3,792
56
3, 090

674
4, 045
57
3,396

696
4, 161
55
3,403

710
4, 136
52
2, 965

694
4,043
52
2, 444

762
4, 003
59
2,117

753
4, 045
59
2, 214

744
4, 074
57
2,316

14, 331
1,574
25, 690
16, 547
6,091
135

13, 417
1,437
25, 704
17,937
5,970
133

13, 509
1,601
25, 533
16,818
6, 024
134

13,557
1, 573
25, 725
16,082
6, 060
136

13,326
1,579
25, 469
16,096
6, 108
134

13,212
1,514
25, 222
16, 775
6,125
133

13,307
1,509
24, 662
15,936
6, 067
136

13, 477
1, 562
24,997
17,230
6, 085
133

14,027
1, 554
25,393
17, 244
6,217
133

14, 669
1,619
26, 202
15, 958
6, 198
134

16, 177
1,591
26,221
15,541
6,132
135

16,540
1, 660
26,356
15,857
6, 084
138

20, 171

19, 906

19, 093

17, 701

18,072

23,907
758

4, 046
53
2, 450
16, 758
1, 684
26,791
17,086
6, 022
138

18,892

18,832

1S, 629

18, 864

18,059

18,600

196,585 197, 269 198, 820 201, 396 202, 764 203, 880
21,687 24, 777 28,514 31, 147 33,072 32, 798
3, 623
3,795
3,825
3,811
3,739
3,572
7, 528
7, 796
7,217
7, 465
7,380
7,455
10,821 10,813 10, 438
9,677 10,205 10, 459
745
762
767
787
767
710
48,371 48,506 45, 027 43, 511 41, 349 40, 659
1, 960
1, 431
1, 844
1, 910
2,262
1, 861
12,939 13, 619 15,592 13,938 13,688 15, 315

204, 041
33, 024
3,735
8, 054
11,376
806
45,577
2, 626
15, 132

203, 858
32,387
3,900
8,278
11,813
889
54, 734
2,737
13,368

204, 121
31,608
3,970
8, 683
11,953
988
59, 238
9,870
14, 818

205, 074
29, 971
3,927
8,934
12, 114
1,030
61,510
11,298
14, 134

204, 469
27,387
3,801
8,851
11, 884
1,039
60,082
10,653
13,524

58, 836
12,910
5,115
1,733

85,138 134,857
13,537 12, 530
5, 356
4, 423
1,720
1,705

218,359
20,411
3, 741
1,708

230,497
22, 842
3,399
1,723

134,414
22,620
3, 695
1,763

70,820
20, 384
3,638
1, 838

51,723
16, 924
3, 694

2, 907
25,801

2. 979
24, 713

3, 031
24, 284

3,011
24, 139

3,106
24, 768

3,081
25, 469

3,500
26,851

3,759
27,588

3, 774

3,253
3,050

3,149
3,010

2,937
2,859

2,862
2,931

2,728
2, 987

2,915
2,973

3,288
3,167

3, 421
3, 394

3,508

24, 131
2,526

24, 230
2, 586

24,387
2,573

23,956
2,616

23, 062
2,635

21,235
2,701

21,556
2,767

22, 744

26,012

2,761

2,883

12,650
15, 743
19,504
6, 965
4, 467
57, 538
2,512

12, 649
15,871
21, 743
7,056
4, 688
57,557
2,597

12,482
15,918
23, 372
7,158
4,998
57,330
2, 622

12,470
16, 150
23, 962
7, 556
5,310
57,675
2, 704

12,410
16,277
25, 744
7,724
5,538
57,844

12,096
16,818
26, 170
7,930
5,517
58, 107
2, 656

12, 249
17,022
26,134
7,870
5, 413
59,202
2, 749

12,315
16, 942
25, 629
8,821
5,278
61,998
2,738

12,393
17,289
25,943

2,712

7,842
5,072
65,265
2, 671

3, 476

°
Q

1, 843
ti

27,404
3,490
3, 223

29,674
3,114

31,926

12,527
17,158
25,237
6, 655

12, 614
17,132

3, 179

23,737
6, 969

4,753

4, 642

66,887
2, 632

66,813

2,483

00

FIG.

9

THE

2-B

SE0.

FORMAL TABLE ; STRUCTURE

FIGURE 9 . - T H E

{Titles, headnotes, and

Example 9 - A . Normal type, with tracer numbers; stub omitted from right-hand
a left-hand page and ending on a right-hand page. Title and headnote center
both pages. Compare with multi page narrow table which repeats stub at left of
Building contractors

All contractors
Total
Division and State

Resident

Nonresident

Resident

t

Nonresident

Num- Value
ber of
work
estab- of perlish
men.ts formed

Num- Value Num- Value
ber of
of
ber of
of
estab- work estab- work
lishperlishpermen.ts formed men.ts formed

Num- Value Num- Value
ber of
of
of
ber of
estab- work estab- work
lish
perper
lisb
ments formed ments formed

1,877 21,882
1, 332 17,282
923 110,615
8,812 46,245
1,847 30, 090
6, 036 99, 182

1, 694 15,851
989 10, 441
739
6,624
8,483 132,054
1,557 22, 844
5,397 83,610

183 6, 031
343 6,841
184 3,991
329 14, 191
290 7, 246
639 15,572

178
120
93
1,071
309
816

4,227
3,182
2, 412
36, 513
8, 324
28, 495

34
68
36
60
59
118

1,680
1,173
1, 300
1,979
1,181
4, 099

25, 155 665, 932 24,561 608, 657
9,880 164,399 9,099 128,337
17,814 314, 934 17, 182 271,758

594 57, 275
781 36, 062
638 43,176

2,437 124,496
1, 002 36,376
1, 881 70,630

113
131
109

8,374
7, 925
7,624

12,815
6, 74G
13,395
9, 180
7,375

444
588
594
365
382

17,640
31,550
29,554
16, 930
11,933

1,623
771
1, 457
1, 334
816

65
97
96

3, 040
9, 903
7, 682
4,256
3, 674

5,566
389 12, 993

27,786
21,207

49
53

22, 437

43

2, 469
2, 434
8,462
10, 092

10
71

3,355
1, 870
606
294
286
5,272

N e w England:
1
2
3
4
5

Maine_-New Hampshire_
Vermont_Massachusetts ___
Rhode Island ___
Connecticut_Middle Atlantic:

7
8

9

New York_-New Jersey__Pennsylvania_-E. North Central:

10
11
12
13
14

Ohio-Indiana_Illinois_Michigan__Wisconsin_-

227,447 12,371 209, 807
116, 625 6, 158 85, 075
288, 496 12,801 258,942
179, 995 8,815 163, 065
112,140
6,993 100,207

67,186
29, 874
60,128
52, 228
25, 265

63

58

W. North Central:
15
16
17
18
19
20
21

Minnesota_-Iowa_Missouri_
North D a k o t a - South Dakota--Nebraska_-Kansas_-

6,179 98, 102
5,428 78,185
6, 764 97, 322
666 10,832
817 10,213
2,291 40, 346
3,271 43, 605

5,971
5, 039
6, 480
552
703
2,157
2,847

7,650
8, 252
31,092
29,866

114
115
134

3,182
1, 961
9, 254
424 13,739

8
575
801
65
95
211
395

1,057
3,974
1,420
3, 736
1,731
2,610
1,059
2, 135
2,782

17, 629
79,304
60,787
91, 321
34,778
76,164
35,149
53, 717
72,657

808
3,278
1,113
3,131
1,421
2, 406
840
1,924
2, 602

13,104
59,970
40,412
64, 173
24,138
63, 357
21,939
46,979
63,512

249
696
207
605
310
204
219
211
180

4, 525
19,334
20,375
27,148
10,640
12, 807
13,210
11,738
9, 145

148
588
215
737
227
612
203
452
752

3,871
21, 964
10,711
23,551
8,128
27,273
9, 446
17,129
25,288

33
123
43
118
51
37
29
43
36

1, 146
4,683
4, 788
7, 951
3,120
2,952
1, 975
7, 438
3, 249

3,080
2,571
1,409
1, 315

55. 988
52. 628
48,501
40, 941

2,749
2,358
1,224
1,109

42, 420
42,927
29,973
15,478

331 13,568
213 9,701
185 18,528
206 25,463

404
392
310
204

15,399
16,578
12,986
6,170

43
39
28
21

3,186
1,002
6,950

1,334 24,545
1,858 71,164
2,242 37,305
8,380 199, 540

1,117 12,985
1,635 48, 924
2, 091 30,632
8,144 170,138

217 11,560
223 22,240
151 6,673
236 29,402

201
331
339
1,840

5, 069
20,680
11,618
55, 283

28
37
31
46

1,683
4,260
1,940
7, 446

897
873
658
2,112
606
945
1,148
326

766
704
556
2,026
501
864
1,095
218

131
169
102
86
105
81
53
108

134
171
90
375
136
290
198
56

5, 455
3,101
2, 499
10,486
3,924
5,879
5, 499
1, 935

18
22
11
14
15
6
8
12

1,711
1,194
281
1,656
506
237
567
525

654 14, 978
9,055
377
5, 921 134,395

34
15
14

1, 939
605
329

92,536
65, 192

84, 176

284 13, 146

20
15

823

South Atlantic:
22
23
25
26
27
28
29
30

Delaware ___
Maryland_Dist.ofCoL
Virginia_West Virginia_North Carolina ___
South Carolina_Georgia__Florida_--

E, South Central:

31
32
33
34

Kentucky_-Tennessee __
Alabama_
Mississippi.-W. South Central:

35
36
37
38

Arkansas_Louisiana __
Oklahoma_-'Texas_Mountain:

39
40
41
42
43
44
45
46

Montana_Idaho_
Wyoming_Colorado
New Mexico_-Arizona_Utah_-Nevada_--

22,703
13, 739
14, 439
40, 399
15, 031
18,286
18,665
9,669

15,903
8,537
8,121
34,788
10,810
14,395
16,308
6,339

6,800
5,203
6,318
5, 611
4,221
3,891
2,357
3, 330

Pacific:
47
48
49

Washington-Oregon __
California_-

3,637 90,678
2, 133 37,061
23,670 408,137

22

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

3,464 78,336
2, 029 32,553
23, 566 397, 949

173 12,342
104 4,508
104 10,188

SEC.

2-B

STRUCTURAL T Y P E S

FIG.

2

PARALLEL TABLE
footnotes omitted]

page. (Parallel tables should always a p p e a r in pairs of facing pages, s t a r t i n g on
across each pair of facing pages a n d each line of t a b l e must be lined up across
right-hand page, a n d repeats title a n d h e a d n o t e on each page.)
Heavy contractors

Highway contractors

Special trade contractors

Resident

Nonresident

Resident

Nonresident

Resident

N onresident

Num Value
ber of
work
estab- of per
lish formed
ments

Num
ber of Value
work
estab- of perlishments formed

Num- Value
ber of
work
estab- ofper
lish
ments formed

Num- Value
ber of
work
estab of per
lish
ments formed

Num- Value
ber of
work
estab- of perlishments formed

Num- Value
bcr of
work
estab- of perlishments formed

22
13
15
101
15
70

2,142
1,154
649
12,192
2,356
11,166

9
19
19
11
8
18

1, 581
2,179
1,704
1, 907
357

48, 015
109
227

37,608

35
22
47

6,321
1,905
13, 363

231
97
229
109
182

24,270
12, 131
32, 254
13,114
11,777

17
24
33
15
22

2,323
1,929
872
895

244
144
84
26
42
73
60

14, 676
12,532
6, 749
3, 728
4, 042
5, 487
6,173

16
24
21
11

14, 684

986
330
125
13,596
994
3,139

17
16
6
31
30

56

1, 632
1, 685
159
6,825
4, 212
2,611

220 108, 736
9,045
64
85 20, 046

52
72
69

85
29
110
64

14, 270
4,740
26,347
12,235
9,890
9,327
5, 681
5, 237

2,082
2, 120

44
31
41
1
1
24
22

1,871
236
2,991
l, 989
4, 557
5,076
1,208
1,085

2
24
6
45
17
25
4
21
39

3, 899
1,309
5,469
2,904
5,120
340
4,025
4,821

36
26
25
22

107
3,827
6,040
7, 706
2,738
2,508
4, 310
943
2, 163

33
28
25
44

4. 601
4, 099
5, 614
7, 095

1,421

1,838

1, 704
1,072
1,769

20
20
19

7
5
86
9
34

5,280
2,524

8, 496
5,775
3,438
69, 753
11,170
40,810

123
240
123
227
193
447

1, 138
1, 804
828
3, 480
1, 496
7, 441

1
2
3
4
5
6

15,226
15,631
11,973

21,608 327,410
7, 924 68, 232
14,989 143, 474

394
556
407

27,354
10,601
10,216

7
8

42
70
83
46
47

6,082
10, 254
11,413
5, 942
4, 437

10,432 104,081
5,261 38,330
11,005 140,213
7,308 85,488
5,930 53,275

320
397
382
241
255

6,680
9, 070
8,530
5, 860
2,927

10
11
12
13
14

25
48
47
11
14
33
33

1, 378
4, 234
6,564
4 1, 005
373
5, 590
2, 901

4,885
4, 289
5, 554
460
565
1, 849
2,370

40,747
25, 772
49, 753
2,541
2, 693
11, 863
11,077

118
264
173
72
66
71
301

1, 661
4, 332
2, 943
483
377
1,296
3, 446

15
16
17
18
19
20
21

24
34

4 1,

638
2,615
881
2, 286
1,146
1,737
613
1, 396
1, 768

7,395
28, 857
24, 849
27, 009
7, 003
22, 879
6, 769
15,650
24, 991

184
518
224
422
218
114
136
132
113

2,114
8,953
9, 311
8,500
2,793
2,790
1, 849
2, 149
2, 649

22
23
24
25
26
27
28
29
30

2,272
882
876

16,755
17, 998
10,312
4, 104

223
125
101
87

3,363
2, 062
2,015
819

31
32
33
34

887
1,233
1, 701
6,074

3, 993
17,060
11, 346
55, 433

117
126
84
131

1,718 35
6,277 36
1,344 37
4,577 38

591
514
441
1,602
343
553
872
156

5,421
3, 491
3,062
13,992
2,780
4,907
6,333
2, 256

76
110
57

2,677

0, 979

1, 576 123,990

99
59

17, 235 106,149

66

31
32
20
55
43

1. 996
5, 250
3,543
8, 144
6, 103
8,085
5,384
10, 175
8,412

61
33
20
23

7,968
5, 162
4, 472
3,500

22
21
31
54

2, 418
2,538
3, 949
16, 886

12
17
12
6

2, 298
3,189
2,203
1,704

15
35
26
113

2, 767
5,356
4,581
40,373

21
10
7
8

2, 086
2, 216
695
2,806

14
36
25
117

1,156
5, 828
3,087
19,049

51
50
51

6,073
9,487
2, 694
14, 573

31
13
15
32
18
16
19

2, 465

1, 819
7, 889
3,649
3,146

10
6
10
]7
4
5
6
2

1, 651
362
741
2, 421
457
463
830

7

1,900
3, 436
713
450
1,240
91
4 3, 106

16
13
16
11
21
9
10
4

1,870
1, 448
2. 211
2,367
2,827
1, 667
1,205
41, 195

20
12
5

1, 953
572
581

56
27
177

32,453
3, 490
42,611

20
18
19

5, 230
2, 648
5, 586

20
51
11

77
49

3

3,376

8
21
12
15
28
25
13

21

1,582
18
5
10

3,646
9, 926
6, 028


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

29

1,487
851
627
7, 225
1,224
4, 477

1, 916

64
56
34
85

754
661
390
875
438
747
494
562

39
40
41
42
43

44
45
46

3,220 47
683 48
3, 692 49

23

FIG.

9

THE FORMAL TABLE : STRUCTURE
FIGURE

SEC.

9.-THE

2-B

PARALLEL

[Titles, headnotes, and

Example 9 - B . With stub repeated on right of right-hand page. (Parallel tables
a n d ending on a right-hand page. Title and headnote center across each
Compare with multipage narrow table which repeats s t u b at left of right-hand
All land in farms
(thousands of acres)

Number of farms
Division and State
1920
United States __

Maine_New Hampshire _
Vermont_-Massachusetts_Rhode Island _
Connecticut_Middle Atlantic __

New York_-New Jersey __
Pennsylvania_Ohio_-Indiana_-Illinois _
Michigan_Wisconsin_West North Central__

Minnesota_Iowa_-Missouri_-North Dakota_South Dakota_Nebraska_-

Kansas_-

South Atlantic_-

Delaware __
Maryland.
District of Columbia_
Virginia_West Virginia_-North Carolina_-South Carolina_Georgia__Florida_-

East South Central.-

Kentucky __
Tennessee_Alabama_-Mississippi__

1930

1935

1940

1920

1925

1930

6, 448, 343 6, 371, 640 6,288,648 6, 812, 350 6,096,799 955, 884 924,319 986,771

N e w England_-

East North Central_-

1925

156,564
48,227
20, 523
29, 075
32,001
4,083
22,655

159, 489
50,033
21,065
27,786
33,454
3, 911
23,240

124,925
39,006
14,906
24, 898
25, 598
3, 322
17, 195

158,241
41, 907
17,695
27,061
35, 094
4,327
32, 157

135,190
38,980
16,554
23, 582
31,897
3, 014
21,163

16, 991
5,426
2,604
4, 236
2, 494
332
1,899

15,858
5,161
2,262
3, 926
2,368
309
1,832

14,283
4,640
1,960
3, 896
2,005
279
1,502

425,147
193, 195
29, 702
202,250

418,868
188,754
29, 671
200,443

357,603
159,806
25,378
172, 419

397, 684
177,025
29, 375
191, 284

348, 100
158, 238
35,835
169,027

40,573 37, 491
20,633 19, 270
1, 925
2.283
17,658 16, 296

35,047
17,980
1,758
15, 309

1,084,744 1,051,572
244,703
256,695
195, 786
205, 126
225,601
237, 181
192, 327
196, 447
193, 155
189, 295

966, 502 1,083,687 1,006,095 117, 735
233,783 23, 516
255, 146
219, 296
184, 549 21,06.3
181,570
200,835
213, 439 31, 975
214, 497
231,312
187,589 19, 033
196, 517
169,372
199,877
181, 767
186, 735 22, 148

112,752 110,891
22, 219 21,514
19, 915 19, 689
30,732 30, 695
18,035 17,119
21,851 21,874

1, 096, 951 1,111,314 1,112,755 1,179,856 1,090,574 256,973 248,081 285, 488
188,231
197, 351 30,222 30,059 30, 913
178, 478
203,302
185, 255
213, 490
213, 318 33,475 33,281 34, 019
213,439
221, 986
214,928
260,473
256,100 34,775 32,642 33,743
263,004
278, 454
255, 940
75,970
73,962 36,214 34,327 38,658
77,690
84,606
77,975
79,537
72,454 34,636 32,018 36,470
83,303
74,637
83, 157
127,734
124,417
129, 458 133, 616
212,062 42, 225 42,025 44, 709
165, 879
165,286
156,327 45, 425 43, 729 46,976
166,042 174,589
1,158,976 1, 108,061 1, 158, 468 1,147,133 1,019,451
8, 994
9,707
10,257
10,140
10,381
49,001
42,110
47, 908
44, 412
43,203
139
65
204
89
104
193, 723
174,885
186,242
197, 632
170, 610
90, 380
99, 282
87, 289
104, 747
82,641
269,763
300,967
283, 482
278,276
279,708
192, 693
165,504
157,931
172, 767
137,558
249,095
255, 598
216, 033
310, 732
250, 544
59,217
72,857
58, 966
62,248
54,005

4

3

18, 561
9,570
20,022
12,427
25,441
6,047

17,210
8,980
18, 594
10,639
21,945
5,865

16, 729
8,802
18,055
10,393
22,079
5, 027

1,051,600 1, 006, 052 1, 062, 214 1,137,219 1,023,349
258,524
252, 894
270,626
278, 298
246,499
252, 669
247,617
252,774
273,783
245,657
237, 631
256,099
273,455
257,395
231, 746
257,228
272, 101
291, 092
311,683
312,663

78, 897
21,613
19, 511
19, 577
18,197

70,607
19, 913
17,901
16,739
16,053

72,817
19, 927
18,003
17,555
17,332

West South Central__

Arkansas_Louisiana _
Oklahoma_Texas__Mountain_-

Montana_Idaho_-Wyoming__Colorado_-New Mexico_-Arizona_-Utah_
Nevada_Pacific__
Washington_-

Oregon_
California.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

996,088 1,017,305 1, 103, 134 1,137,571
221,991
232,604
253, 013
242,334
132,450
135,463
161, 445
170,216
197, 218
191, 988
203,866
213,325
465,646
436,033
495, 489
501,017

97,775
945
4,758
6

88,589
900
4,433

86,363
901
4,374

964,370 173,449 165,013 183,908
216,674 17, 457 15,632 16,053
9,355
8,838
150,007 10,020
179,687 31,952 30, 869 33, 791
418,002 114,021 109,674 124,707

244,109
57,677
42,106
15,748
59,934
29,844
9,975
25,662
3,163

233, 392
46, 904
40, 592
15,512
58,020
31,687
10,802
25, 992
3, 883

241,314
47, 495
41,673
16,011
59, 956
31,404
14,173
27,159
3,442

271,392
50,564
45,113
17,487
63,644
41,369
18,824
30, 695
3, 696

233, 497 117,337 131,689 157,450
41,823 35,071 32,736 44,659
9,347
8,116
43,663 8,376
15,018 11,809 18, 663 23, 525
51,436 24,462 24,167 28,876
34,105 24,410 27,850 30,822
18, 468
5,802 11,065 10,527
5, 613
5,050
5,001
25,411
4,081
2,357
4, 091
3,573

234, 164
66,288
50, 206
117,670

265,587
73,267
55, 911
136,409

261,733
70, 904
55, 153
135,676

299,567
84,381
64, 826
150,360

276,173
81,686
61, 829
132,658

56, 153
13,245
13, 542
29, 366

54,258
12,610
14,131
27, 517

60,525
13,534
16,549
30,443

SEC.

2-B

STRUCTURAL TYPES

FIG,

9

TABLE-Continued
footnotes omitted]

should always a p p e a r in pairs of facing pages, s t a r t i n g on a left-hand page
p a i r of facing pages a n d each line of t a b l e m u s t be lined up across b o t h pages.
page.)
AJJ land in farms
(thousands of acres)
-continued

Percent of total land area represented by all land in farms

Average acreage per farm

Division

and State
1935

1920

1925

1930

1935

1940

1920

1925

I, 054, 515 1,060,852

148, 2

15, 463
4,722
2,116
4,043
2,196
308
2,080

13,371
4,223
1,809
3,667
1, 938
222
1,512

108.5
112. 5
126. 9
145.
77. 9
81. 2
83.8

145. 1

156, 9

154.8

174. 0

50. 2

99.4
103. 2
107. 4
141. 3
70.8
79.0
78.8

114, 3
119.0
131. 5
156. 5
78.3
84.1
87.4

97. 7
112. 7
119.6
149.4
62.6
71.1
64. 7

98. 9
108.3
109.3
155. 5
60.8
73.6
71.5

42, 8
28. 4
45.0
48. 5
48.6
61.6

36, 455
18,686
1,914
15,855

33, 639
17,170
1, 874
14,594

95.4
106. 8
76. 8
87.3

89.5
102.1
64. 9
81.3

98. 0
112. 5
69.3
88.8

91. 7

96. 6
112.1
72.6
86.3

63. 4
67. 7
47.5
61.5

116, 957
22,858
20, 519
31, 661
18, 460
23,459

113,655
21,908
19,801
31,033
18,038
22,876

108. 5
91. 6
102. 7
134. 8
96. 9
117.0

107. 2
90. 8
101. 7
136. 2
93.8
113.1

114. 7

113. 0

74, 9
90.2
91.3
89.1

273,077
32, 818
34,359
35,055
39,118
37,102
46,616
48,010

274, 423
32,607
34, 149
34, 740
37,936
39,474
47,344
48,174

234. 3
169. 3
156.8
132. 2
466. 1
464. 1
339.4
274.8

95, 987
921

92, 555
896

84. 4
93.1

4, 384

4, 198

99. 3

3
17,645
9,424
19,936
12,330
25,297
6,048

2
16, 445
8,909
18,845
11, 239
23,684
8,338

27.8
99. 7
109.6
74. 2
64.5
81.9
112.0

99. 4
65.6
61. 6
88.1

79, 101
20, 699
19,086
19,661
19, 655

77,086
20,294
18, 493
19, 143
19,156

201, 118
17,742
10, 444
35,335
135, 597

173,88)

1940

105. 6
65.2
82. 9

98. 1
108.4
143. 1
101.1
120.3

107.9
89.6
102.2
136. 9
93. 9
117.4

93. 7
107.3
145.4
96. 2
122. 5

223, 2
159. 7
155. 9
125. 3
451.9
402. 6
329.0
263.6

238, 6
166. 9
158.3
131.8
495.8
438.6
345. 4
282. 9

231, 4
161.4
154. 8
125. 9
462. 4
445.4
348. 9
275.0

251. 6
165. 2
160. 1
135. 6
512. 9
544. 8
391.1
308. 2

79. 9
87. 7
90. 5
27.4

83, 7
88. 7
98. 7
31. 5
89.3
90.0
66.2
74.5

90.8
99.6
99. 7
36.0

99.0

81. 6
92.8
101.3
29. 5
98.1
106. 5
64.5
65.8
86.4
85. 2

101. 0
83. 0

75, 0
79. 9
77.2
76.4
66. 9

70. 2
77.0
70. 8
70.4
62.4

68, 6
80.8
73.3
68.2
55.4

69. 6
74.4
69. 7
71.9
63.1

75.3
80.2
74. 7
82.6

200,527
18, 045
9, 996
34, 803
137, 683

174. 1
75.0
74.0
166. 4
261.5

162. 2
70.4
66. 7
156. 5
235. 5

166, 7
66. 2
57.9
165.8
251. 7

176.8
70.1
61.4

207. 9

47, 512
9,952
28,162
29, 978
34,397
14,019
6,239
3,622

191, 901
46, 452
10,298
28, 026
31, 527
38,860
25, 651
7,302
3,785

480, 7 564. 2 652, 5 640, 7 821.9
608.1 607. 9 940.3 939.6 1,110.7
198. 9 199. 9 224.3 220. 6 235. 8
749. 9 1,203.2 1,469.3 1,610.4 1,866.2
408.1 416. 5 481. 6 471. 0 612. 9
817. 9 878. 9 981. 5 831. 5 1,139.4
581. 7 1,024.4 742. 7 744. 7 1,388.9
196.8 192. 4 206. 7 203.3 287.4
745. 2 1,053.5 1,185.6 979. 9 1,059.4

62, 476
14,680
17,358
30, 438

63,694
15, 182
17,988
30,524

239.8
199.8
269. 7
249.6

88.8

204.3

172.1
252. 7
201. 7


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

231, 2
190.0
300.1
224. 4

94.0
89. 7
67. 7
81. 7
109. 6
133.9

65.8
83.3

165. 6

66.6
193. 7

274.6

329.4

72. 5

1930

1935

1940

48, 6

51.8

55. 4

55. 7

U.S.

40. 0
27.0
39. l
67. 2
46. 0
45.3
59. 4

36, 0
24.3
33. 9
66. 7
39. 0
40. 9
48. 7

39. 0
24. 7
36.6
69.2
42. 7
45.1
67.4

33. 1
21.3
31. 3
61. 8
38.3
32.8
48.2

N.E,
Maine.

58. 6
63. 2

57. 0
61.3
39.8
55.3

52. 3

M. A.
N.Y.
N.J.

40.0

54. 8
59.0
36. 6

56.8

53. 4

71. 7

70. 6
82. 5

Conn.

Pa.

83. 2
85.5

E.N.C.
Ohio.
Ind.

86. 7
49.4
65. 3

Mich.
Wis.

84. 0

W.N.C,
Minn.

72.5

86.3

85.3

85. 7
49.0
61.8

85.6
46.5
61. 9

75. 9
58.1
93.6
74. 2
76.4
65.1
85.5
83.6

81.9
59. 7
95.6
76.6
86.1
74. l
91.0
89.8

83, 5
63.4
96.6
79. 7
87.1
75.4
94. 8
91. 7

51,4
71. 5
69. 7
9. 6
66.8

50. 2
71.6
68. 8
7. 7
64.9
57. 3
57. 9
53.3
58. 7
14.3

55,7
73.3
68. 9
7. 1
68. 5
61.3
63.9
63.2
67.3
17. 2

57. 8
59. 9
57.4
63. 2
24. 0

s. C,

68. 9
80. 5
71.5
59. 9
66.2

66. 7

E. S. C

54. 1

63. 4
77.5
67. 5
53.5
58.4

79.1
68.9
58.6
63.1

G3. 1
51. 9
34.5
71.9
67. 9

60. 0
46. 5
30.4
69. 5
65.3

66. 9
47.8
32.2
76.1
74. 3

73.1
52.8
35. 9
79.6
81.9

21.3
37.5
15. 7
18. 9
36.9
31.1
8.0

24. 0
35.0
15. 2
29. 9
36.4
35.5
15. 2
9. 5

28, 6
47.8
17. 5
37. 7
43.5
39.3
14.5
10. 7

31.6
50. 8
18. 7
45. 1
45.2
43. 9
19. 2
11.9
5. 2

35. 0
49.6
19. 4
44. 9

30. 7
34.3

31, 1
35. 4

28. 4

29. 2
30.4

51. 7
62.6
78. 6

58.4
94.1
79. 1
80. 6

70.4
85. 9

86.8
56.8
75.1

74.8

14. 8
72.0
62.2
64. 2
63. 7
67. 7
17. 2
68. 7
84.0
73.1
59. 7
61. 3

9. 6

58. 4
59.6
54.5

58.4
16. 7
61.5

77.4
67.1
51.0

5.8

5. 8

27.6
31.0
22.1

26, 7

267.8

230. 6
185. 9
290. 0

202.4

230.1

29. 5

27. 6

29, 7
31.6
27.0
30.6

174.0

Mass.
R. I.

74.4
87. 7
88. 9
88.3
50. 2
66.3

3.4

208, 6

56. 0
38. 9
50. 6

N.H.
Vt.

29.5
23.1

30.6

63. 7
95.3
78.4
84.6
80. 6
91. 7
53. 9

70. 7
66.3
6. 0

64.4

72, 7

53.3
34. 6
78. 5
81.6

Iowa.
Mo.

N.Dak.
S. D a k .

Nebr.
Kans.

S. A.
Del.
Md.
D. C.

W. Va.
N. O,

Ga.
Fla.

Tenn,
Ala.

Miss.

w.s.c.
Ark.

La.

Okla.

Tex.
Mt.

Mont.

Idaho.
Wyo.

47.4

Colo.

50.0
35.3
13. 9
5. 4

N.Mex.
Ariz.
utah.
Nev.
Pac.
Wash.

Oreg.
Calif.

25

FAR.

213

THE

FORMAL TABLE : STRUCTURE

SEC.

2-B

A "parallel table" occupying an odd number of pages is a contradiction
in terms. Usually, it represents a divide table incorrectly handled or
is composite in form.
214. Narrow table as the norm.-The narrow table represents t h e
normal case from which all other types are derived. It is not only t h e
most common type; it is t h e simplest in basic pattern and most readily
understood. When defining all other types, the narrow table is
assumed as the point of departure.
In general, the deviations represented by t h e other types are determined by space requirements. Thus, if t h e number of columns (in an
assumed normal table) is too few for the page width, the table may
become a half-measure, doubled; third-measure, tripled; etc. If the
number of columns is too great for t h e page width, the table may become a narrow-divide, broad, or even parallel table, depending upon
other factors in construction and upon t h e size of the page.
SPECIFIC TYPE OF TABLES D E F I N E D (215-226)

215. Fractional-measure tables.-A form used when the table is
not wide enough to be spread across t h e page and is narrow enough to
double up (or triple up, etc.) both the s t u b and the columns. (See
fig. 4, examples 4 - A to 4 - D . ) T h e half measure, doubled, is the
typical case.
a. T h e half-measure, doubled, is set or typed half t h e width of t h e t y p e page and
doubled up with a parallel vertical rule between t h e two sections. Similarly, t h e
third-measure, tripled, is set one-third page width and tripled u p ; t h e fourthmeasure, quadrupled, is set one-fourth page width a n d quadrupled up, etc.
b. Both box and s t u b must be doubled up to fall into this group. If t h e s t u b
is not included in t h e doubling up, t h e result is a form of narrow table; it is not a
fractional-measure table.
c. T h e s t u b may be continuous or in repeated blocks. (See examples 4 - A and
4 - B , respectively.) If t h e s t u b is in repeated blocks, one or more blocks m a y
appear in each table section.

216. Narrow table.-The "normal" or "upright" table; t h e most
common and desirable type. It permits many variations in construction. (See fig. 5, examples 5 - A to 5 - E . )
a. It is called "narrow" or "narrow-measure" because printing runs across t h e
narrow dimension of paper. It contrasts with t h e "broad" or "broad-measure"
table in which printing runs t h e broad or long way of t h e page.
b. It is set or typed full width of t y p e page.
c. It has a single s t u b appearing at left and applying across entire table, and a
single boxhead (simple or compound) at t o p of table applying down full depth of
table or table page.
d. T h e narrow table with field spanners should not be confused with t h e narrowdivide table. (See example 5 - E . ) In a narrow-divide, t h e entire boxhead appears for each deck. (See fig. 6, examples 6 - A a n d 6-B.)

26

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

SEC.

2-B

STRUCTURAL TYPES

PAR.

217

217. Narrow-divide table.-This form is used typically when the
number of columns is found to be too great for the width of the page
but the stub is shallow, thus permitting the box and stub to be repeated below on the same page in the form of one or more additional
"decks." (See fig. 6, examples 6-A and 6-B.)
a. T h e table is broken about halfway across figure columns a n d t h e remaining
columns are brought under t h e first "deck" with both s t u b and box repeated.
T h e same process can be continued to create three or more decks.
b. T h e repetition of box for each deck includes t h e entire box, not merely t h a t
portion which differs from t h e box of t h e upper deck. Otherwise t h e table becomes
a narrow table with field spanners. (See fig. 5, example 5 - E . )
c. For each deck t h e parallel rule is repeated at t h e top of t h e box, and t h e
single rule below t h e box extends through t h e s t u b as at t h e top of t h e table.
d. T h e repetition of t h e s t u b includes t h e wording of t h e s t u b box.
e. In divide tables, t h e spanner at t h e top of t h e table applies only to t h e upper
deck. It has no more influence on t h e lower decks t h a n any portion of t h e boxhead
above one group of columns has on a n y adjoining group of columns which it does
not cover. Contrast this with a narrow table with field spanners (fig. 5, example
5 - E ) where t h e column heads at t h e top of t h e table apply all t h e way to t h e
bottom of t h e table, with t h e field spanners providing additional qualifications
for t h e affected blocks.

218. Multipage narrow table with continuous boxhead.-This is
not a formal structural type but it warrants special attention to avoid
confusion with the parallel table which it resembles. The common
form is the narrow table with a continuous boxhead (two or more pages
required to complete the boxhead classification) and a I-page stub
repeated on each continued page. (See fig. 7.) Basically, it is a
"divide" table with each individual page constituting a deck. Distinctions between this form and the parallel table are as follows:
a. P a g i n g . - I t may comprise an even or odd number of pages; need not appear
on facing pages; may start on either a right-hand or left-hand page. T h e parallel
table must comprise an even number of pages; must appear on facing pages;
must start on a left-hand and end on a right-hand page.
b. Table t i t l e . - T h e title is repeated separately on each page, followed by
" - C o n t i n u e d " on continued pages. In a parallel table, t h e title and headnote
run across, a n d are centered upon, each pair of facing pages; t h e title is followed
by " - C o n t i n u e d " on continued pairs of pages.
c. Stub. The s t u b is repeated on each page, always appearing in its normal
position at t h e left of t h e field. In a parallel table, t h e s t u b normally appears on
t h e left-hand page only; tracer numbers appear at t h e left of t h e even-numbered
(left-hand) pages and at t h e right of t h e odd-numbered (right-hand) pages. In
a parallel table with repeated s t u b (a special form) t h e s t u b on t h e odd-numbered
(right-hand) page is always at t h e right of t h e field, never at t h e left.
d. Specific r u l e . - I f t h e s t u b is placed at the left of t h e right-hand page, t h e
table is not a parallel table, and t h e title should not be centered across t h e facing
pages.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

27

PAR.

219

T H E FORMAL

TABLE ; STRUCTURE

sEc. 2-B

219. Broad t a b l e . - T h i s form is used when the table is too wide for
the normal width of the page, cannot be run as a narrow-divide,
and is not wide enough to make a parallel table necessary. (See fig. 8.)
The broad table should be avoided where possible since the publication
must be rotated to read it.
a. It is set or typed t h e long or broad way of t h e page.
b. I t s width is less t h a n normal page depth because t h e page folio line retains
its normal position (thereby running at right angles to t h e table) a n d must be
allowed for.
c. It is invariably placed so it can be read after t h e book is rotated 90 degrees
clockwise. T h a t is, regardless of whether it falls on t h e left-hand or right-hand
page, t h e top of t h e table is always at t h e left when t h e book is in t h e normal
upright position. T h e reason for this becomes apparent where a broad table
continues over several pages. Specifically If t h e t o p of t h e table is placed at t h e left (as it should be), t h e upper a n d lower
pages (facing pages) may be read in proper sequence from top to bottom. For
example, if t h e facing pages are pages 16 and 17, t h e rotation clockwise will place
page 16 (the left-hand page) on t h e top and page 17 (the right-hand page) on
t h e bottom.
If t h e t o p of the table is placed at t h e right, it will be found t h a t t h e bottom
page must be read before t h e t o p page if t h e statistics are to be read in proper
order. T h a t is, after t h e necessary rotation counterclockwise, page 16 will be on
t h e bottom and page 17 on t h e top, an unnatural arrangement for sequential
reading.

220. Broad-divide table.-Generally, this form is undesirable. The
figure columns are broken the same way as for a narrow-divide table,
but the table runs the broad way of the page. (Not illustratedrarely used.)
221. Parallel table: ° Normal c a s e . - T h i s form is used where the
table is so wide that it spreads across two pages with no room for
repeating the stub on the second page. It should be avoided where
possible. It is hard to read and the requirement that it appear on
facing pages complicates make-up and table sequence (fig. 9, examples
9-A and 9-B). The following statements refer to the normal case of
the parallel table. (See example 9-A.)
a. It is printed on facing pages with each pair of facing pages considered a
single u n i t ; must begin on a left-hand page.
b. T h e title and headnote run across t h e two facing pages.
2 Parallel is a term which, in printing, refers to the treatment given any pair of facing tabular pages whereby
a stub appears on the left-hand page, but is omitted on the right-hand page. Tracer numbers are run on
the outside of both left and right pages to aid the user to keep his place while reading across. However, in
the field of tabular presentation this term is descriptive of a table which consists solely of facing pages run in
parallel. Accordingly, in this manual, the term parallel table is limited to tables set entirely in parallel;
that is, in pairs of facing pages. (See par. 225.)

28

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEc. 2-B

STRUCTURAL TYPES

PAR.

221

c. T h e figure columns are broken at a convenient, logical, or physically practicable point. T h e columns to t h e right of t h e division are printed on t h e righthand or odd-numbered page.
d. T h e s t u b appears only at t h e left edge of t h e left-hand (even-numbered)
page. Tracer numbers are assigned to each entry line if t h e s t u b is not repeated
on t h e facing page. T h e tracer numbers appear at t h e left of t h e s t u b on t h e
even page and on the right-hand side of t h e odd or facing page (example 9-A).
e. Note: If the stub is repeated in t h e normal position at the left side of the odd
or facing page, the table is a multipage narrow table (fig. 7 ) . It is not a parallel
table and the title and headnote should not be run across the two facing pages.

222. Parallel table: Special t y p e . - I n a special type of parallel
table, the stub is repeated at the right-hand side of the right-hand
(facing or odd-numbered) page (fig. 9, example 9-B.). In this case,
tracer numbers are not used.
This form (with repeated stub placed at the right) should be used
sparingly. I t s justification rests upon the importance of horizontal
continuity across the page break, not upon space demands. Obviously, if the stub appears at all, there is as much room for it on the
right as on the left.
a. It is most justified where it is imperative that the specific columns on either
side of t h e space 'break be interpreted as a coordinated unit.
b. It is rarely justificd where the columns separated by the page break are
components of a simple coordinate series. Here, the additional separation caused
by insertion of the repeated stub in normal position on the facing page is probably
less harmful t h a n t h e necessity of tracing the line back from a s t u b appearing at
t h e right-hand edge of the page.
c. It is least justified where the d a t a presented on each page constitute independent, though related, presentations, such as a page of data for white families
facing a page of data for Negro families.
This injunction is not intended to discourage presentation of corresponding
figures for related universes on facing pages. In itself, this is a sound practice.
However, this objective may be accomplished by placing t h e s t u b on both pages
in its proper position and providing a continued title on t h e right-hand page.
In t h e example cited above (white families and Negro families), confusion may
arise if t h e data for white and Negro families require two pages for each race
group. Here, the parallel presentation will require an examination of alternate
pages to obtain white d a t a and alternate pages to obtain Negro data. In effect
this means t h a t alternate pages of two distinct tables have been interleaved.

223. Parallel-divide table.-This form is the same in principle as
the narrow-divide but the decks run all the way across the two facing
pages. Tracer numbers are used as in parallel tables. (Not illustrated-rarely used.)
224. Tables without stubs.-This is a special form used largely for
one-dimension listings; that is, listings intended to be read vertically
only, not horizontally. Such tables, usually confined to text, may


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

29

PAR.

224

T H E FORMAL TABLE ; STRUCTURE

SEC.

2-B

run full width of the text column; or they may run full width of the
page, across both columns. However, if t h e table has only a few
columns, it may be run indented on both sides, thus:
TABLE

A.-

H O U S E

N U M B E R S ON

East side

West side

5101
5103
5105
5109
5113
5115
5117
5119
5121
5123

5102
5104
5106
5108
5110
5112
5114
5122
5124
5128

OAK

STREET

225. Composite f o r m s . - T h i s group comprises tables which include
features of several of the standard types as outlined above. Thus, a
table with three decks m a y start out as a narrow-divide and end as a
third-measure, tripled. Again, a 3-page table may begin as a narrow
table (first page) and have t h e next two pages (facing each other) set
in parallel. Just as t h e first example is not a narrow-divide table, as
such; neither is t h e second a parallel table, in the sense that t h e term
is used here. Both tables are composite.
In general, composite forms are undesirable because of t h e need for
the user to find his way through a shifting pattern. In practice, they
sometimes are necessary, particularly in mass-production work where
the possibilities for space-saving (an important factor in printing costs)
must at times outweigh the reading problems of the user.
226. Tables in r u l e s . - I n general, tables should not be enclosed in
rules; t h a t is, vertical side rules should not be added. T h e practice
of enclosing an entire table in a box (including title, headnote, and
footnotes) also should be avoided.
3 A possible exception is the use of vertical side rules to set off tables without stubs. (For example, see
G P O Style Manual, p. 133.) Apparently, the'purpose of the vertical side rules is to warn the reader that
the first column is not a stub column. The style of the Bureau of the Census does not include use or side
rules in such instances.

30

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Chapter 3

THE TABLE TITLE (301-374)
Sec. 3-A. General (301-309)
301. Definition.-The table title, placed at the top of the table, is a
brief statement of the nature, classification, and time reference of the
information presented, and the political division, geographic area, or
physical plant to which the statistics refer. These points are sometimes referred to as the " W h a t , " "How classified," "Where," and
"When" of the table.
302. Function. The title's function is twofold. It provides a
catalog of table content and serves as a purpose indicator and reference
guide. The wording and arrangement of the component parts of the
title depend upon the comparative significance of these two factors in
the given presentation.
303. Component parts or "title segments."-Each title should
indicate, specifically or by implication, four basic characteristics of
the tabular data. (See below.) The statements on these points, and
other points peculiar to the given table, are joined together by appropriate connectives and punctuation to form the title as a whole.
These component parts may be conveniently referred to as "title
segments."
a. Universe s e g m e n t . - S p e c i f i e s t h e n a t u r e and l i m i t s of t h e group or subExamples: W 1 T E EMPLOYED
group covered by t h e table. (See sec. 3 - C . )
W O R K E R S ; G E N E R A L R E V E N U E ; I M P O R T S OF M E R C H A N D I S E ; D O M E S T I C A N I M A L S

ON FARMS; MATERIALS CONSUMED; INDEPENDENT TORES; etc.
b. Classification s e g m e n t . - S p e c i f i e s how t h e universe d a t a are classified and
cross-classified. (See sec. 3 - D . ) Examples: BY AGE AND SEX; BY SOURCE; BY
COMMODITY GROUPS AND ARTICLES; BY K I N D A N D AGE GROUPS; BY KIND,
QUANTITY, AND COST; BY SIZE OF STORE; etc.
c. Area or space s e g m e n t . - S p e c i f i e s t h e political division or subdivision,
geographic area, or physical p l a n t to which t h e d a t a refer. (See sec. 3 - E . ) Examples: FOR THE STATE; FOR CITIES OF 500,000 OR MORE; FOR THE UNITED
STaTEs; etc.
NOTE.-Throughout this manual, examples of titles appear in caps and small caps, the style of the Bureau
of the Census appropriate for document-size pages. For discussions of capitalization, indent, and type
size for table titles, as such, see paragraphs 308 and 309. In general, titles would appear in caps on censussize pages, except the word "Table" would be in caps and lower case. (See fig. 10, p. 34.)


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

31

rAR.

303

THE

TABLE TITLE

sEC.

3-A

d. Time-refere nce s e g m e n t . - S p e c i f i e s t h e point in t i m e to which t h e d a t a
refer. M a y be a single d a t e , a t i m e period, or a series of d a t e s or t i m e periods.
(See sec. 3 - F . ) Examples: 1940; 1935 To 1940; 1940 AND 1930; 1910, 1920, AND
1940; WEEK or MARcn 24 To 30, 1940; etc.

304. Relationship between content and arrangement.-The wording
of each title segment reflects the content of the table. The order in
which they appear reflects the purpose or conditions of presentation.
Hence, the cataloging function of the title is fulfilled by the segment
wording; the reference-indicator function is provided for by the title
arrangement.
For a given table there may be a number of equally correct arrangements of the title. A primary problem in title writing involves selection of the most suitable arrangement for the given purpose and under
given presentation conditions. Only rarely is there a single title
wording or arrangement that is "correct" for all purposes and under
all conditions.
Example: Assume a table showing the population by age, race, and
sex, each cross-classified with the other. The following listing shows
several possible titles for such a table. Each of these is "correct"
under the given conditions although the tabular content is identical
for all these tables.
a. AGE, BY RACE AND SEX
H e r e , t h e purpose of t h e t a b l e is to show an
a g e distribution by race a n d sex. N o t e , however, t h a t if a g e d a t a were shown in
all tables in t h e publication, t h e difference between tables might well be stressed
by m a k i n g t h e title read RACE AND SEX, BY AGE
b. RACE, BY AGE AND S E X . . . An a p p r o p r i a t e title if t h e t a b l e were pres e n t e d as p a r t of a discussion of racial characteristics in o r d e r to s h o w a g e a n d
sex differentials for each race.
C. NEGROES BY AGE AND SEX, WITH COMPARABLE DATA FOR TOTAL POPULATION AND OTHER RACE GROUPS . . . H e r e , t h e s a m e d a t a a r e being used in a report
on t h e Negro population. T h e interest lies in showing d a t a for Negroes w i t h t h e
o t h e r information inserted primarily to lend significance to t h e Negro statistics.

.

305. Treatment of individual segments.-Both construction and
content of individual segments will vary widely in accordance with the
customs and needs of the general field, the nature of the given report,
and differences in content among the tables in the group and the
content of the given table.
T h e various analytical factors are too numerous and varied to permit establishing specific rules or even to provide a series of specific
examples. In sections 3-C to 3-F (following) a few observations are
made concerning practices generally in vogue. In addition, a few
general statements on content are included.
306. Coverage and brevity.-Whether the title be written as a
catalog of content or as a reference guide, and irrespective of the title-

32

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEc. 3-A

GENERAL

PAR.

306

writing scheme employed, it must reflect the material shown in the
table.
a. Content may be understated; never overstated.-The table should contain
t h a t which t h e title advertises. T h e table may contain more; it should never
contain less. T h e outstanding exception to this is where t h e headnote is used to
qualify t h e title, but its use to qualify overstatement should be kept to a minimum.
Hence, if a question arises concerning overstatement or understatement, t h e only
safe course is understatement.
b. Titles should be brief, but adequate.-Bureau of Census table titles are
frequently criticized because they are (1) too long, and (2) not sufficiently descriptive. To an extent, both criticisms are justified. More important, however, is
t h e basic conflict exemplified in these criticisms.
Title brevity and completeness of coverage can be combined easily for simple
tables; t h a t is, those with few classifications. It is not easy for t h e more complex
materials where t h e difference between tables in a series may lie in exact detail
of classification for a closely defined subgroup; or t h e comparative significance of
t h e d a t a m a y depend upon t h e combinations shown, rather t h a n upon t h e individual classifications as such.
Minor sacrifices in coverage are warranted in order to shorten titles. In general, a less complete title which will be read is better t h a n a complete title which
will be ignored because of length or complexity.

307. Title wording.-Table titles are written in telegraphic style,
not in complete sentences. Verbs are commonly omitted, as are articles and other parts of speech not absolutely essential to understanding. Classification statements (except those relating to area) are
expressed in the singular unless the plural form is essential to clarity.
Standard conventionalized phrasing should be used wherever possible.
(See par. 373 for terminology in percentage titles.)
All of these factors, and other peculiarities of title writing, place a
heavy premium ona. Careful and consistent use of terms employed,
b. Meticulous phrasing, a n d
c. Precise punctuation.

308. Capitalization and indent of titles.-In general, the title of the
statistical table should be displayed in a distinctive manner. This
serves to emphasize its function and to differentiate it from the headnote and the remainder of the table.
a. Capitalization.-Census Bureau practice is characterized by t h e use of full
capitals (caps), or of caps and small caps, for t h e title proper, with t h e choice made
dependent upon t h e width of t h e given table. ( T H I S S T A T E M E N T IS IN
F U L L CAPS; TH1s Is 1N CAPS AND SMALL CAPS.)
Where full caps are used, t h e word "Table" is shown in lower case and t h e table
number appears in boldface, where available, as Table 5 . - P O P U L A T I O N . . ..
Where caps and small caps are used, t h e word "TABLE" is also printed in caps
and small caps, with boldface for t h e table number, as TABLE 5.-POPULATION.• •


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

33

•

PAR.

308

T H E TABLE T I T L E

SEC.

3-A

Where small caps a n d boldface are not available, as where t h e standard typewriter is employed for typewriter-offset work, full caps are used for t h e title
proper; t h e word "Table" is shown in lower case; and t h e table number is in
s t a n d a r d numerals, as Table 5. -POPULATION. . . . . (See fig. 10, examples
A - 4 and B-4.)
b. Indent.-In Census Bureau practice, t h e indent of titles is dependent upon
t h e number of lines in t h e given title. (See fig. 10.)
(1) One-line t i t l e . - C e n t e r over table.
(2) Two-line t i t l e . - R u n t h e first line flush to edge of table, both left and
right; center t h e second line.
(3) Three-line t i t l e . - T i t l e s occupying more t h a n two lines are undesirable
since they are hard to read. However, where a three-line title appears, t h e first
line is run flush, left and right. T h e second line is indented 2 ems (4 typewriter
spaces) at t h e left and runs flush to t h e edge of t h e table at t h e right. T h e third
line also is given a 2-em indention at t h e left; it is permitted to run short (left
dangling) at t h e right.
c. Division of w o r d s . - I n titles, as in text material, word divisions are appropriate at t h e end of title lines, as necessary. Normal rules for syllabication should
be applied.

F I G U R E 1 0 . - T A B L E T I T L E S : C A P I T A L I Z A T I O N AND I N D E N T

Style A: On tables less than 30 picas (5 inches) in width as printed:
Example A - 1 . One-line title:
T A B L E 1 5 . - P O P U L A T I O N OF T H E U N I T E D S T A T E S :

1790

To

1940

Example A - 2 , Two-line title:
TABLE

16.-AGE

OF T H E P O P U L A T I O N , BY M A R I T A L S T A T U S , R A C E , A N D S E X ,
F O R T H E U N I T E D S T A T E S : 1940 A N D 1930

Example A - 3 . Three-line title (letterpress) :
T A B L E 1 7 . - A G E OF T H E P O P U L A T I O N , BY M A R I T A L S T A T U S , R A C E , A N D S E X ,
F O R T H E U N I T E D S T A T E S , U R B A N A N D R U R A L , A N D F O R C I T I E S OF 100,000
I N H A B I T A N T S OR M O R E : 1940

Example A - 4 . Three-line title (typewriter-offset) : Note t h a t all lines except t h e
first are indented 4 typewriter spaces. The style shown here is identical with
t h a t shown in example B - 4 in Style B; the only difference is in t h e width of
the title; that is, in the assumed width of the table.

Table 18.-AGE OF THE POPULATION, BY MARITAL STATUS, RACE, AND
SEX, FOR THE UNITED STATES, URBAN AND RURAL, AND FOR CITIES
OF 100,000 INHABITANTS OR MORE: 1940

34

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Figure 1 0 . - T A B L E TITLES: CAPITALIZATION AND INDENT-Con.
Style B: For titles of tables 30 picas (about 5 inches) or more in width as printed:
Example B-1. One-line title:

Table 1 5 . - P O P U L A T I O N OF T H E UNITED STATES: 1790 TO 1940

Example B - 2 . Two-line title:

Table 1 6 . - A G E OF T H E POPULATION, BY MARITAL STATUS, RACE, AND SEX, FOR T H E UNITED STATES,
URBAN AND RURAL: 1940 AND 1930
Example B - 3 . Three-line title (letterpress):

Table 1 7 . - A G E OF T H E POPULATION, BY MARITAL STATUS, RACE, AND SEX, FOR T H E U N I T E D STATES, URBAN
AND RURAL, FOR PRINCIPAL METROPOLITAN DISTRICTS, AND FOR CITIES OF 100,000 INHABITANTS OR
MORE: 1940
Example B - 4 . Three-line title (typewriter-offset) : Note t h a t all lines except t h e first are indented 4 typewriter spaces. The capitalization shown should be employed on all typewriter-offset work where small caps a n d bold face are not available, irrespective
of width of table. T h e line-up at right is achieved by insertion of extra typewriter spaces between words as necessary to make
lines come out even; this technique should not be used for t h e last line, which is always permitted to run short.

Table 18.-RELATIONSHIP TO HEAD OF HOUSEHOLD OF NATIVE WHITE WOMEN 18 TO 64 YEARS OLD, BY LABOR FORCE
AND EMPLOYMENT STATUS AND NUMBER OF CHILDREN UNDER 10 YEARS OLD, FOR METROPOLITAN DISTRICTS, AND
FOR SPECIFIED CITIES: MARCH 1940

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

PAR.

309

THE

TABLE TITLE

sE0. 3-A

309. Type s i z e . - W h e r e different type sizes are available, table
titles are normally shown in type one size larger than the size of type
employed in the table proper. Thus, at the Bureau of the Census,
8-point type is usually employed for titles of 6-point tables.
Sec. 3-B. Methods of Title Writing (311-314)

311. Three basic methods.- Various methods of title writing have
been devised, distinguished largely by the mechanical order or placement of the title segments. The objective of each is the same; that is,
to emphasize the content and purpose of the table and the difference
between one table and another.
"Difference stress" is comparatively simple when only the universe
or the major classification is involved. The difficulty increases with
classification complexity; or when the difference is a function of area
or time, largely because the "where" and "when" are usually placed at
the end of the title.
Two methods of title writing are most commonly used in Bureau of
Census publications: (a) The conventional, and (b) the subject-phrase.
Although different in detail, both methods use a single unified title.
A third method, restricted largely to specialized fields, particularly
tables of the Census of Business, is (c) the separated-unit scheme,
characterized by a physical separation of one or more segments from
the main title.
The outstanding general example of the separated-unit method of
table titling is the use of a subtitle to carry a portion of the title
burden. The subtitle is not normally used by the Bureau of the
Census, largely because of the demands it makes on vertical space.
For discussion and examples of the use of subtitles in general statistical practice, see Walker, Helen M., and Walter N. Durost, Statistical Tables, Their Structure and Use, Teachers College, Columbia
University, 1936, p. 22; and Mudgett, Bruce D., Statistical Tables
and Graphs, Houghton Mifflin Company, 1930, p. 45-46.
These methods are described below in terms of construction of the
title as a whole. More specific questions concerning which title
segment should be emphasized are discussed in the following sections
(secs. 3-C to 3-F) in this chapter.
312. Conventional method.-So-called because most commonly used
throughout the statistical world. In its simplest form it merely
expresses the "What, how classified, where, and when" in that order,
as POPULATION BY AGE, RACE, AND SEX, FOR THE UNITED STATES:
1940. Frequently, the "how classified" is split, as AGE OF THE
POPULATION BY R A C E AND S E X

36

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEc. 3-B

METHODS OF TITLE WRITING

PAR.

312

a. Advantages and disadvantages.-This method has several signal advantages,
of which the second and third (in order of listing) result from its frequency of use:
(1) Flexibility.-Ideal for small simple tables, t h e conventional method is
capable of considerable manipulation and expansion to cover the more complex
tables. By careful selection of the opening phrase, many of the advantages of
the subject-phrase method may be attained without the attendant difficulty of
expressing relationship between the subject-phrase and the remainder of the
title.
(2) Easiest m e t h o d to learn and to handle.-Since most tables are titled
by the conventional method, the beginning title-writer is able to draw upon his
own experience as a user. Minor errors in judgment concerning arrangement
are unlikely to cause confusion since the average reader is familiar with the
method.
(3) Conforms to normal expectation of u s e r . - T h e user is accustomed
to titles prepared in t h e conventional manner and takes their construction principle for granted. This aids him in noting differences and grasping content.
Coupled with these advantages are several distinct disadvantages which other
systems attempt to overcome. These disadvantages may become acute in involved
series of tables, particularly if t h e title work is performed hastily or without
proper regard to intertable relationships.
(4) Danger of undue title l e n g t h . - T h e conventional method readily
lends itself to long and complicated titles which t h e reader tends to ignore rather
t h a n read.
(5) M a y obscure i m p o r t a n t differences.'-While it easily handles table
differences confined to a single title segment, t h e conventional method may
obscure differences involving two or more segments. Differences in area segment are particularly hard to bring out since t h e area segment normally appears
near t h e end. If t h e only difference between two or more tables lies in t h e difference in area covered, t h e reader m a y be faced with seemingly identical titles
with t h e difference between them submerged.

313. Subject-phrase or key-phrase method.-So-called because it
invariably starts t h e title with a key-phrase set off from t h e remainder
by a dash or colon. This permits increased emphasis on purpose of
presentation and difference between tables. Examples # follow :
1 Occasionally, an attempt is made to emphasize important words of the title by using boldface, capital
letters, or other contrasting type. This practice, rarely used by the Bureau of the Census, requires cautious
handling lest oddity in appearance distract, rather than aid, the reader. For examples of limited use of
bold face type for emphasizing table titles, see U . S . Works Progress Administration, Workers on Relief in
the United States in March 1935, Washington, D. C., 1939. For intensive i s e of capital letters for emphasis
of important words in table titles, see annual issues of Statistical Abstract of the United States, 1878 to 1904.
2 The conventional method would tend to submerge the main subject of the tables and the intertable differences as follows:

T A B L E 1. - A C R E A G E , PRODUCTION, AND VALUE OF POTATOES, BY STATES:
T A B L E 2 . - A C R E A G E , PRODUCTION, AND VALUE OF CORN, BY STATES:

1930

1930

TO

TO

1941

1941

TABLE I 1 . - N A T I V E W H I T E W O M E N BY EDUCATION OF WOMAN, MARITAL STATUS, AND A G E , FOR THE
UNITED STATES: 1940
TABLE 1 2 . - N A T I V E WHITE W O M E N BY EDUCATION OF WOMAN, MARITAL STATUS, AND A G E , FOR THE
UNITED STATES: 1930


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

37

PAR.

313

TABLE

1.

THE

TABLE TITLE

SE0.3-B

P O T A T O E S - A C R E A G E , PRODUCTION, AND V A L U E , BY STATES:

1930

TO

1941
TABLE 2.-CORN-

ACREAGE, PRODUCTION, AND V A L U E , BY STATES:

T A B L E 1 1 . - E D U C A T I O N OF W O M A N ,

1940:

1930

TO

1941

N A T I V E W H I T E W O M E N BY M A R I T A L

STATUS AND A G E , FOR THE U N I T E D STATES

T A B L E 1 2 . - E D U C A T I O N OF W O M A N ,

1930:

N A T I V E W H I T E W O M E N BY M A R I T A L

STATUS AND A G E , FOR THE U N I T E D STATES

a. Advantages and disadvantages.-The advantage gained by giving greater
prominence to the key phrase carries with it several disadvantages which should
be taken into account. Both advantages and disadvantages are given below:
(1) R e fere nce g u i d e . - T h e separation of key phrase facilitates reference.
(2) Most useful when presenting involved s e r i e s . - T h e key-phrase
method tends to be most successful in complex presentations, as in an involved
series of complicated tables; t h a t is, at t h e point where t h e conventional method
is weakest. It is least useful for a table standing alone, or in a simple series,
where t h e conventional method is strongest.
(3) Requires skillful h a n d l i n g . - A s compared with t h e conventional
method, more skill and experience is required when using t h e key-phrase
method, particularly to maintain parallel treatment between tables in a series.
T h e seeming ease of selection (and wording) of a key phrase is deceptive. Poor
selection can make a well-organized series of tables look disconnected when t h e
titles are listed in a table of contents, or when t h e reader is leafing through t h e
report. Finally, t h e relationship between t h e key phrase a n d t h e remainder
of t h e title m a y be hard to express because t h e break in continuity tends to
obscure back-reference to t h e key phrase.
(4) Brevity of k e y phrase e s s e n t i a l . - I f t h e key phrase is not kept short,
it will defeat its own purpose. Maximum length should be one-half of t h e first
line; better if kept to one-fourth of t h e first line. A " k e y phrase" longer t h a n
t h e remainder of t h e title usually means bad selection.
(5) Danger of long t i t l e s . - I f mishandled, t h e key-phrase method may
result in even longer titles t h a n t h e conventional method. A common fault
leading to t h i s is t h e complete repetition of t h e " k e y phrase" in t h e remainder of
t h e title. A key phrase tacked-on at t h e front of. a conventional title merely
distorts both methods with dubious advantage to t h e user.

314. Separated-unit method.-See facing page.

38

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

314. Separated-unit m e t h o d . - T i t l e s from the Census of Business
may be used as examples. T h e separated-unit method is so-called
because certain title segments are physically separated from the
title proper and relegated to a subtitle, to an area-indicator or class-

indicator at one side, or distributed between both. This method
provides a separate observation point for primary features of the
title or for points of differences between tables.
Examples:

Table 1. - C I T I E S OF 25,000 I N H A B I T A N T S OR M O R E
E S T A B L I S H M E N T S , R E C E I P T S , P E R S O N N E L , A N D P A Y R O L L , BY K I N D S OF B U S I N E S S

ARKANSAS

Table 2 7 . - M O N T H L Y E M P L O Y M E N T A N D S E X OF E M P L O Y E E S
BY GEOGRAPHIC DIVISIONS AND STATES

0

_BARBER SHOPS

a. Advantages and disadvantages:

CD

(1) M o s t s u i t a b l e for s u m m a r y t i t l e s . - T i t l e segments
must be brief; hence this scheme is most effective for multipage
tables of simple classification, or where the title specifies t h e
general field covered rather t h a n catalogs detailed content.
( 2 ) Least s u i t a b l e f o r i n t e n s i v e a n a l y s i s . - W h e r e a series
of tables are presented which explore a given field or subgroup
intensively, this scheme is least adaptable because the differences
between tables m a y be a m a t t e r of cross-classification detail.
Such differences are frequently difficult to express except within
title context; hence they do not separate readily.
( 3 ) D e m a n d s u n i f o r m i t y of t r e a t m e n t . - T i t l e s written by
t h e separated-unit method m a y confuse the reader if t h e pattern
is not maintained rigorously. In general, a shifting of content


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

pattern between set-off indicators, subtitle, and title may distract
t h e reader as he moves from table to table.
( 4 ) Space c o n s u m i n g . - T h e arbitrary division of the space
above t h e table into sections is space-consuming and complicates
typesetting a n d typing. Thus, in table 27, above, the main
title is forced into two short centered lines to avoid conflict with
t h e class-indicator. If a subtitle also were used, still more space
would be required and t h e width and number of lines of t h e subtitle would be determined by t h e same factor. This type of title
is most practicable, therefore, for tables presented on large pages.
Thus, this example is r u n broadside because these titles cannot
successfully be presented across t h e width of a document-size
page.

0

PAR.

321

THE

TABLE TITLE

sEc. 3-C

Sec. 3-C. Universe Segment (321-325)

321. Definition.-The universe segment is that part of the table
title specifying the nature and limits of the group or subgroup covered
by the table. Examples: W H I T E E M P L O Y E D W O R K E R S ; G E N E R A L
R E V E N U E ; I M P O R T S OF M E R C H A N D I S E ; D O M E S T I C A N I M A L S ON F A R M S ;

etc.
322. Placement.-Usually appears as the first or second segment.
The universe segment precedes the segments on time and area.

MATERIALS CONSUMED; INDEPENDENT STORES;

a. Universe s t r e s s . - P l a c e m e n t of this segment at beginning of title is suitable
1n

(1) A series of tables in which each table involves a different basic universe.
(2) A table standing alone where t h e purpose of the presentation is to provide
general information on t h e universe without stressing a n y particular characteristic.

323. Curtailment of universe segment by use of headnote. A long
universe description may be shortened by relegating part of it to the
headnote. The headnote, or that portion of it, then qualifies or defines the title. This technique demands careful handling to avoid
misleading the user. Thus, in a given series of tablesa. Universe qualifications common to all tables m a y be placed in t h e headnote
a n d omitted from t h e title; but
b. Universe qualifications distinguishing one table from another should not be
placed in the headnote; they should be kept in t h e title.
Example: A presentation is confined to 3 tables, presenting respectively data
for total, white, and nonwhite workers 25 to 34 years old. Here, if desired, t h e
age qualification could be moved from t h e title to t h e headnote in all tables,
with t h e headnote reading "[Workers 25 to 34 years old]," since this universe
qualification is identical for all tables. It would not be permissible to relegate t h e race qualification (total, white, nonwhite) to t h e headnote, since t h a t
qualification differs for each table.

324. Omission of universe segment or portion thereof.-Where the
basic universe is identical for all tables in a series, and the nature of the
universe is made clear in the report title or running head, the universe
statement may be omitted completely from the title and headnote.
325. Stating unit of presentation.-If the universe segment (or some
other portion of the title) does not specifically indicate the unit of presentation, it is imperative that it be made clear at some point in the
table. (See pars. 512a and 1225a, and secs. 9-F and 14-C.)
Sec. 3-D. Classification Segment (331-336)

331. Definition.-The classification segment is that part (or parts)
of the title specifying how the universe data are classified and crossclassified. Examples: BY A G E A N D S E ; BY S O U R C E ; BY C O M M O D I T Y

40

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEc.

3-D

CLASSIFICATION S E G M E N T

PAR.

332

G R O U P S A N D A R T I C L E S ; BY K I N D A N D A G E G R O U P S ; BY K I N D , Q U A N -

etc.
332. Placement.-May precede or follow universe segment, or a
portion may be put in front and t h e remainder behind the universe
segment. T h e classification segment precedes t h e time and area
segments.
T I T Y , A N D C O S T ; BY S I Z E OF S T O R E ;

a. Classification s t r e s s . - P l a c e m e n t of classification segment at beginning of
title is most suitable in(1) A series of tables with the same or similar universe where the distinction
rests on differences in classification of data.
(2) A table standing alone, such as any special purpose table, where the purpose of presentation is to stress a particular classification aspect of t h e data.

333. Splitting of classification segment.-When cross-classifications
are shown (such as population by age, race, and sex), t h e classification
considered most important is frequently placed before the universe
segment and the remainder placed afterward. Two factors should be
taken into account in thus splitting t h e segment:
a. Permits stressing of one classification feature.-Here, t h e most important
classification (in terms of table purpose or table difference) is placed at the front
of the title, as AGE OF THE POPULATION BY RACE AND SEX.
b. Permits stressing major cross-classification.-Splitting t h e segment indicates automatically that the characteristics preceding the universe segment are
classified by all characteristics following t h a t segment. Nonstatistical users tend
to be confused by the multiple use of " b y " to indicate cross-classification. Thus,
AGE OF THE POPULATION, BY RACE AND SEX is less confusing t h a n POPULATION
BY AGE, BY RACE AND SEx and means t h e same thing.
(See par. 336.)

334. Showing variation in classification for two groups.-The extent
to which varying detail of classification should be shown in the title
depends largely upon whether the primary purpose of the table is such
as to make detailed description essential. Where possible, the title
should be kept short by indicating no more than t h e basic classifications. If this results in an overstatement, however, the situation
should be spelled out in more detail in the title or a qualification should
be placed in t h e headnote.
Example: In a given table, statistics for 1940 are shown for age
by race and sex, and 1930 statistics for age are shown only by race.
Here, three general solutions are possible:
a. A G E
, BY R A C E A N D S E X , 1940, A N D A G E BY R A C E , 1930, F O R T H E
Gives the 1940 picture in full with an additional statement for the 1930 picture.
Normally, in Bureau of Census practice, when figures are shown for two censuses,
the data for the current census are stressed in the title.
BY R A C E , 1940 A N D 1930, A N D BY S E X , 1940, FOR T H E
b. A G E
Describes the common situation in full and adds a statement on the sex factor
for 1940. This title stresses the comparable material rather t h a n the most recent
information.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

41

PAR.

334

sEc. 3-D

THE TABLE TITLE

C. AGE OF THE POPULATION, BY RACE, FOR THE UNITED STATES: 1940 AND 1930.
Includes in t h e title only the information common to all groups. This is undesirable in this instance since the d a t a for the current census should ordinarily be
stated in full; more liberties can be taken when describing d a t a for a previous
census which are included for the reader's convenience. This type of t r i m m i n g
of title would be more appropriate, therefore, if it were the 1930 data t h a t were
understated.

335. Titles of "composition tables." Sometimes called "General
Characteristics Tables," tables of this type are commonly used in
census work to bring together summary or limited information for a
given universe or area. Characteristically, this results in a table
showing limited information for many different universes or subuniverses with only a general relationship.
When all groups (and their classifications) cannot be mentioned in
the title, the title may be generalized as GENERAL CHARACTERISTICS,
INDUSTRIAL CHARACTERISTICS, COMPOSITION OF, etc.
a. A given "composition" title may be used only once in a given s e r i e s . - T o
avoid confusion, a GENERAL CHARACTERISTICS title may be used only once for
a n y given universe or area in t h e same report. Two separate tables for the same
area entitled, for example, GENERAL HOUSING CHARACTERISTICS, and containing
altogether different groupings of housing d a t a for the same time period for t h a t
area, would confuse t h e reader.
b. Partial solution when two "composition tables" shown.-Ordinarily, when
two tables of this type are required for a given area, a common general feature
can be found in the groups shown in each as distinct from t h a t in t h e other.
If so, t h a t feature should be stressed in each title. If this proves impracticable,
and t h e tables cannot be merged into a single table, a partial (but undesirable)
solution may be found by specifying t h e first groups listed in each table, followed
by "etc.," as in the following examples:
TABLE 4.-OCCUPANCY, TENURE, TYPE OF STRUCTURE, ETC., .
TABLE 5,-NUMBER OF ROOMS, LIGHTING EQUIPMENT, SIZE
BY C O U N T I E S
ETc., .

. BY COUNTIES
OF

HOUSEHOLD,

336. Limiting explanation of cross-classification.-Two different
objectives, frequently conflicting, are involved in title description of
cross-classification: First, strict statistical accuracy of statement;
and second, provision of the essential clue to lead the user to the right
table for his information. In simple cases, both of these objectives
can be met without resulting in undue length or complexity of title.
Where multiple cross-classification is involved, particularly where
classification variation appears within the table, insistence on strict
statistical accuracy may destroy the usefulness of the title as a guide.
(See also par. 373 ff. in respect to percentage descriptions in titles.)
a. Statistical accuracy versus reader-patience.-Statistical accuracy in a title
is of dubious value if the reader cannot grasp t h e meaning or loses patience with it.

42

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

SEC.

3-D

CLASSIFICATION SEGMENT

PAR.

336

Omission of cross-classification detail is desirable if it means that the title will
become more useful.
T h e ability to trim the classification segment of t h e title may depend largely
on whether "this table" is the one in tbe series to which the reader would naturally
refer for a given classification, even though it is not mentioned in t h e title.

Example A: Universe t a b l e s . - I f the presentation is confined to one table
each for several distinct universes, minimum specification of detail is
required in the titles. "Composition" titles reflect this situation. The
reader interested. in some aspect of a n y one universe will naturally t u r n to t h e
table for t h a t universe. If his particular information is not there, he will not
expect to find it elsewhere in the presentation. This works both ways, however;
t h a t is, it is important to specify clearly in the title any feature appearing in
t h e table which t h e reader would not normally expect to find there.
Example B: Variation in color d e t a i l . - I f , for counties, only one
table is presented on marital status by color, it is not essential to specify in t h e
title t h a t marital status for nonwhite is shown separately only for certain of t h e
counties. Such a qualification m a y best be placed in t h e headnote and t h e
title worded simply BY CoLoR. T h e reader will expect to find in this table all
t h e "color" information presented on the subject and will t u r n to it automatically, whether or not t h e detailed qualification is specified in t h e title.
b. Restrict use of "by" to major cross (frequency tables).-Where possible,
use " b y " to indicate t h e major cross-classification only. Let t h e reader look at t h e
table to observe t h e detailed cross-classification. This is not desirable, of course,
where t h e distinction between tables lies in t h e exact detail of cross-classification.

Example C; AGE, BY RAcE, BY S E x . - T h i s phrasing is unnecessarily
complex if t h e table is t h e only one in t h e report presenting age d a t a by race
and sex, or if it reflects t h e "normal" pattern of t h e report. T h e real question is
"Would t h e reader seek some other table for additional detail if t h e title were
phrased: AGE, BY RACE AND SEX, Or even AGE, RACE, AND SEX"? If not, t h e
multiple use of " b y " m a y be an ill-advised nicety not likely to be appreciated
by t h e using public.
Example D-1: AGE, Br RACE AND Br S E x . - H e r e t h e table title shows t h a t
age is crossed separately by race and by sex, but not by both simultaneously.
Such a distinction is desirable only if it constitutes a deviation from a standard
p a t t e r n ; t h a t is, if it is an unusual feature. I t s principal value lies in its being
different from AGE, BY RACE AND SEx, so t h a t t h e reader may reasonably
assume t h a t the table must be different also.
Example D-2: A G E , F O R EACH RACE AND SEX GROUP.-This is another
approach to the same title problem. This expression may be considered more
specific t h a n that in example D - 1 , above, but it may leave t h e reader in doubt as
to whether t h e age distribution is shown for t h e totals (all races combined and
both sexes combined).
In a n y case, these distinctions tend to be over-subtle a n d should not be relied
upon too heavily. If the distinction needs to be made to point out inter-table
differences, then make it. Otherwise use t h e briefer form, even though it may be
considered statistically incomplete.
c. Use of " b y " when "universe total" is not included in c r o s s i n g . - O n e of
t h e most common title problems arises where only certain segments (but not t h e
total) of group A are shown classified by B.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

43

PAR.

336

THE

TABLE TITLE

SEC.

3-D

Problem: Table 4 shows an age distribution for t h e total population (not by
sex) with accompanying percentage distributions. Table 5 then shows t h e
d a t a for male and female separately, but not for t h e total. Should t h e classiBY
fication in table 5 (male and female table) be described as A G E OF
SEx; t h a t is, should t h e same terminology be used as would be used if both
tables were combined? Would not t h e correct title for table 5 be MALE AND
FEMALE
BY A G E ; S P E C I F I E D SEX G R O U P S OF
BY A G E ; OR A G E OF
OF E A C H SEX?
Solution: This title distinction is useful where t h e title is otherwise short.
It need not be made where t h e remainder of t h e title is long or complicated. It
would be well, however, to include a headnote in table 5 saying "See table 4
for comparable age d a t a for both sexes combined."

Sec. 3-E.

Area or Space Segment (341-352)

341. Definition.-The area or space segment is t h a t part of the
title which specifies the political division or subdivision, geographic
area, or physical plant to which the d a t a refer. Examples: FoR THE
STATE; FOR CITIES OF 500,000 ORMORE; FOR THE UNITED STATES; etc.
342. Placement.-Normally placed toward the end of the title
immediately preceding the time-reference segment. It is placed elsewhere only when the various portions of d a t a described in the title
refer to different areas, etc.
343. Standard terminology.-Use standard terminology wherever
possible. Reference to previous census reports in the given field will
usually provide a standard phrase for any given situation. A few
special features of phrasing and usage are discussed below. (For metropolitan districts, see par. 1114.)
344. The United States.-The method of referring to the United
States is virtually standardized in Census Bureau publications.
a. FOR THE UNITED STATES.-Normally signifies continental United States
only.
b. F o R CONTINENTAL UNITED STATES.-Restricted almost entirely to cases
where it is necessary to emphasize t h a t the given table is restricted to continental
United States, whereas other tables in t h e given series include figures for t h e
Territories and Possessions. In such cases, t h e distinction should be made clear
on each table. If, however, all but one of t h e tables are for continental United
States only, use FOR THE UNITED STATES in all but t h a t one. In t h e odd table
specify t h e "difference"; t h a t is, use t h e term FOR THE UNITED STATES, TERRITORIES, A N D POSSESSIONS.
Note t h a t t h e definite article ("the") is never used when referring to continental
United States.
c. FOR THE UNITED STATES, TERRITORIES, AND POSSESSIONS.-Used whenever
information is shown (within t h e same table) for both continental United States
and the outlying Territories and possessions, whether combined or separately.

44

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEc. 3-E

AREA OR SPACE SEGMENT

PAR.

345

345. The State.-Two principal uses should be noted:
a. Specifying State n a m e . - U s e the term FoR ILLINOIS, not FOR THE STATE OF
ILLINOIS. Exception: For t h e State of Washington, use FOR THE STATE OF WASHINGTON Or FOR WASHINGTON STATE wherever necessary to avoid confusion with
Washington, D. C.
b. Generalizing the r e f e r e n c e . - F o R THE STATE may be used where t h e name
of the State is carried in a page running head, or where the given release or report
is confined to d a t a for that particular State with the State name clearly specified in
the report or release title. Note, however, t h a t the expression FOR THE STATE
means that d a t a are shown for the State as a whole unless additional qualification is
provided in the title.

346. Regions and divisions.-Inclusion of the term GEOGRAPHIC is
optional in titles. Thus, one may say either FOR GEOGRAPHIC D1VISIONS or FOR DIVISIONS. When information is shown by both divisions and States, it is better to say FoR D1vis1oNs ANDSTATES omitting
the "geographic" reference.
347. Individual cities.-When presenting data for individual cities
(one table for each city), several special problems arise:
a. When specifying city n a m e . - U s e FoR THE C1TY or X, rather t h a n FOR X,
FOR THE CITY OF CHICAGO, not FOR CHICAGO. Exception: Where the term
"city" is already a part of the city name, it is enough to say FOR JERSEY CITY
rather t h a n FOR THE CITY OF JERSEY CITY.
Area designation (city, etc.) is particularly important where various types of
areas of t h e same name are presented, as Washington city and Washington State,
Chicago city and Chicago metropolitan district. When presenting city and county
data, it is important to distinguish clearly t h e following: New York City and New
York County, St. Louis city and St. Louis County, Baltimore city and Baltimore
Countv etc.
b. Specifying State name as well as city n a m e . - U n n e c e s s a r y for cities of
500,000 inhabitants or more since their names and general location tend to be
generally familiar. Optional, but desirable, for cities of 100,000 to 500,000 inhabitants. The State name should invariably be included for cities of less t h a n
100,000 inhabitants. Certain rules should be noted:
(1) S t a t e n a m e required in special cases.-Always include the State
name (abbreviated) when presenting d a t a for any city with a name which is the
same as t h a t of a major city in another State. The most common of these are:
Kansas City, Kans.
Springfield, Ill.
Charleston, S. C.
Kansas City, Mo.
Springfield, Mass.
Charleston, W. Va.
Springfield, Mo.
Portland, Maine
Columbus, Ga.
Springfield, Ohio
Columbus, Ohio
Portland, Oreg.
aS

(2) Abbreviate S t a t e n a m e . - A l w a y s abbreviate t h e S t a t e name in titles
of city tables. Use standard abbreviations. Note t h a t Idaho, Iowa, Maine, Ohio,
and Utah are always spelled out.

348. City size-groups.-When presenting data (in a single table) for
a list of cities of a given size, a common Census Bureau practice is
to omit mention of the basis of the size classification, such as INHAB-


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

45

PAR.

348

THE

sEc. 3-E

TABLE TITLE

in the title segment FOR C I T I E S OF 25,000 OR M o R E . The
inclusion of this term is essential, however, when there is a possibility
of misunderstanding.
Thus, if the title states that the universe is W O M E N OF C H I L D B E A R I N G A G E , B I R T H S , D W E L L I N G U N I T S , etc., the reader may assume
that the cities referred to are those with the given number of women
of childbearing age, births, dwelling units, etc. The danger of confusion is increased when the data are confined to aggregate figures
for all cities combined. In the latter case the data are not shown
separately for each city; hence the reader, if he is uncertain, is unable
to satisfy his doubts by noting the entries for a familiar city.
Where data are shown separately for each city, the omission of the
size-classification base from the title is a matter of judgment. In
general, the smaller the population of the area and the greater the
number of cases of the given type, the more important is specification
of the basis for the classification by size. Thus, when presenting
statistics on marriage for cities of 1,000,000 inhabitants or more it
might not be essential to specify "inhabitants." On the other hand,
when presenting data on dwelling units for urban places of 2,500
inhabitants or more, it is wise to specify "inhabitants" since many
cities have more than 2,500 dwelling units.
Examples: Appropriate terminology, including and excluding sizereferent (basis of size classification), is as follows:

1TANTS

a. Without size-referent:
Right: F o R C r T 1 E s oF 25,000 on M o R E : 1940
Right: F O R C I T I E S OF 10,000 TO 25,000: 1940
b. With size-referent: Use I N H A B I T A N T S rather t h a n P o P U L A T I O N . 3 Use
I N H A B I T A N T S OR M O R E instead of OR M O R E I N H A B I T A N T S .
Note t h a t the phrase
OR M O R E is essential where only the lower limit of the city size-group is mentioned.
Right:
Right:

25,000 I N H A B I T A N T S OR M O R E :
10,000 To 25,000 I N H A B I T A N T S :
C. S.: F O R C I T I E S OF 25,000 OR M O R E I N H A B I T A N T S :
Wrong: F O R C I T I E S 0F 25,000 I N H A B I T A N T S : 1940 ( " O R
Wrong: F O R C I T I E S OF 25,000 P O P U L A T I O N OR M O R E :
Wrong: F O R C I T I E S OF 25,000 OR M O R E P O P U L A T I O N :
Wrong: F O R C I T I E S OF 25,000 P O P U L A T I O N : 1940
F O E C I T I E S OF

F O R C I T I E S OF

1940
1940
1940
MoRE"

is essential)

1940
1940

a The reason for use of the term "inhabitants" rather than "population" bas been stated by Dr. Leon E.
Truesdell, Chief Demographer, in an office memorandum, as follows: " T h e word 'population' is
a collective noun in the singular and it may not properly be used immediately following a numeral adjective. Do not say, for example, 'in a city of 10,000 population.' Say either 'a city of 10,000 inhabitants,'
or 'a city with a population of 10,000.' This word may occasionally be used in the plural, though ordinarily
as a collective noun it carries the sense (though not the construction) of a plural. We say ' T h e population
of all these cities bas increased rapidly since 1930,' not T h e populations * * * have increased.' We
may, however, say ' T h e populations of India and of Australia are so radically different that comparisons
of this kind are not significant.""
• Contrary to style advocated in this manual, not wrong as such.

46

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

..

SEC.

3-E

AREA OR SPACE SEGMENT

PAR.

349

349. Urban and rural.-The expression used to designate urban
and rural data differs according to whether or not data are shown
(within the same table) for the State as a whole.
a. Data included for State as a whole.- U s e FOR THE STATE, URBAN AND
RURAL, if the data are shown separately for the State as a whole, for the urban
area, and for the rural area. The common separation of rural data into "Ruralnonfarm" and "Rural-farm" does not affect title description unless the table is
being shown for one of these parts only.
b. Data for urban and rural, but not for the S t a t e . - U s e FOR URBAN AND
RURAL AREAS OF THE STATE if both are shown, Or FOR THE URBAN AREA OF THE
STATE, FOR THE RURAL-NONFARM AREA OF THE STATE, etc., if only one such area
is shown.
c. Appending part of area reference to universe s e g m e n t . - F o r many types of
d a t a , t h e expressions "Urban," "Rural," "Rural-nonfarm," and "Rural-farm"
may be prefixed to the universe statement if t h e data are shown for only one of these
areas, as URBAN POPULATION, BY
FOR THE STATE, RURAL-NONFARM DWELLING UNITS, BY
FOR THE STATE, etc. T h e expression FOR THE STATE (Or
FOR ILLINOIS, etc.) is needed here to make clear t h e general area for which t h e
urban or rural part is being shown.

350. Omission and curtailment of area reference.-Under certain
conditions the area reference may be omitted entirely or may be
curtailed, as follows:
a. O m i s s i o n . - I f all tables in a given report are for t h e same area as a whole,
t h e area reference m a y be omitted from t h e individual table title, provided it is
plainly stated in the report title.
Example: If all tables in a given report are for t h e United States as a whole,
t h e expression FOR THE UNITED STATES may be omitted from t h e table titles.
This does not apply, however, to a United States summary report which recapitulates data presented in a series of individual S t a t e reports. Regardless of
whether t h e series of reports is to be bound eventually into a single volume, t h e
area reference should appear in all table titles, subject to curtailment methods
outlined below.
b. Curtailrnent.-When a report or chapter is restricted to d a t a for a single
major area, with some tables showing d a t a for t h e area as a whole a n d others for
t h a t major area classified by subareas, t h e titles of t h e latter tables need not
mention t h e major area as long as it is specified in t h e running head or in t h e
report title.
Examples:

United S t a t e s S u m m a r y . - U s e FOR DIVISIONS AND STATES, nOt FOR THE
UNITED STATES BY DIVISIONS AND STATES.
S t a t e r e p o r t . - U s e BY COUNTIES, not FOR THE STATE BY COUNTIES.
C i t y r e p o r t . - U s e BY WARDS, not FOR THE CITY BY WARDS.

351. Using "for" and "by" in area reference.-Distinguish
between the meaning of "for" and " b y " in indicating presentation of
subareas, particularly when the specification of the area being divided
has been omitted.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

47

PAR.

351

THE

sEc.

TABLE TITLE

3-E

a. "By" indicates inclusion of totality d a t a . - U s e BY if (within t h e same table)
d a t a are presented for t h e major area in t h e same or greater detail as for its
subareas.
Example: If d a t a are shown for t h e United States as a whole, followed (in
t h e same table) by comparable d a t a for each division and State, use BY DIVISIONS AND STATES, rather t h a n FOR DIVISIONS AND STATES.
Similarly, in a table presenting d a t a for t h e State as a whole, followed by
comparable data for each county, use BY COUNTIES, nOt FOR COUNTIES.
b. "For" indicates presence of subarea data without totality f i g u r e s . - U s e
Fon in t h e following situations:
(1) Where data are shown (in a given table) for each subarea (such as divisions, States, counties, etc.) but not for t h e major area (such as t h e United
States, or t h e State). T h e titles should then read FOR DIVISIONS AND STATES;
FOR COUNTIES; etc.
(2) Where only a grand total line (or column) is shown for the major area
in combination with a detailed distribution for the subareas.
(3) Where cities are listed, since the city d a t a will never add to a State total
line or block. Use FOR even though a total is shown for all city d a t a combined.
(See below for area reference for a table with both city and county data.)
c. City and county data in same t a b l e . - I n this case use FOR COUNTIES or BY
CouNTIES depending upon the presence of State total d a t a in t h e same detail as
shown for the counties. Use FOR CITIES, invariably. Combine as follows, being
careful to see to it that the city-size reference is plainly applicable to the cities
only, not to tbe counties.
Right: BY COUNTIES, AND FOR CITIES OF 25,000 OR MoRE: 1940
R i g h t : FOR COUNTIES, AND FOR CITIES OF 25,000 OR MORE: 1940

352. Use of terms "Selected" or "Specified."-When showing a
major area classified by subarea, use t h e term S E L E C T E D or S P E C I F I E D
when t h e listing of subareas for which d a t a are shown separately does
not include all of the subareas of the type described in t h e title.
These terms should never be employed where t h e stub (or box) listing
includes all subareas of the type described in the title.
a. " S e l e c t e d . " - B y definition, this term makes clear t h a t some specific criterion has been applied in determining which areas to include and to exclude. If
SELECTED is used in the title, the basis of selection should be stated specifically
in t h e title or in the headnote, or t h e headnote should refer the reader to the
appropriate portion of the text for the explanation.
Examples: FOR SELECTED COUNTIES
W I T H D A T A FOR N O N W H I T E FOR SELECTED CITIES
FOR SELECTED C I T I E S OF 10,000 OR M O R E
b. " S p e c i f i e d . " - U s e d ordinarily when t h e criterion is too complex to explain;
where enumeration or processing difficulties are the determining factor; or where
the d a t a would not be significant for the remaining areas, but the circumstances
do not warrant detailed explanation. SPECIFIED does not demand explanation
in the title, headnote, or text, although a generalized text statement may be made
on the subject if desired.
Examples: FOR SPECIFIED COUNTIES
FOR S P E C I F I E D C I T I E S OF

48

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

10,000

OR M O R E

SEC.

3-F

TIME-REFERENCE SEGMENT

PAR.

361

Sec. 3-F. Time-Reference Segment (361-368)

361. Definition.-That part of the title which signifies the point in
time to which the figures refer. May be a single date, a time period,
or a series of dates or time periods. Examples: 1940; 1935 ro 1940;
1940

AND

1930; 1910, 1920,

AND

1940;

W E E K OF M A R C H

24

To

30,

1940; etc.

362. Placement.-The time-reference segment is placed after that
portion of the title to which it applies.
a. Same time-reference for all portions of t i t l e . - P l a c e at end of title, preceded
by a colon.
Right: AGE OF THE POPULATION, BY R A C E A N D S E X , FOR THE UNITED STATES:
1940

AND

1930

Wrong: AGE OF THE POPULATION, BY RACE AND SEX, 1940 AND 1930: FOR THE
UNITED S T A T E S
b. Different time-reference for different portions of t i t l e . - P l a c e each reference
after portion to which it applies. Usually each is set off by commas, although it
may be introduced by "in," "for," or "during." Even though one of the timereferences may thereby appear at the end of the title (see last example, below),
the colon must not be used unless the reference applies to the entire title.
Right: AGE BY RACE AND SEX, 1940, AND AGE BY RACE, 1930, FOR THE
U N I T E D STATES
Right: A G E BY R A C E A N D S E X F O R THE UNITED S T A T E S , 1940 A N D 1930,
1940
1940 A N D 1930;
S T A T E S , 1940

A N D BY R A C E FOR D I V I S I O N S A N D S T A T E S ,

Right:

A G E BY R A C E A N D S E X , FOR
A N D BY

Confusing: AGE

BY

THE

U N I T E D STATES,

R A C E , FOR D I V I S I O N S A N D

RACE AND SEX FOR THE UNITED STATES, 1940 AND

1930,

A N D BY R A C E F O R D I V I S I O N S A N D S T A T E S :

1940

363. Punctuation.-As illustrated above, an interspersed timereference is usually set off by commas. A terminal time-reference is
set off either by a colon or a comma.
a. Colon.-Precedes terminal time-reference only when t h e reference applies to
the title as a whole; otherwise use a comma. This distinction is important.
Incorrect usage leads to confusion. (See example under par. 362b, above.)
b. C o m m a . - U s e commas to set off interspersed time-references; t h a t is, to set
off each time-reference which applies only to that portion of the title immediately
preceding the given reference, regardless of whether such reference be within or
at the end of the title.
c. Semicolon.-Use a semicolon after interspersed time-references whenever
clarity will thereby be increased.

364. Exact dates not essential in table titles.-Although precise
time specification is imperative in any report, particularly for statistics
subject to seasonal or other time fluctuation, time description should
be held to an absolute minimum in table titles. The burden of detailed statement of the time factor should be borne by the accompanying text, if any, or by the headnote or some other portion of the table.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

49

PAR.

364

THE

TABLE TITLE

sEc.

3-F

In general, dates or time periods which depart from the normal
expectation of the user should be spelled out in full in the title. Those
which reflect a standard or expected situation should be curtailed
sharply."
In particular, in any series of tables with identical time-reference,
only the most imperative part of the reference should be included in
the table titles unless it is important to distinguish that series from
some other series (in the_same_report) which has a different timereference.
a. Regular census d a t e s . - S u c h dates as those for t h e Censuses of Population,
Agriculture, etc., may be shown merely as 1940, 1930, 1920, 1910, rather t h a n
APRIL 1, 1940, APRIL 1, 1930, JANUARY 1, 1920, APRIL 15, 1910. These dates
are so well-known t h a t there is rarely a n y point in specifying in f u l l in table titles.
In tables showing comparative data for several censuses, however, the exact dates
should be shown in the boxhead or s t u b if the difference in mouth of enumeration
needs to be taken into account by the user. In any case, the exact dates should
be stated clearly in the text of the report.
b. Regular census periods.-For statistics which cover a regular survey period
(rather t h a n being as of a given date) such as information gathered in the Biennial Census of Manufactures or the Census of Business, the time-reference is
normally stated merely as 1939, or 1937, rather t h a n in the longer and more precise forms D U R I N G T H E YEAR 1939, Or F O R T H E YEAR 1937,5
c. Other d a t e s . - S p e c i f y exact d a t e or period in table title only in those cases
where misinterpretation of the statistics would arise if it were not stated exactly;
where it is impracticable to s t a t e it elsewhere in the table; or where it is insufficient
to confine the exact time-reference to the accompanying text.

365. General problem of listing two or more dates in table titles.Basically, the order of listing dates in the title should reflect the
order in which statistics for those dates appear in the table (from
the top reading down and from the left-hand side reading across).
In the arrangement of the figures (see pars. 1122-1124) two methods
are in common use in the statistical world, as follows:

-

a. Direct method.-This method places t h e earliest statistics first and the most
recent statistics,last. Thereby, t h e historical sequence is'stressed and an identical
arrangement of dates in t h e table title is made possible.
b. Indirect method.-This method reverses t h e order of presentation_by placing
t h e most recent figures first and the earliest figures last. Thereby, the most
recent information is stressed. However, since such an expression as 1940 TO
1870 is unsatisfactory, the indirect listing of dates in the table title is generally
restricted to cases where only two or, under certain conditions, three dates are
to be cited. Where four or more dates are involved, t h e dates are listed forward
in time in t h e title (resorting to t h e direct method) without disturbing t h e backward arrangement of figures in t h e table.
In the volumes of the Agriculture Census, the term CENsus or 19- is commonly employed in table
titles where a single table presents both inventory data as of a given date and statistics for the preceding
crop year, In such an instance, specific identification is provided within the table. This practice represents a solution of a problem similar to that cited in par. 364b.

50

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

3-F'

sEC.

TIME-REFERENCE SEGMENT

PAR.

365

T h e general rule for title description of statistics arranged backward in time
may be stated as follows:
General rule: Where statistics are arranged backward in time, t h e same arrangement should be used in the table title as long as brevity is maintained:
c. Bureau of Census practice.-The Bureau of the Census uses the indirect
method, t h a t is, it normally places t h e most recent information first in t h e table and
arranges t h e dates in the title accordingly.6 T h e primary reason for publication
of Census statistics is to present t h e information from t h e most recent census or
survey; statistics from other censuses are characteristically inserted for the
convenience of the reader.

366. Order of listing two or more dates.-In general, dates should
be listed in the table title in the same order that they appear in the
boxhead or stub. If the earliest year appears at the top in the stub (or
at the left in the boxhead), the dates should be listed forward in time
in the title, as 1930 A N D 1940.
If the latest year is placed at the top in the stub (or at the left in the
boxhead), the title listing should be backward in time if not more than
two dates are involved, as 1940 AND 1930; but forward in time if more
than two7dates, as 1920 TO 1940 (decennial statistics), or 1939 To 1941
(annual statistics). The following discussion assumes normal census
arrangement in the tables; that is, that the statistics are shown with
the latest year first.
a. Two dates or time p e r i o d s . - L i s t them backward in time; t h a t is, show the
most recent date first, followed by the earlier date, and join them with A N D .
Right: 1940 A N D 1930
C. S.: 8 1930 A N D 1940
Wrong; 1930 To 1940 (implies presence of figures for intervening dates)
Right: 1940 A N D 1920
C. S.: s 1920 AN 1940
Wrong: 1920 TO 1940 (implies 1930 shown also)
Wrong: 1920-1940 (implies 1930 shown also)
b. Three dates in an incomplete series; or not in standard series.-Where
three dates are shown which represent three independent dates (or periods) not
comprising a standard time-series, or where they represent three out of four points
in a standard time-series, list each date separately with the most recent placed first.
(1) Incomplete standard s e r i e s . - I f decennial census statistics on a given
subject are available (for example) for only three out of t h e four censuses from
1910 to 1940, the listing should be as follows:
Right: 1940, 1920, A N D 1910 (if 1930 not shown)
Right: 1940, 1930, A N D 1910 (if 1920 not shown)
Wrong: 1910 TO 1940 (implies all census years appear)
Wrong: 1910-1940 (implies all census years appear)
For comment on usage in the Statistical Abstract of the United States, see footnote 7, par. 1122b.
exception, see par. 366b, below.
• Contrary to Bureau of Census style, not wrong as such. Actually, in volumes with historical emphasis,
this style (direct method) frequently is desirable.
7 For

77655449

5


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

51

PAR.

366

THE

TABLE TITLE

SEC.

3-F

(2) Independent d a t e s . - F r e q u e n t l y when statistics are presented for a
special census or survey, comparable information is available for two or more
past dates or periods which m a y or m a y not include regular censuses. T h e
three dates should be listed as follows, with inclusion of exact dates depending
on analytical needs, opportunity to express them elsewhere in t h e table, and
presence of accompanying text.
Right: SEPT. 1, 1944, APR1L 1, 1940, and FEB. 18, 1932
Right: 1944, 1940, AND 1932
C. S.: ° 1932, 1940, AN 1944
Wrong: 1932 TO 1940 (time coverage open to doubt)
c. Three or more dates in continuous s e r i e s . - W h e r e three or more dates or
time periods comprise all, or a complete segment, of a continuous time-series (no
- s t a n d a r d d a t e or period missing within t h e group), list them forward in time; t h a t
is, list first t h e earliest period and place t h e most recent date or time period last,
separating them by TO. This method does not emphasize t h e most recent or
current d a t e but is required by t h e need for brevity.
Right: 1910 TO 1940 (if 1910, 1920, 1930, and 1940 are all shown)
C. S . : 4 1910-1940 (use r o , not hyphen)
Wordy: 1910, 1920, 1930, AND 1940
Wordy: 1940, 1930, 1920, AND 1910
d. Combinations involving future t i m e . - I n all combinations which include a
reference to future time, list in chronological order beginning with t h e past and
ending with t h e future. T h e question of a separate listing for each d a t e or time
period shown depends upon whether a continuous or broken series is involved.
Right: 1910 To 1980 (if figures are presented for each standard period, such
as every 2-, 5-, or 10-year period)
Right: 1910 To 1940, 1960, AND 1980 (if figures not shown for 1950 a n d 1970)
Right: 1900, 1920 TO 1940, 1960, AN 1980 (if figures not shown for 1910,
1950, and 1970)
Right: FOR SPECIFIED PERIODS, 1910 To 1980 (avoids detailed listing to
indicate gaps; headnote explanation not r e q u i r e d - s e e par. 352b)
Right: FOR SELECTED PERIODS, 1910 To 1980 (requires headnote explanation
of basis of selection-see par. 352a)

367. Use of TO, hyphen (-), AND, and DURING.-These connectives should be used carefully. The following may prove helpful. In
general, the examples relate to decennial census dates. However, the
same principles apply to other dates or time periods.
a. T O . - F o u r points govern t h e use and meaning of TO in Census practice.
(1) Signifies t h a t all specified and intermediate standard dates are
s h o w n . T h e use of TO between dates or time periods signifies inclusion of
figures for both dates mentioned and also for all standard intermediate dates,
unless preceded by t h e term FOR SPECIFIED DATES Or FOR SPECIFIED PERIODS.
Example: T h e expression 1910 TO1940 means t h a t figures are shown separately
for 1910, 1920, 1930, and 1940, and not merely for 1910 and 1940. (For exceptions, see par. 368c, below.)
(2) M a y represent " c h a n g e " involving t i m e . - T a b l e s restricted to
"increase," "decrease," or other types of "change" in terms of time require
Contrary to style advocated in this manual, not wrong as such.
Contrary to Bureau of Census style, not wrong as such. This is correct when using the direct method.

52

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEc.

3-F

TIME-REFERENCE SEGMENT

PAR.

367

t h e use of TO (as 1930 TO 1940), or such expressions as BETWEEN 1930 AND
1940. If three points in time are involved, this may be stated either simply
and briefly, as (example) 1920 TO 1940, or more explicitly as 1920 To 1930,
A N D 1930 TO 1940.
(3) Three or more consecutive dates or t i m e periods m u s t be inv o l v e d . - U s e ro in only those cases where three or more dates are presented in
a continuous series. If only two dates are involved, use AND, except as provided in the paragraph immediately below.
(4) Requires listing dates forward in t i m e . - W h e r e ro is used, t h e
listing must necessarily be forward in time; t h a t is, t h e earliest date is listed
first, as 1920 To 1940, never 1940 TO 1920.
b. Hyphen ( - ) . - I t is suggested t h a t t h e hyphen not be used except where
extreme demand for space necessitates. However, where used by the Census
Bureau, the hyphen @ (as in 1870-1900) is subject to the same rules as r o ; i t should
not be substituted for A N D or DURING.
c. A N D . - T h e use of AND conforms to t h e ordinary rules of rhetoric.
(1) To join a n y combination of t w o d a t e s . - I f t h e listing is comprised
solely of t h a t combination, the current or most recent date is normally placed
first in Census Bureau practice.
Right: 1940 AND 1980 (if statistics are shown only for 1940 and 1980)
C. S.:11 1980 AND 1940 (wrong order)
Wrong: 1940 TO 1980 (implies figures are shown for 1950, 1960, and 1970,
also; t h a t is, for more t h a n the two dates)
Right: 1940 AND 1930
C. S.;11 1930 AND 1940 (wrong order)
Wrong: 1930 TO 1940 (use AND for two dates)
(2) To join t h e last t w o m e m b e r s of a s e r i e s . - T h i s represents t h e
normal practice of using AND in a series. Note that both AND and TO may be
used in a series of dates where the use of TO covers three dates or periods.
Right: 1880, 1900 TO 1920, AND 1940 (if 1890 and 1930 are not shown)
Right: 1940, 1930, AND 1910 (if 1920 is not shown)
d. DURING.-"During" implies an action t h a t was taking place continuously
during the specified period; or which might have occurred at a n y time, or at varying combinations of times, during that period. A void this term where possible.
I t s primary use is as a part of the universe-classification segment where it limits
t h e group to which reference is made. It is rarely used as a part of the timereference segment as such; but when used, it always reads in;12 that is, no comma
or colon separates it from the rest of the time specification.
Right:

AGE OF EMPLOYED WORKERS WHO WORKED LESS THAN 12 MONTHS
1N 1939, BY RACE A N D SEX, FOR THE UNITED STATES: 1940
Right: AGE AND SEX OF WORKERS EMPLOYED DURING WEEK OF MARCH 24
To 30, 1940
Wrong: AGE AND SEX OF EMPLOYED WORKERS: DURING WEEK OF MARCH
24 TO 30, 1940

10 Technically, in typeset composition, the "I-en dash" (-) is substituted for the hyphen (-) in tabular
work. However, since the distinction involves a printing technicality, and the difference between these
characters is scarcely distinguishable to the nonprinter's eye, the term hyphen is used throughout this
manual.
11 Contrary to Bureau of Census style. "1940 and 1980" and "1930 and 1940" would be correct with the
direct method.
12 For definition of "read in," see par. 927.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

53

PAR.

368

THE

TABLE TITLE

sEC.

3-F

368. Omission and curtailment of time-reference.-In certain cases
the time-reference may be omitted completely from the table title,
or curtailed sharply therein, even though this may mean a minor
sacrifice in the cataloging function of the title.
a. Complete omission from title.-The time reference may be omitted if t h e
statistics shown in all tables in the given report are for the same date or time
period, and the date or time period is specified in the running head of the page.
This type of omission is risky if statistics for a n y other period appear in a n y table.
If the time-reference is brief, it is better to include it in all tables.
b. Omission of all reference to incidental statistics.-Omit all reference to
incidental statistics within a given table which have a time-reference other t h a n
t h a t for the main body of the table, particularly when the classification is not
as detailed. Thus, if all but a few lines of a full-page table are for 1940 and the
few lines (for 1930) have been added merely as a convenience to the user, no
reference to the 1930 statistics need be made in the table title unless the items are
of paramount interest or are not a logical part of the table in terms of subject
matter.
c. Disregarding missing members in a time-series.-Use of such expressions as
1850 TO 1940 is undesirable unless statistics are shown for all intervening censuses,
but occasionally may be justified under the following conditions:
(1) One missing m e m b e r in a long-term standard s e r i e s . - I f only one
member is missing in a standard long-term time-series, and if the remainder of
t h e title is necessarily long, the time-reference may be shortened by ignoring
t h e gap, provided t h a t the headnote specifies the missing member. The example assumes statistics for decennial census periods.
Right: 1850 TO 1880, AND 1900 TO 1940
Permissible: 1850 TO 1940 (with headnote reading "Statistics not shown
for 1890")
Inappropriate: F o R SPECIFIED CENSUsEs, 1850 To 1940 (use of SPECIFIED
is inappropriate when only one member is missing)
(2) Two or more m e m b e r s missing in a long-term standard series.If two or more nonconsecutive members are missing, one of two procedures
is possible, listed below in order of preference. The example assumes t h a t
statistics are shown for all decennial censuses from 1850 to 1940 except 1890
and 1910.
Right: 1850 TO 1880, 1900, AND 1920 TO 1940
Permissible: FOR SPECIFIED CENSUSES, 1850 To 1940
Avoid: 1850 To 1940 (even with headnote explanation)
If all of the missing members are consecutive, specify the censuses shown.
T h e following example assumes that all members from 1850 to 1940 appear
except 1890 and 1900.
Right: 1850 TO 1880, AND 1910 TO 1940
Inappropriate: FOR SPECIFIED CENSUSES, 1850 To 1940
Avoid: 1850 ro 1940 (even with headnote explanation)

d. Omitting mention of minor variations in dates in a time-series.-In a timeseries where all but a few points are for the same day in t h e month (or same
month in a year) and the remainder vary as to d a y and month, do not mention
t h e variation in the title. If necessary, make it clear in a headnote or elsewhere
in the table.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

SEC.

3-F

TIME-REFERENCH SEGMENT

PAR.

368

e. Including only major aspect of t i m e - r e f e r e n c e . - I n general, give only t h e
most imperative part of a time-reference in the title. If the date is an exact
day, it usually is sufficient to mention only the month in the title, or it may be
reduced to t h e year if it is a standard census date. Similarly, use only the year
if t h e time-reference is to a month, unless the month specification in the table
title is absolutely essential.® Frequently it is sufficient to mention the month
in the accompanying text a n d . refer only to the year in the table title. (See
par. 364.)

Sec. 3-C. Title Description of Percent Data (371-374)

371. General.-Correct identification of the base is the paramount
consideration in describing percentages. A second requirement is
correct description of the class-groups into which each universe, or
subuniverse, is distributed. Since these objectives must be achieved
by fine distinctions in phrasing and word-order, flexibility in title
treatment is reduced drastically in comparison with titles for frequency
tables.
Percentage description usually must be independent of description
of frequencies presented in the same table; that is, although certain
words or phrases may be identical for both descriptions, they usually
must be repeated separately for each. The need for independent
description for each type of data naturally adds to title length and
complexity.
372. Percentages and frequencies appearing in same table.-Many
tables contain both percentage and,frequency data. In general,
title treatment depends on the primary purpose of the table as a
whole and the conditions or circumstances of inclusion of the two
types of data.
a. Frequency table including percentages.-Here the primary purpose of the
table is to show frequencies. The percentages are of secondary importance and
are inserted for reader-convenience or to clarify, or add significance to, the frequency information.
Rule: Confine title to description of frequencies. Omit title-reference for the
percentage data unless they(1) Are of outstanding significance in themselves, or
(2) Are not a natural byproduct of t h e frequencies shown; t h a t is, they
comprise percentages which would more logically be placed elsewhere and which
have been included in the given table only because of available space.
b. Percentage table including f r e q u e n c i e s . - H e r e t h e primary reason for table
existence lies in the percentage presentation. T h e frequencies are of secondary
significance. Thus, it may be desirable to include in a percentage table the frequency rims used as the bases of the percentages.
Rule: Confine title to description of percentages. Omit title-reference for the
frequency data unless they13 The interpretation of "imperative" and "essential" depends upon the nature of the statistics. Thus,
for highly seasonal phenomena, a failure to specify the month or season might be a serious error.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

55

PAR.

372

THE

TABLE TITLE

sEc. 3-G

(1) Are of outstanding significance in themselves, or
(2) Are not t h e bases of the percentages but comprise "supplementary"
frequency information which appear in the given table because of difficulty
of inclusion in some other table where they logically belong, or because their
nature is such that they cannot be presented in a separate table.
c. Doubtful c a s e s . - W h e r e t h e amount of frequency and percentage d a t a is
about equal, or where the primary purpose of t h e table (in terms of presenting
frequency or percentage data) is in doubt, the general rule is to assume t h a t it is
a "frequency table including percentages" and title it according to the rule in
paragraph 372a, above.

373. Problem of distinctive phrasing of percentage titles.-When a
number of percentage tables which have been titled at various points
in time are compared with each other, it may be found that the
descriptive terminology has not been used consistently, even where
all the tables were titled by the same person. Thus, identical phrasing
(in respect to essentials) may have been used to describe basically
different distribution patterns, or different phrasing may have been
used to describe identical patterns, or both. Obviously, then, some
of the tables have not been titled properly, since the reader cannot
ascribe the same meaning at all times to the same phrasing.
a. Use of a predetermined method.-Theoretically, t h e solution to t h e above
problem would be to title all tables in accordance with some predetermined
scheme broad enough in scope to provide distinctive phrasings for at least t h e
most common situations.
In this connection, it must be remembered t h a t t h e title-writer and t h e user
approach t h e title from opposite directions. T h e title-writer works from table
to title; t h a t is, knowing t h e content of t h e table, he tries to describe it. T h e
user, however, works from title to table; t h a t is, he tries to deduce t h e content
from t h e descriptive title. It follows t h a t t h e title-writer should avoid terms
which he a n d t h e user will not interpret in t h e same way.
T h e first prerequisite to a common understanding is the consistent use of
descriptive phrases. Use of a standard scheme would promote consistency not
only among titles written by the same person, but also among titles written by
different persons. However, a n y scheme which attempts to differentiate between
all major possibilities tends to become cumbersome where applied. Also, the
distinctions in phrasing may become too fine-drawn to be useful to most readers.
(See par. 336.)
An experimental method is illustrated in fig. 11, pp. 57-58. This scheme states
the problem; it does not solve it. T h e phrasings provided there are better
suited for the description of simple t h a n of complicated tables.

374. Appending percentage description to frequency titles.-Where
it becomes necessary to call specific attention to the presence of percentages in a frequency table, the reference usually is inserted preceding the area segment (assuming area specification is the same as
for the frequencies) and is introduced by W 1 r H .
Right: AGE

OF

THE

56

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

P O P U L A T I O N BY C O L O R AND S E X , W I T H P E R C E N T D I S TRIBUTION BY A G E , FOR THE STATE: 1940

Figure 1 1 . - E X P E R I M E N T IN FORMAL TERMINOLOGY FOR PERCENT TABLES
T h i s figure illustrates t h e problems involved in t h e development
of a n y s t a n d a r d scheme for phrasing of titles of percentage tables.
In each case, t h e first title stresses t h e class being distributed (the
base) ; t h e second title stresses t h e characteristic by which classified.
T h e s e examples comprise an experiment in a search for(1) A s y s t e m a t i c presentation which focuses attention on basic
problems of percentage description;
(2) A testing g r o u n d for a n y system of terminology it m a y
seem desirable to develop; a n d
(3) A guide for optional use where circumstances seem to
warrant.

T h e skeletal titles shown represent t h e basic terminology for t h e
percentages described. T h e t h r e e d o t s at t h e beginning of each title
(except tables 5 a n d 9) represent t h e words PERCENTAGE DISTRIBUTION; those at t h e e n d represent t h e area a n d t i m e specification.
T h r o u g h o u t , X represents t h e universe t o t a l (as t o t a l employed
workers) ; A represents sex; B, color; a n d C, age.
Example: Assuming d a t a a r e for employed workers in t h e
United S t a t e s , 1940, t h e title for t a b l e 1 would read Percentage
Distribution OF Employed Workers BY Co L O R , for the United
States: 1940 (or) Percentage Distribution BY COLOR FOR Employed

(2) X classified by characteristic A,
a n d each resultant group distributed
according to characteristic B:
(1)

Universe t o t a l ( X ) distributed
according to characteristic B:

TABLE 1.
OF (THE UNIVERSE)
BY COLOR
(or) . . . BY COLOR
FOR (THE UNIVERSE)
Color
All classes_-White.
Nonwhite_-(See examples 4 and 5 below.)


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Percent
100.0

89. 8
10. 2

TABLE 2.
OF SEX GROUPS OF
(THE UNIVERSE) BY COLOR
(or) . . . BY COLOR FOR SEx
GROUPS OF (THE UNIVERSE)

(3) X cross-classified by A a n d C,
a n d each resultant group distributed
according to B :
TABLE 3.
OF AGE-SEX GROUPS
OF (THE UNIVERSE) BY COLOR
(or) . . . BY COLOR FOR AGE-SEX
GROUPS OF ( T H E UNIVERSE)
[Compare with table 7]

Color and sex

Percent
Color and sex

Total_-

100. 0

White_Nonwhite ___

89. 8
10. 2

Male __
White_Nonwhite_-Female __
White_Nonwhite ___

i::::,

Workers in the United States: 1940.

100.0
90. 0
10.0
100. 0
89. 6
10. 4

AU
ages

Under 45 and
45 years over

Total_WhiteNonwhite--

100.0
89. 8
10. 2

100. 0
88. 8
11. 2

100. 0
92. 4
7. 6

Male_-White_-Nonwhite.--

100. 0
90. 0
10. 0

100.0
89.8
10.8

100. 0

100. 0
89. 6
10. 4

100.0
88. 5
11. 5

100. 0
92. 6
7.4

Female __
White_
Nonwhite_--

92. 2

7.8

0
0

( 4 ) X distributed according to groups
formed by cross-classification by A a n d
B (single-column arrangement):
TABLE
COLOR
COLOR

OF ( T H E U N I V E R S E ) BY
AND
SEX
(or) . . . BY
AND S E X FOR (THE UNIVERSE)

( 6 ) X distributed according to A and B, (7) X classified separately by A and C and
taken separately:
each group distributed by B:
T A B L E 6.
OF ( T H E U N I V E R S E ) BY T A B L E
OF S E X A N D A G E G R O U P S
C O L O R A N D BY S E X
{or) . . . BY
(or)
OF ( U N I V E R S E ) BY C O L O R .
C O L O R A N D BY S E X F O R ( T H E U N I V E R S E )

BY
OF

C O L O R FOR
(UNIVERSE)

SEX

AND

AGE

GROUPS

[Compare description w i t h t h a t for table 6]
(Compare description with t h a t for table 4]
Percent

Color and sex
T o t a L __

89.8
10. 2
50. 2

Male
White_
Nonwhite __

45. 1

5. 0

Total

100.0

Total_White_
Nonwhite_-

89.8
10. 2

Male_
Female_-

50. 2 White_-49. 8 Nonwhite __

Total_-

45
Fe- Under and
M a l o male
45 over

100. 0 100. 0 100. 0 100. 0
89. 9
10. 2

90. 0
10. 0

89. 6
10. 4

88.8
11. 2

100. 0
92. 4
7. 6

49. 8

Female
White_
Nonwhite.-

44. 6
5. 2

• Omission of lines for these categories would not affect
t i t l e , since these are subtotals or prime cells.

(5) Variant of table 4 with another title
approach:
TABLE

Color

100. 0

White
Nonwhite

Age (years)

Sex

Percent

Color a n d sex

5 . - P E R C E N T OF
(UNIVERSE)
EACH COLOR-SEX GROUP

(8) 2-way distribution in same table:
OF S E X BY A G E ,
T A B L E 8.
(or)
OF A G E BY S E X , F O R
BY A G E F O R S E X , A N D BY S E X
A G E , FOR

IN

Sex by age

AND

(9) Cell-to-cell relationships between data
in two frequency tables, expressed as
percentages (not. a complete distribution):

FOR
TABLE 9 . - P E R C E N T
AGE-SEX GROUP

Total_-White_
Nonwhite.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Total

Male

Fe Total Male Fe
Total Male male
male

Female

100. 0

50. 2

49. 8

89. 8
10. 2

45. 1
5. 0

44. 6

2

Total_Under 45 years_45 and o v e r -

100. 0 100. 0 100. 0 100.0
73.3
26.7

EACH

Age by sex

Age
Color

W H I T E IN
OF ( T H E

73.1 73.6 100.0
26. 9 26. 4 100. 0

50. 2 49.8
50. 0 50.0
50. 6 49. 4

Sex

Total ___
Male
Female...

.All
ages

Under
45
years

45
and
over

89. 8

65. 1

24. 6

45.1
44. 6

32. 7
32. 5

12. 5
12. 2

Chapter 4

THE TABLE NUMBER (401-435)
Sec. 4-A. General (401-407)
401. Definition and purpose.-The table number is an indicator of
relative position of the table within a series. As an identifying
device, it provides a convenient means of table indexing and reference.
The use of prefix and suffix symbols makes it possible to indicate
relationship between and within tables and series of tables.
In Bureau of the Census practice, the table number immediately
precedes and is placed on the same line as the opening of the title.
It is separated from the title proper by a period and a dash. The
number appears in boldface type where boldface is available.
402. Symbols u s e d . - F o r practical purposes the types of symbols
used for numbering are restricted to four, discussed here in order of
frequency of use:
a. Arabic n u m e r a l s . - M o s t desirable because most familiar to the user.
b. Capital I e t t e r s . - M o s t common use is for text tables when Arabic numerals
have been used in the tabular section of the report. Capital letters should be
avoided for a n y series with more t h a n 26 tables; the doubling-up (such as TABLE
AA or Aa for the twenty-seventh (or the second) table tends to be confusing.
For use of capital letters as prefixes, see p a r a g r a p h 403 a n d s e c t i o n 4 - C .
c. Roman n u m e r a l s . - A l t h o u g h frequently used for text tables, Roman
numerals have two serious disadvantages:
(1) Hard to r e a d . - F e w users can read Roman numerals easily. In general,
Roman numerals should be avoided in numbering schemes. In a n y case, they
should not be used in a n y series of tables numbering more t h a n 39, since the
meaning of the L in XL is rarely understood.
(2) Verbal c o n f u s i o n . - R o m a n and Arabic numerals of t h e same value are
readily distinguished in writing but are identical in speech. T h e use of Roman
and Arabic numerals to distinguish two series of tables m a y prove troublesome. No user can refer to them verbally (on the telephone, or when giving
instructions, or dictating) without either specifying Arabic or Roman, or risking
confusion.
d. Small letters (lower case).-Undesirable when standing alone, particularly
where, as in Bureau of Census reports, table titles are printed or typed in caps,
or in caps and small caps.! In particular, the use of small letters should be confined to a series of not more t h a n 26 tables. Small letters are most useful as suffixes
to table numbers. (See sec. 4 - D . )
T H I S IS IN CAP8. THIS IS IN CAPS AND SMALL CAPS.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

59

PAR.

403

THE

TABLE N U M B E R

sEc.

4-A

403. Symbol combinations.-In order to distinguish subseries of
tables, or to indicate direct table relationships, certain symbols may
be used in combination. In general, those combinations are best
which tend to be most familiar to the user. In the following list a
means any small letter, 2 means any Arabic numeral, etc.
Preferred

Acceptable

Not acceptable

2a
A-2
II-A

2-A
A-II
Ila

2 - 2 or 2-II
A-A or A-a
I I - I I or II-2
a-2, a-A, or a - a

404. Omit number for single tables.-When the entire tabular
presentation consists of a single table, the table number may be omitted since it rarely performs a useful purpose. Note, however, that
this does not apply to an unrelated table among a group of related
tables. It refers to such cases as a release containing only one table,
or a single table typed for inclusion in correspondence.
405. No duplication.-Within a given report, no two tables should
bear the same number unless a standard table scheme is involved.
(See sec. 4-B.) However, table numbers differentiated by prefixes
or suffixes do not constitute duplications unless the entire combination
is identical.
406. Normal numbering scheme.-Unless there are specific reasons
to the contrary, tables should be numbered in a single series of continuous Arabic numerals beginning with 1.
407. Tables in text and in tabular section.-In a report which
presents analytical or summary tables within the text with the bulk
of t h e statistical presentation segregated in a tabular section of detailed tables, two numbering schemes are available. (See also par.
423.)
a. Numbering in same s e r i e s . - A s s i g n number 1 to t h e first text table and continue the series without a break through the tabular section.
(1) Most suitable when text tables and detailed tables, taken together, constitute a logical development.
(2) Least suitable when text tables constitute a summary or analytical series
covering t h e entire group of detailed tables.
(3) M a y cause operating difficulties. In general, the text material tends to
be subject to change after t h e detailed tables in t h e tabular section have been
frozen. A change in the total number of text tables (by adding, omitting, combining, or dividing a table) will throw off t h e numbering of all the detailed
tables.
b. Numbering in separate s e r i e s . - N u m b e r text tables in one series; s t a r t a
new series for the tabular section.
(1) Most suitable where t h e two series are not continuous in development, as
when the text tables comprise figures selected from the tabular section either
for summary purposes or to illustrate analytical points discussed in t h e text.

60

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

•

src. 4-A

GENERAL

PAR.

407

(2) Least suitable when text tables and detailed tables, taken together,
develop t h e materials seriatum.
(3) Most convenient system for operating purposes since revisions in text do
not affect the tabular section.
(4) Requires two different types of symbols to avoid duplication of table
numbers. The preferred symbol (Arabic numerals) should be used for the detailed tables. Either capital letters or Roman numerals m a y be used for the
text series, with capital letters preferred. However, see also section 4-C,
paragraph 423.

Sec. 4-B. Numbering Standard-Series Tables (411-414)

411. Standard series defined.-A method of tabular organization in
which a series of table forms is established for a given area or analytical
group, and the same series is repeated for each comparable area or
group.
Example: Any Bureau of the Census volume with State chapters (or sections)
in which tables identical in form are presented for each State.

412. Purpose and use of standard series.-Facilitates analysis,
reference, and production.
a. Insures parallelism of presentation.-Within obvious limits, a characteristic
or classification presented for a n y one group or area is automatically presented
for all comparable groups or areas. In this way, t h e more important features are
assured a uniform coverage. Moreover, this method does not prevent insertion of
additional tables for features peculiar to a given area. (See sec. 4 - D , pars.
433-435.)
b. Facilitates r e f e r e n c e . - T h e standardization as to content tells t h e reader
referring to a given area the minimum he may expect to find for any similar area
or group for which t h e same problem exists. The identity of table number also
provides him with the table number for all similar tables for*all similar areas or
groups.
c. Simplifies production.-Instructions common to all similar areas or groups
can be prepared at one time. Problems common to all can be handled by general
rulings. The attention can then be focused on individual problems peculiar to
specific areas or groups. Worksheets can be reproduced in quantity with common captions already entered instead of being written by hand on each sheet.

413. Numbering scheme.-The standard numbering scheme is an
important feature of the standard series. Its existence emphasizes
the standardization and makes it apparent to the user. The numbering method should conform to the following requirement, stated here
from two different points of view:
a. Tables with identical numbers should have comparable content and coverage.
b. Tables with comparable content and coverage should have identical numbers.

414. General consequences of standard table numbering. Certain
general consequences follow from the use of a standard series numbering scheme, and certain obligations are automatically incurred:


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

61

Pax.

414

THE

TABLE N U M B E R

sEc.

4-B

a. Automatically arranges report in parts or sections, irrespective of formal
designation. T h e formal organization of the report should be adapted to this
arrangement.
b. Requires clear designation, either in the running head or in t h e table title,
of t h e particular universe for which each table is presented. T h a t is, since there
may be a number of tables numbered 1, the distinctive feature of each must be
made clear at a glance.
c. Prevents continuous numbering in those sections where a table is omitted
from the series because it is not applicable for the given area or group covered
by the section. That is, if table 4 is omitted for Idaho, table 5 of the standard
series should not be renumbered as table 4 since then table 4 (and following tables)
for Idaho would not be comparable with table 4 (and following tables) for other
States.
d. Requires specific explanation of table omission at the point where the
omitted table would have appeared. T h a t is, the explanation should appear at
the point where the user would expect to find the table, not merely in the table
of contents.
Examples of statements inserted in t h e tabular section following tables 3 and
34, respectively.
(Table 4 of the standard series for the S t a t e omitted because there
are no cities of 50,000 or more in Idaho reported by wards)
(Tables 35 to 52 of t h e standard series omitted because there are no
cities of t h e given size-group in this State)
e. Converts the listing of tables on the contents page of the combined volume
to a tabular arrangement. Where all table titles and numbers have been standardized, there is no point in repeating t h e same series of titles for all similar areas,
with the page number opposite each listing. Instead, the contents page should
specify t h e important features of t h e standard scheme and provide the list of
standard titles. A tabular index should then be presented showing, for example,
t h e S t a t e names in the stub, the standard table numbers in the boxhead, and t h e
applicable page numbers in the cells.
f. Requires establishing of standard titles within close limits. Only minor
variations are permissible. Outstanding differences tend to run counter to the
purpose a n d function of standardization.
g. M a y demand specialized treatment for page running heads. If, as usually
is true, t h e standard titles used omit the specific name of t h e area or group common to t h e entire section, t h a t name must appear in the running head of the page.
Thus, if t h e titles end with the words
. F O R THE STATE: 1940, the running
head of the page should provide the State name.

Sec. 4-C. Use of Prefix Letters (421-424)

421. Definition.-A capital letter (rarely a small letter) placed in
front, and considered a part, of the table number. Example: The
A and J in TABLE A-1 and TABLE J-12, respectively.
422. General use.-The use of prefix letters usually is restricted to
two situations:
a. To distinguish a series of tables appearing in the text, or in appendixes, from
the main series appearing in the tabular section of a report; or
b. To identify a standard table scheme of a compound nature. (See par. 424.)

62

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEc. 4-C

USE OF PREFIX LETTERS

PAR.

423

423. Text and appendix tables.-Here, the use of prefixes may arise
from a desire to use an Arabic-numeral scheme but to avoid any implication that the text or appendix tables are part of the same basic
series as the detailed tables. (See par. 407.)
a. Text t a b l e s . - N u m b e r s may be run in a single continuous series using Arabio
numerals, all prefixed by the capital letter A, as TABLE A-1, A-2, etc.
Again, if the text is divided into chapters or readily recognizable sections of
any length, prefix A may be used for the first chapter or section, prefix B for the
second, etc., starting with 1 again for each such section, as TABLE A-1, TABLE
A-2 . . ., TABLE B-1, TABLE B-2 . . ., etc. However, this system has the obvious
disadvantage of complexity and, in the case of numbered chapters, the prefix will
not tie in with the chapter number.
b. Appendix t a b l e s . - I n a single appendix, prefix A may be added to all table
numbers, running them as TABLE A-1, TABLE A-2, etc. This makes clear that
the appendix tables are not continuous in development with the detailed tables.
This method is impracticable, however, if prefix A already has been assigned to
text tables. In that case, the use of prefix B for appendix tables might imply a
relationship to text tables which is unlikely to exist.
T h e prefix scheme is most useful where two or more appendixes exist which
might appropriately be designated as appendix A, appendix B, etc. Tables in
appendix A may then be numbered TABLE A-1, TABLE A-2, . . .; and those in
appendix B numbered TABLE B-1, TABLE B-2 . . ., etc.

424. Standard-series tables.-Where possible, prefix letters should
be avoided in standard-series arrangements. As in any reference-code
scheme, that which seems simple and obvious to the designer may
prove confusing to the user who is interested in occasional quick reference rather than in intensive study or analysis. Where used, the
prefix scheme should be as easy to grasp as possible. Also, it should
be explained clearly in the table of contents or introduction.
a. Basic principles.-In any prefix scheme for table numbers, reader-grasp will
be helped by maintaining the following principles:
(1) T h e n u m e r i c a l c o m p o n e n t (as 2 in TABLE A-2) represents an identity
of table form and classification of data. T h a t is, all tables with the same
numerical component in their table number (all tables numbered 2) should
provide strictly comparable data, with the only variable that of universe or area
as symbolized by the alphabetical prefix.
(2) T h e alphabetical c o m p o n e n t (as prefix A in TABLE A-2) represents an
identity of universe or area. T h a t is, within a given section of the report all
tables in series A, as indicated by the prefix, should be for the same universe or
area. Here the variable is one of content and classification as symbolized
by the numerical component.
(3) S u m m a r y e x a m p l e . - I t follows t h a t tables A-1 to A-10 should always
be for the same major group or area, but they may be expected to differ in the
subgroup shown or in the way the data are classified. Contrariwise, tables
A-4, B-4, C-4, D - 4 , etc., should always present the same subgroups classified
in the same way, but they may be expected to differ in the universe or area. for
which the data are distributed.
b. Circumstances of u s e . - P r e f i x schemes are not advisable in standard-series
tables except as a last resort and then only in a compound series. In general,


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

63

PAR.

424

THE TABLE NUMBER

sEc. 4-C

prefix schemes in standard-series tables should be confined to cases where all data for a
given universe or area must appear together, and the universe or area designation
cannot be shown effectively in the running head.
(1) Not needed in simple s e r i e s . - P r e f i x e s should not be used in a simple
standard series such as a standard series repeated for each State. Here the report
tends to organize in State sections, and the resultant presence of the State name
in the variable running head makes a prefix scheme unnecessary.
(2) Combining tables to avoid prefixes.-When comparable d a t a are to
be presented for a number of areas in the form (for example) of standardized
tables 1 to 10, separate tables for each area need not be established if the
TABLE 1 for the first area is followed immediately by TABLE 1 for each other
area; and the resultant group of tables numbered 1 is followed by TABLE 2 for
each area, etc. Here the normal expedient of combining all tables numbered 1 into a single multipage table solves the problem. Assigning prefixes
to the table number is then unnecessary since there is only one TABLE 1 , one
TABLE 2, etc.
(3) Use of prefixes in a c o m p o u n d series of standard t a b l e s . - O n occasion, a report will consist of a section for each State with a standard series
of tables presented not only for the State as a whole, but also for subareas
within the State. In cases where the nature of the data makes it essential t h a t
all d a t a for a given subarea be kept together, standardized tables 1 to 10 may
be presented for the first subarea; followed by tables 1 to 10 for t h e second
subarea; etc.
Here the running head of the State section distinguishes tables for one S t a t e
from those for another. But the running head cannot also be expected to
differentiate, within each State section, between the table 1 for each subarea
(such as table 1 for the State as a whole) and table 1 for each other subarea
(such as t h a t for the first major city in the State, for the second major city,
etc.).

This problem may be solved by assigning prefix A to the tables for the first
subarea; prefix B to the tables for the second subarea; etc. Then standardized
tables A-1 to A-10 are followed by standardized tables B - 1 to B-10, etc.

Sec. 4-D. Use of Suffix Letters (431-435)

431. Definition.-A small letter placed after, and constituting a
part of, the table number. Example: The a and c in TABLE 2a and
TABLE 8c, respectively.
432. Basic principles of table relationship and position. The use of
suffix letters is properly limited by the following principles of table
relationship and position.
a. Table relationship.-Tables with suffixed table numbers should bear an
easily noted, preferably obvious, relationship to their " m a s t e r " table (that is,
to the table of the same number without a suffix), even though the master table
does not appear.
b. Table position.-Tables with suffixed table numbers constitute a subseries
and should not be separated from each other. The master table should appear
first, followed by its suffixed tables. In turn, t h e suffixed tables should be
arranged in alphabetical order of suffix, as table 5, 5a, 5b, etc.

64

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

•

sEc. 4-D

USE OF SUFFIX LETTERS

433. General objectives in usage.
automatically assigned whenever the
appear. Even within the framework
lationship and position, they are best

PAR.

433

Suffix letters should not be
qualifying table relationships
of the basic principles of reused only-

a. To bring out the table relationships in a series where the relationships might
be seriously obscured by the normal numbering scheme; or
b. In standard-table schemes to permit insertion of added information for certain areas or universes without disrupting either the standard table form or the
standard numbering scheme.

434. Types of relationship to master table.-Three general types of
relationship to the master table are most frequently found, as follows:
a. Component-part tables.-These are tables which basically constitute asingle
whole but are split apart (but follow one another) for reasons of convenience.
T h e relationship is commonly stated as "Tables 5a, 5b, 5c, and 5d add up to
table 5." More precisely, the addition of corresponding cells in tables 5a, 5b, 5c,
and 5d totals to the entry in the corresponding cell of table 5. Where the d a t a
are more significant when presented by subgroup t h a n when combined, the master
or totality table may not appear.
(1) When t o t a l i t y table appears.-Suffixes identify t h e component-part
tables as subordinate parts of the totality table.
(2) When t o t a l i t y table does not a p p e a r . - S u f f i x e s unify t h e componentpart tables by indicating that they are all part of the same basic presentation.
b. Expansion table following master table. Here t h e suffixed table provides a
more detailed distribution of only one of several physical parts of the master
table; or it presents comparable data for only one of several components of the
master-table data. Suffix use for this situation is largely confined to "extra"
tables in a standard-table scheme.
(1) More detailed distribution for one part of a m a s t e r table.

The

suffixed table (such as table 5a) shows one panel (or block) of table 5 in greater
detail t h a n t h e form of table 5 permits or t h a n it is practicable to present for all
panels (or blocks) of table 5.
Example: Table 5 shows 5-year age groups by race, with panels for all classes,
white, Negro, and other races. Additional age detail (single years) is desired
for the white, not for the other race groups. A new table is inserted, therefore,
showing single years of age for the white population, and becomes table 5a.
(2) Comparable figures for one c o m p o n e n t ' of a m a s t e r t a b l e . - A

special case of the component-part table scheme. Here, the figures are required
separately for only one of the several components which make up the combined
figures shown in the master table.
Example: Table 6 shows a given distribution for all races combined. Table
6a shows the same distribution for the nonwhite only. No table is presented
for the other component, the white.
c. Derived table following a frequency t a b l e . - W h e r e a frequency table is
followed immediately by a table presenting a percent distribution for those figures
(or medians, rates, etc., relating to them, or derived from them) the relationship
may be expressed by a suffix letter.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

65

PAR.

434

THE TABLE NUMBER

sEc. 4-D

This use of suffixes is common in standard table schemes where it is desired to
present derived data for certain tables in certain areas only. In general, these
added suffix tables should be treated as standard tables when so used. T h a t is,
table 7a for one area should correspond to table 7a for another.
Examples: (1) Table 5 presents frequencies. Table 5a presents a percentage
distribution of table 5. (2) Table 7 presents frequencies. Table 7a presents
the percentage relationship of cell frequencies in table 7 to corresponding cell
frequencies in some other table. Table 7b presents averages for data in table 7.
Note t h a t the relationship between tables 7a and 7b is not direct, but both
suffix tables are obviously related to table 7.

435. Summary and examples of suffix numbering.-The following
summarizes and illustrates suffix numbering.
a. Component-part tables following totality t a b l e . - A s s i g n t h e regular, or
master, number to the totality table. Add a suffix to each component-part table
and arrange alphabetically in order of suffix.
Right: Table 8 (totality) ; table 8a (male) ; table 8b (female)
Wrong: Table 8a (totality) ; table 8b (male) ; table 8c (female)

b. Component-part tables without totality t a b l e . - F r e q u e n t l y these are mIsnumbered. If the totality table is not present, the regular or master number
(standing alone) should not be assigned to any table. However, the componentpart tables should carry suffixes as though the master table were present.
Right: Table 8a (male) ; table 8b (female)
Wrong: Table 8 (male) ; table 8a (female)

c. Derived tables following frequency t a b l e s . - H e r e , t h e master or base table
should carry a regular number without a suffix. The derived table should carry a
suffix. If there is more than one derived table, arrange them in order of analytical
or reference convenience and assign suffixes alphabetically.
Right: Table 8 (frequency or base table); table 8a (percent distribution);
table 8b (medians)
Right: Table 8 (frequency table) ; table 8a (medians); table 8b (cells in
table 8 as percent of cells in some other table)
Wrong: Table 8a (frequency table) ; table 8b (first derived table) ; table 8c
(second derived table)

d. Expansion tables (all types).-The same principle applies as for derived
tables. T h e master table should carry a regular number without a suffix. The
expansion table should carry the suffix a. If more than one expansion table
appears, arrange them in analytical or reference order and assign suffixes alphabetically.
Right: Table 8 (master t a b l e - a g e by race) ; table 8a (single years for white)
Wrong: Table 8a (master table) ; table 8b (single years for white)
Right: Table 8 (all races combined) ; table 8a (nonwhite only)
Wrong: Table 8a (all races) ; table 8b (nonwhite only)

66

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Chapter 5

THE HEADNOTE (501-523)
Sec. 5-A. General (501-506)
501. Definition and function.-A statement (or statements) enclosed in brackets and appearing between the table title and the top
rule of the table.
In general, the headnote is a means of providing information (essential to the understanding of the data) when the point in question cannot be made clear elsewhere in the table or is too general for effective
handling in footnotes.
Specifically, the headnote provides information qualifying, explaining, or otherwise relating toa. The table as a whole (primary function);
b. An easily identified major portion of the table; or
c. A constantly recurring specific term, abbreviation, or situation.

502. Relationship of headnote to title.-Both the headnote and
title have as their primary function the explanation or description of
the table as a whole. Their close relationship is manifested in two
principal ways; that is, the headnote may acta. As a limited form of s u b t i t l e . - T h e headnote frequently provides additional
title information crowded out of the title because of the need for brevity; or
b. As a title f o o t n o t e . - T h e headnote may define a term in t h e title or qualify a
specific statement made there. In this way it acts as a footnote to the title without the necessity of a reference symbol. This is one of the most important uses of
the headnote.
Formal footnotes to titles (reference symbol in title and footnote at bottom of
table) are undesirable. Usually, they reflect a contradiction in function. Ordinarily, a title footnote represents a misplaced headnote.

503. Relationship of headnote to footnote.-A sharp distinction
must be drawn here if confusion in usage is to be avoided. The footnote is used to explain or qualify a specific cell, line, column, or portion
of the table which is clearly identified by placement of the given
reference symbol. In contrast, the headnote primarily qualifies the
table as a whole; its secondary use to explain a particular part of the
table should be sharply restricted since the user is not given a warning

776554
49

6


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

67

PAR.

503

THE

HEADNOTE

SEC.

5-A

at the place where the data actually appear. These restrictions are
indicated below. Where the specified situations do not exist, eXplanations of major portions of tables, specific terms, etc., should be
handled in footnotes.
a. Headnote relating to an easily identified major portion of the table.Properly restricted to cases where either(1) T h e given major portion is referred to in a title statement which, in
itself, needs qualification or explanation, or
(2) The table arrangement is such t h a t the table portion referred to cannot
be clearly indicated by placement of a footnote reference symbol. In some
cases, there may be no place to put the symbol within the table. In other cases,
the limits of the mass of data to which the symbol refers may not be clear.
b. Headnote explaining specific terms, abbreviations, e t c . - P r o p e r l y restricted
to cases of constant recurrence; that is, to cases where the given item keeps recurring
a n d the resultant multiplicity of identical recurrent footnotes would detract from
the presentation rather t h a n add to it.
A special case of this type, frequently encountered, is the headnote explanation
of a standard reference symbol (such as an asterisk), a mathematical sign (such as
a minus sign appearing in a column headed "Increase"), or a standard code
symbol used throughout the table. Here the symbol or sign is not only recurrent;
it is so distinctive that once noted the reader tends to carry its meaning in his
mind. Note, however, that recurrent reference symbols in the form of numbers () are properly explained in foot.notes; never in headnotes.

504. General restrictions on headnote u s e . - T o be included in a
headnote, a statement should be both (a) essential to the understanding of the data and (b) a point which cannot be made clear in the title
or in the table proper. The ideal table would carry no headnote
since in the ideal situation all of these points would be brought out
• clearly in the title, box, or stub. Following are several general
injunctions as to headnote inclusion.
a. A void using headnotes where possible but do not hesitate to insert them
where needed.
b. Avoid using headnotes for general discussion or for explanations of general
interest. Such discussions or statements usually belong in the text.

Note: If it is impracticable to include text in t h e report, or the text must be
closely restricted as in the Statistical Abstract of the United States, t h e burden of
general definition and qualification must fall largely upon the headnote. This
does not affect t h e general rules provided here. It merely exemplifies necessary
adaptation to an extreme situation.
c. Rely upon text for definitions of terms. Headnote only where the term is
not used as defined in text or is subject to gross misinterpretation.
d. Do not present in the headnote an explanation common to all tables in the
series unless the omission is likely to result in a definite misapprehension. Place
t h e explanation in the text.
e. In general, do not use headnotes to refer to text for an explanation which (1)
the user would normally expect to find in the text, or (2) could not logically be
found anywhere else.

68

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEc.5 - A

GENERAL

PAR.

504

f. Unless a given table has been designed specifically to make possible the presentation of a chart, avoid headnotes which imply t h a t t h e chart is the reason for
the table's existence, such as "Statistics for chart IX."
Where charts are prepared to facilitate grasp of general relationships, the user
should be directed to a source table for the figures. Indicate the tabular source
upon the chart as "SOURCE: TABLE 12" or "See table 12." The headnote of
table 12 might well include t h e statement "See also chart IX."

505. Repetition on "continued pages" of multipage table. Four
practices are described here, listed in order of preferred usage.
a. Repeating headnote in f u l l . - A common Census Bureau practice is to repeat
the headnote in full on each page of the table. This method simplifies operations,
particularly when standard table forms 1 are used. It is least successful in the
case of extremely long headnotes.
b. Omitting headnote on continued pages.-This method is theoretically sound
since a headnote placed at the beginning of the table covers t h e entire table. In
practice, short headnotes can be repeated without material trouble and the
user is helped by them. Omission on continued pages is best practiced where
the headnotes are necessarily long and detailed.
c. Inserting reference to first page.-Here, the continued pages carry a headnote reading "See headnote, p.
or "See headnote at beginning of table."
This warns the user, and the reference occupies little space on the continued
pages.
d. Repeating pertinent portion o n l y . - T h i s practice provides for repetition (on
continued pages) of only that portion of the headnote applying specifically to the
given page. In using this scheme several precautions are necessary.
(1) First establish the basis of selection. Then make sure t h a t the same
criteria are applied for each continued page. (The full headnote is shown on
the first page.)
(2) Review each continued page after it is in type, or typed for offset, to
make sure t h a t paging and make-up adjustment have not so shifted t h e d a t a
t h a t t h e headnote portion shown no longer applies.
(3) Avoid footnotes on continued pages reading merely "See headnote."
Where used, be sure t h e portion of t h e headnote at t h e top of t h a t page contains t h e information needed. It may not occur to t h e reader to t u r n back
to t h e first page where t h e entire headnote appears. (See par. 615b.)

506. Practices to avoid. Several questionable practices are sufficiently common to warrant special attention.
a. Arbitrary restriction of headnote to a single s t a t e m e n t . - B a s e d on a false
premise, this idea is incorrect in theory and causes serious trouble in application.
T h e headnote should be kept as brief as possible, but as many statements should
be included as the needs of t h e presentation demand.
b. Footnoting h e a d n o t e s . - A footnote to a headnote is like a footnote to a
footnote. Both are incorrect.
c. Beginning with " N o t e : " . - T h e headnote should not begin with "Note:" or
"General note:". T h e existence and placement of t h e note are sufficient to
identify it as such; t h e use of brackets makes clear it is not a subtitle. There
1 That is, page forms printed in advance with identical stubs, boxheads, headnotes, and titles. Then the
cell entries are filled in for different areas. Where the same preprinted page form is used for all pages of a
multipage table, the complete headnote will automatically appear on all pages.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

69

PAR.

506

THE HEADNOTE

SEC.

5-A

is no more reason to identify it by starting with " N o t e : " t h a n there is for placing
"Title:" at t h e beginning of each title.
(See also pars. 615a-615c, which discuss objections to the footnoting of titles
and t h e use of footnotes reading "See headnote.")

Sec. 5-B. Specific Use of the Headnote (511-513)
511. G e n e r a l . - T h e following lists of specific uses, and the examples
cited, are subject to two important qualifications.
a. Illustrative o n l y . - T h e categories listed, and the examples cited, are merely
illustrative of headnote problems frequently encountered in Census Bureau work.
(1) N o t all-inclusive.-Equally appropriate headnote instances may arise
which are not covered here.
(2) Not m u t u a l l y exclusive.-Examples cited under one category may
apply equally well under another category, depending upon t h e wording of
t h e table title, the treatment of d a t a in t h e table, or t h e point of view.
(3) W o r d i n g n o t s t a n d a r d i z e d . - T h e wording of a n y particular example
is not to be construed as representing t h e "standard" wording for t h a t particular
situation. Such wording may properly vary depending upon t h e purpose a n d
circumstances of t h e presentation. In part, such variation tends to be a function of t h e given report. Within a n y one report identical' situations should
be handled identically.
b. Headnoting permissible; not m a n d a t o r y . - T h e cases cited represent permissive situations. A headnote should not be inserted in a n y table merely
because t h e prototype is included in these lists. Headnotes should be used when
they are essential to t h e understanding of t h e d a t a , not otherwise.

512. Qualifying table as a whole or an easily identified part of it.
a. To define or specify the unit of presentation.

In a long headnote, place t h e
unit of presentation at t h e beginning, preferably in boldface type.
[In tons of 2,000 pounds)
[Number of persons 14 years old and over working the equivalent of 2 or more days during
specified weeks]

b. To define, clarify, or qualify the title as a whole, or any part of i t :
[Data for irrigated farms limited to 20 specified States in 1940]
[Exclusive of stillbirths]

c. To describe the enumerative or reporting base or coverage, particularly when
it differs from table to table:
[Based on Sample D]
[Complete census returns]
[Based on Incomplete reports; totals may differ from those in table X which represent complete
coverage]

d. To indicate limitations of the data or reliability of estimates:
[Statistics for 1946 are subject to sampling variation. Figures identified by an asterisk (*) are
subject to particularly large sampling variation and should be interpreted with special
caution. See text statement "Reliability of estimates"]

e. To indicate the basis of selection of areas or items shown:
[Figures for nonwhite shown separately for areas having, in 1940, 25,000 or more nonwhite
women 15 to 49 years old]
(Limited to stores which report an analysis of their sales by commodities]

70

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEC.

5-B

SPEOIFIO USE OF THE HEADNOTE

PAR.

512

f. To describe the method of reading an involved table:
[The first column shows the number of births which occurred in the given city, regardless of
whether the mothers' permanent home was in that city. The second column shows the
number of births to mothers whose permanent home was in that city, regardless of whether
the births occurred in that city or elsewhere]

To refer the user to the text or to a related table:
[For definitions of terms, see text, p. 24]
[For total number on all farms (base for percentages), see table 652]

h. To qualify or explain the significance of an implicit characteristic of the
presentation:
[Figures prior to 1916 relate to years ending June 30, thereafter to calendar years]

j. To warn against an unsound interpretation:
[Relative wealth of the several States is not shown by these figures because of differences in the
basis of assessment and practices thereunder]
[Increase from census to census includes that due to annexation of territory as well as to direct
growth}

k. To specify source of data where a source note does not appear at the bottom
of the table:
[Based on detailed table 24]
[Figures for 1940 are those reported in the census for that year. Estimates of future population
are reproduced with the cooperation of the National Resources Planning Board and of
Warren S. Thompson and P. K. Whelpton, of the Scripps Foundation for Population
Research, who prepared them for the National Resources Committee]

513. Qualifying a constantly recurring factor.-A headnote may be
used to qualify or explain a constantly recurring specific term, abbreviation, or situation in order to avoid the multiple recurrence of a
standardized footnote.
a. To explain the significance of a standardized reference symbol such as an
asterisk (*):
[For explanation of asterisk (*), see text statement, "Coefficient of Variation"]
[Places marked with an asterisk (*) were classified as urban in 1940]

b. To explain significance of a mathematical sign, such as a minus sign ( - )
in an "Increase" column:
[A minus sign ( - ) denotes decrease)

c. To indicate that derived data are not shown where less than a specified
value or where based on less than a given amount or number:
[Percent not shown where less than 0.1 or where base is less than 100]

d. To define or qualify a term constantly recurring in the box or stub:
(Figures for "All dwelling units" include vacant units]

Sec. 5-C. Wording of the Headnote (521-523)

521. Frame of reference.-Headnotes can be shortened, and phrasing simplified, by taking into account the basic' frame of reference of
the table. The headnote is an integral part_of/the table and should
be handled accordingly.
a. Basic reference is to table.-Headnote statements, by definition, refer
solely to information shown in the given table unless otherwise specified. This
basic principle affects headnote wording in two ways.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

71

PAR.

521

THE HEADNOTE

sEc. 5-C

(1) It reduces the need for specifying "These statistics" or "Statistics
in this table" when referring to t h e data in general.
Terse: [Based on Sample D]
Verbose: [The statistics in this table are based on Sample D]

(2) It requires specification of the category or item referred to if the statement does not apply to the entire table.
Vague: (Based on Samples C and W]
Clear: [Statistics for 1940 based on Sample C; those for 1910 based on Sample W]
Clear: [Statistics based on Sample D except "total population" which represents result of a
complete census]

b. Order of w o r d i n g . - T h e headnote should start with that which the
can see, or is likely to assume, and move to that which is not apparent
known to him. T h e reader should not be asked to read an explanation
he discovers what is being explained; or to read a qualification of an item
he discovers what the item is.

reader
or not
before
before

Accurately stated: [Because of differences in the basis of assessment and practices thereunder, figures
do not show the relative wealth of the several States]
But this is better: [Relative wealth of the several States is not shown by these figures because of differences in the basis of assessment and practices thereunder]

522. Need for brevity.-Under the best of circumstances, headnotes
are hard to read. Special attention needs to be paid to brevity, but
clarity must come first. The following should be kept in mind:
a. T h e coverage of many different points prevents rapid reading and reduces
the chance t h a t any one part will be noted.
b. Reader patience is quickly exhausted because of(1) Mechanical reasons: T h e smallness of t y p e and great length of line
make headnotes hard to read.
(2) C o n t e n t a n d f o r m : Headnotes are parenthetical in nature. Frequently, they comprise a series of unconnected ideas. This makes them hard
to comprehend.

523. Style.-Headnotes should be written in telegraphic style.
Articles and other parts of speech not essential to understanding may
be omitted. Standard conventionalized phrasing should be used
wherever possible. Use the same abbreviations and phrasings in the
headnote as in the remainder of the table. Unrelated statements
should not be joined by a semicolon merely to reduce the number
of headnote "sentences." Use as many sentences as required.
Lower case is used throughout the headnote; do not use caps and
lower case.2 The entire headnote is enclosed in brackets. Where
individual tables are typewritten for inclusion in correspondence,
parentheses may be used instead of brackets, provided that parentheses
are not also used within the headnote.
No period at e n d . - N o period should appear at the end of the headnote even though the last statement is a complete sentence. Periods
are used internally where sentence structure warrants them. (In
contrast, footnotes always end with a period.)
This Is in Caps and Lower Case. T h i s is in lower case,

72

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Chapter 6

THE FOOTNOTE (601-665)
Sec. 6-A. General (601-606)
601. Types of footnotes.-As the term implies, any statement or
note inserted at the foot (or bottom) of the table may be considered a
"footnote.
In practice, it may take one of three distinct forms, each subject to
its own rules and conditions of proper usage. The three forms may be
identified as follows:
a. The specific footnote, involving reference to a specifically designated portion
of the table.
b. The source note, used to indicate the source of the data. This might
reasonably be considered a distinctive component of statistical tables rather t h a n
a type of footnote. It is included here as a matter of convenience in discussion.
c. The general footnote, involving reference to the table as a whole or major
portion of it.

602. Scope of discussion.-Most problems of footnoting arise in
connection with the specific footnote. The general footnote is virtually never used by the Bureau of the Census since its normal
function is adequately taken care of by the use of the headnote. The
source note involves the dual problem of appropriateness of insertion
and correctness of wording. The latter problem (correctness of wording) involves technical practices in the field of bibliographical reference,
a subject not within the scope of this manual. However, a general
discussion of the source note is included as section 6-C.
In the present section (sec. 6-A) each of these three forms of footnotes is defined. The definition of the general footnote is accompanied by a brief statement of appropriate use. No further discussion
of it is provided in this manual.
603. The specific footnote defined.-Located at the bottom of the
table, the specific footnote is a "keyed" statement which qualifies,
describes, or explains the information presented in (or omitted from)
a specific cell, column, line, or group of columns or lines; or defines or
clarifies the meaning of a stub or box entry. "Keyed" means that the
item (or items) thus qualified is assigned a reference symbol which is
repeated at the beginning of the footnote, thereby relating the footnote


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

73

PAR.

603

THE FOOTNOTE

SEC.

6-A

to the point (or points) of reference. For detailed discussion of the
specific footnote, see section 6-B; for discussion of the use of reference
symbols, see section 6-D.
Example:

Includes 489 farms reporting no land owned or leased.

For the purposes of this manual, the specific footnote is considered
the "normal case." References to footnote practices appearing
throughout this book relate to the specific footnote unless otherwise
specified.
604. The source note defined.-Generally located at the bottom of
the table, the source note is a statement indicating the specific source
of the statistics, or of a designated portion of them. (See sec. 6-C.)
On occasion, it may comprise the headnote, or appear as a part of it.
General example:
Source: Federal Security Agency, Social Security Board, Bureau of Employment Security;
Social Security Yearbook, 1942.

605. The general footnote defined.-(Rarely used by the Bureau ot
the Census.) Located at the bottom of the table, the general footnote
is a statement which qualifies, describes, or explains the table as a
whole, or an easily identified major portion of it. It is introduced by
the word "Note" or "General note" followed by a colon.
Comparison of the above definition with that for the headnote (see
par. 501) will show a virtual identity in concept. This fact, a cause
of considerable confusion in the field of table design, may be attributable in part to historical development. The headnote appears to be a
comparatively recent innovation intended to bring to the early attention of the reader important information which otherwise might not
be brought to his notice until the end of the table is reached. The
reasoning involved may be likened to that which has established as
Census Bureau practice the placement of group totals above, rather
than below, their component parts.
606. The general footnote: Restricted u s e . - T h e general practice
of the Bureau of the Census restricts the use of the general footnote
toa. Tables presented without titles, such as text tabulations, or the special case
of the release table without title (see par. 125). Here, the headnote cannot be
used since it is physically an adjunct of the title. Where t h e d a t a affected are
sufficiently localized within the table (which has no title) to permit placement of
a reference symbol, such headnote-type of statement ordinarily should be inserted
as a specific footnote. Where t h e localization of reference is impracticable, or
where a multiplicity of identical footnote reference symbols would result, the
general footnote may be used.
Examples (placed at bottom of tables without titles):
Note: Statistics exclude 6 counties for which data were incomplete.
Note: Median not shown where base is less than 100.

74

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

SEC.

6-A

GENERAL

PAR.

606

b. Tables presented without text; specifically, tables accompanying correspondence. At times it is desirable to confine a transmitting letter or memorandum to nontechnical material and to include detailed technical explanations or
notes on the tabular attachment. T h e use of t h e general footnote is appropriate
where the desired explanation is too long to permit incorporation in the headnote
or the nature of the discussion makes headnoting inappropriate. Usually such
notes consist of a discussion or detailed explanation of the d a t a in technical terms.

Sec. 6-B. The Specific Footnote (611-615)
611. Purpose of specific footnote.-The specific footnote is a means,
of· providing information essential to the understanding of the particular datum or data to which it is keyed by the given reference
symbol. (The term "footnote" as used in the following materials is
to be interpreted as meaning "specific footnote.")
612. Relationship of footnote to headnote.-A sharp distinction
must be drawn here if confusion in usage is to be avoided. The
footnote is used to explain or qualify a specific cell, line, column, or
portion of the table which is clearly identified by placement of the
given reference symbol. The headnote primarily qualifies the table
as a whole. Its secondary use to explain a particular part of t h e
table is sharply restricted. These restrictions are indicated in some
detail under the discussion of the headnote (see par. 503). Where t h e
situations specified in that statement do not exist, explanations of
major portions of tables, specific terms, etc., should be handled in
footnotes.
613. Relationship between footnote, headnote, and t i t l e . - T h e title
describes the main features of the table. The headnote may relieve
the title of a portion of this burden by describing general features of
the table or by qualifying specific title statements. The footnote is
required when a highly specific part of the table (below the headnote)
must be explained. The relationship between the title and headnote
is frequently close; that between the title and footnote is usually
remote. Title-footnoting is undesirable since one of the primary functions of the headnote is to act as a footnote to the title without use of
a reference symbol.
614. General restrictions on footnote u s e . - T o be included in a
footnote a statement must be both (a) absolutely essential to the
understanding of the data, and (b) a point which it has not been practicable to make clear in the title, headnote, boxhead, stub, or field.
The ideal table would carry no footnotes since in the ideal situation
all such points would be brought out clearly elsewhere in the table.
The general injunctions concerning headnote usage described in
paragraphs 504a to 504e, inclusive, may be applied also to the use
of footnotes.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

75

PAR.

615

THE FOOTNOTE

SEO.

6-B

615. Practices to avoid.-Several questionable practices are sufficiently common to warrant special attention.
a. Footnoting t i t l e s . - T i t l e footnotes represent a contradiction in function.
Qualification of t i t l e is one of t h e most important and obvious purposes of t h e
headnote. Practically all headnotes qualifying t h e table as a whole would
qualify as title footnotes if such a practice were adopted generally.
An extreme case, sometimes encountered, is where a given qualification is
placed in a headnote where it belongs, but t h e title is also footnoted with t h e
footnote reading merely "See headnote." This interrupts t h e user when reading
t h e title and sends him to t h e bottom of t h e page (or to t h e end of t h e table if
t h e footnotes are assembled there) only to tell him to look up to t h e top (or at
t h e beginning) again.
b. Footnotes reading "See h e a d n o t e . " - A n extreme case of this type has been
cited immediately above in connection with title-footnoting. T h e more common
case arises from overcaution of t h e table-designer. T h a t is, after inserting a
headnote to qualify a recurrent item, column head, etc., t h e table-designer fears
t h a t it m a y be overlooked by t h e reader. Therefore, he also inserts a footnote
reference symbol against t h e particular item. Recognizing t h e absurdity of
repeating· t h e statement at t h e bottom of t h e page he tries to get around t h e
problem by making t h e footnote read "See headnote."
Such overcaution may defeat its purpose since t h e result is likely to distract
a n d even annoy t h e reader. Where this overcaution is coupled with concent r a t i o n of notes at t h e end of a multipage table t h e result can be infuriating.
If, in addition, t h e full headnote is not repeated on every page, t h e footnote
injunction "See headnote" may be misleading since t h e reader is likely to glance
at t h e portion of t h e headnote appearing at t h e top of t h e last page (where t h e
footnote statement appears) and find t h a t no part of t h a t headnote applies.
Solution: Either t h e statement should appear in t h e headnote a n d t h e footnote reference omitted from t h e item or column head; or t h e headnote statement
should be omitted and t h e explanation restricted to t h e footnote.
Undesirable method:
T A B L E 5 . - P O P U L A T I O N BY A G E A N D C O L O R :

1940

AND

1930

[Figures for 1930 white population revised to include Mexicans classified as nonwhite in the 1930 reports)
1940

1930

Age
Total
All ages_.

Under 45 years_45 and over_Not reported __
1

White

Nonwhite

Total

White I

Nonwhite

1, 243

957

286

1, 126

842

284

759
484

592

167

681

365

119

432

490

191

5

85
8

13

347

See headnote.

Preferred methods: (1) Omit the footnote reference symbol and t h e footnote;
rely upon t h e headnote. (2) Another solution would be to omit t h e headnote and
retain the footnote. In the case illustrated, the first solution is better since,
actually, both t h e white and nonwhite figures are affected.
c. Footnoting h e a d n o t e s . - A footnote to a headnote is like a footnote to a
footnote. Both are incorrect.

76

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

8EC.

6-B

THE SPECIFIC FOOTNOTE

PAR.

615

d. Beginning with " N o t e . " - W r o n g for a specific footnote since t h e presence of
the reference symbol identifies it sufficiently. Necessary for a general footnote.
e. Footnoting beyond the level of significance.-Avoid footnoting for detail
beyond t h e level of significance. This can become a nuisance to both the producer
and the user. Thus, when presenting d a t a in considerable detail it may be
necessary to call attention to deficiencies of statistics in the more detailed level of
classification. But when the qualified data are submerged in a presentation
confined to totals for major groups, qualification of details not shown separately
may become statistically insignificant or even inapplicable. In such cases the
footnotes may be omitted.
Example A: When dealing with areas, it is reasonable to footnote data up to
the county level for deficiencies concerning the minor civil divisions listed.
Carrying such footnotes to t h e State totals would rarely be appropriate;
carrying them to geographic division or region totals, or to United States totals,
would rarely be necessary.
Example B: When showing d a t a for manufacturers classified according to
detailed lists of products, it may be important to indicate that products of a
particular type which might be expected to be included in subgroup A are
actually included in subgroup C. If both of these subgroups are in the same
major group, then this footnote would be inapplicable in a n y presentation confined to major group totals with the affected subclassifications not shown
separately. However, shifts of this type involving subgroups in two different
major groups must continue to be noted.
f. Failure to footnote at all points of recurrence.-A footnote applies only to t h e
items covered by t h e reference. I t s effect is not carried to other points of recurrence of the same s t u b or column listing. Therefore, in a recurring situation
the footnote should appear at each point of recurrence, not merely the first time.
This goes back to a basic principle of footnoting; that is, t h a t the absence of a
footnote may provide as specific information as its presence but in the opposite

direction!
(1) Omission of footnote w i t h o u t w a r n i n g . - T h e examples assume t h a t
the stub listing "Widowed and divorced" occurs twice in a given table, footnoted
the first time but not the second. The footnote gives no warning t h a t it is
supposed to apply throughout the table.
Example (C-1. First occurrence:
Single_--

Married_Widowed and divorced i _ - - 1 Includes unknown marital status.

Example C-2. Second occurrence:
Single__-

Married _
Widowed and divorced __

Interpretation: This combination clearly states two things to t h e reader.
On t h e first occurrence, t h e presence of the note warns him specifically t h a t
"Widowed and divorced" includes persons of unknown marital status (not
normally so included in t h a t category). On the second occurrence, the absence
of the note (in view of its presence on t h e first occurrence) clearly implies t h a t
"Widowed and divorced" covers exactly what t h e listing states; t h a t is, it does
not include persons of unknown marital status.
(2) Warning t h e reader of subsequent o m i s s i o n . - T h e situation is
scarcely improved by including in the above footnote a warning t h a t it covers
all recurrences, thus:
Example D:
1 Throughout this table (or series of tables) the category "Widowed and divorced" includes
persons of unknown marital status.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

77

•

PAR.

615

THE FOOTNOTE

SEC.

6-B

Firstly, t h e reader is given no warning at t h e point of recurrence t h a t t h e
affected statistics are qualified elsewhere. This is particularly likely to be
serious in a multipage table.
Secondly, this is obviously a headnote-type of statement.
Exception: On rare occasions, this practice may be justified where all three
of t h e following conditions exist: (a) No other practicable means is available,
( b ) t h e qualification is noted solely for t h e sake of the record, and (c) its effect
on the d a t a is statistically negligible.
(3) Solution: Either use a recurrent footnote or a headnote. If neither is
practicable, examine t h e possibility of inserting an explanation in t h e accompanying text, if any. It is more reasonable to expect the reader to examine
t h e text t h a n to expect him, upon referring to certain figures, to examine all
footnotes on all previous pages on the chance t h a t a qualification of the given
figures may have been stated there but not repeated. This difficulty does not
arise with t h e headnote since t h e reader is obligated to read it, just as he is
obligated to read the table title.

Sec. 6-C. The Source Note (621-624)
621. The source note defined.-Generally located at the bottom of
the table, the source note is a statement indicating the specific source
of the statistics or of a designated portion of them. On occasion,
it may comprise the headnote, or appear as a part of it.
General example:
Source: Federal Security Agency, Social Security Board, Bureau of Employment Security;
Social Security Yearbook, 1942.

622. General principles.-Where a table presents survey results
directly and the report is a primary source, source notes are not needed
since the report itself is the source.
Where the tables include data from other surveys for background
or comparison, or where the report is secondary in nature, each table
should have a source note so that the user may, if he wishes, examine
the figures in the original setting or obtain additional detail.
By "primary source" is meant, generally, the first report of a survey
(or given phase thereof) prepared by the responsible agency. By
"secondary source" is meant any document making use of, or presenting, data previously published elsewhere; that is, in an earlier report
of the same agency or in a report issued by another agency.
Most publications of the Bureau of the Census constitute primary
sources since their function is to report to the using public the basic
results of censuses and surveys conducted by the Bureau. For this
reason, the source note, as such, is given only cursory treatment here.
The Statistical Abstract of the United States may be referred to for its
many examples of source notes.
623. Appropriate u s e . - I n general, the source note serves two
functions: (a) It refers the user to the original source, and (b) it
gives credit to the originating person or organization.

78

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEc. 6-C

THE SOURCE NOTE

PAR.

623

From a statistical reference standpoint the first function is more important. However, the second function must not be overlooked, particularly where the information originates outside the Bureau of the
Census.' W h e r e Bureau records and reports are involved, the statistical reference aspect assumes paramount importance.
General rule: If material is used originating outside the Bureau of the
Census1 always give credit in Bureau reports, even though statistically
the reference is neither significant nor particularly useful. If the data
originate inside the Bureau of the Census,1 let the statistical reference
usefulness decide.
Specific rules: Ordinarily, in Bureau of Census operations, the
source is cited only when all, or a portion, of the statistics shown in the
table have been transcribed, abstracted, or adapted. a. From a document or volume published outside of the Bureau of the Census.If the figures as cited differ in a n y way (other t h a n rearrangement) from those in
the report from which taken, the source note should so state, or at least imply, by
use of such opening terms as shown in the following examples:
Examples:
Source: Based on data from_Source: Adapted from_
Source: Selected from __

Includes Bureau of Census adjustments for __

b. From information furnished in unpublished form by a source outside the
Bureau of the C e n s u s . - H e r e t h e citation should refer to t h e organization or
government department or bureau from which the information has been obtained.
If based on general worksheets or records the source note may so indicate by
using the terms " N o t published elsewhere" or "Official records." If transcribed
or adapted from a manuscript not yet published, the formal name of t h e manuscript may be cited with an appended statement " I n press," " I n preparation,"
or " U n p u b l i s h e d m a n u s c r i p t , " as t h e c i r c u m s t a n c e s w a r r a n t .

(See p a r . 624.)

Examples:
Source: U . S . Department of the Interior, Bureau of Reclamation. Official records.
Source: U . S . Works Projects Administration, Division of Research, Statistics, and Finance;
Manual of Instructions for the Preparation of Text, Tables, Charts, Maps, etc., for Publication
Purposes. Unpublished manuscript.

c. From information prepared by, or abstracted from the published reports of,
Subject Divisions of the Bureau of the Census other than the Division issuing the
table which is being footnoted.-A rule of reason should be followed here with
primary emphasis upon direct usefulness to t h e user. Thus, in a report of the
Bureau of t h e Census, there is little point in indicating source for a single population figure. On the other hand, there may be a great deal of value to the user in
an indication of the source of figures involving a detailed subclassification in a
specialized field.
Examples:
Source: Bureau of the Census, Sixteenth Census Reports. Agriculture. Territories and
Possessions, p. 112.
Source: Bureau of the Census. Facts for Industry, Series M35a. Farm Machines and
Equipment, 1945, p. 24, table 7.
Source: Bureau of the Census. State and Local Government Special Study, No. 20 (final).
Financing Federal, State, and Local Governments, 1941, p. 17.
1

Assuming that the Bureau of the Census is the agency preparing the table on which the source note is to

appear.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

79

PAR.

623

THE

FOOTNOTE

sE0.

6-C

d. From the detailed tables of a given report for use in text tables at the beginning of the same report.-Here again, t h e use of a source note should be restricted
to cases where it will be of significant help to the reader. This t y p e of source note
is frequently inserted in the headnote.
Examples:
Source: Detailed table 19.
Source: Derived from basic data shown in table 32 for Illinois.

624. Citation of manuscripts not yet published.-Two rules should
be observed rigorously. First, a manuscript or report which has not
been published should never be cited without appending an indication
of status, as " I n press," " I n preparation," etc. Second, misuse of
the term " I n press" should be guarded against. It means exactly
what it says; that is, that the report or volume is physically in the
printing office with instructions to print. It does not mean that it
is ready for the printer or that it will be sent to the printer shortly.
The reason why the term " I n press" should be maintained inviolate
should be clear. From the standpoint of the future user, there is
little distinction between a volume indicated as existent (in published
form) in a source note written five years earlier, and a volume described as " I n press" at that earlier date. In either case, the future
user expects that the volume cited will be in existence in printed
form (at that future time), that it will be titled exactly as cited, and
that it will contain the material credited to it.
The following usages are suggested:
a. Use "In press" if t h e manuscript is physically in the hands of the printer
with instructions to print.
b. Use " I n preparation" where work on the report is actually under way and
printing plans are reasonably definite.
c. Use "Unpublished manuscript" if t h e work has been halted after substantial
completion (including at least a first draft of t h e text), and final completion is
uncertain. Use this term also where a report has been completed but printing
plans are uncertain.
d. Use "Official records" if the material cited is from material which it is not
planned to publish, or where a report on t h e subject is merely contemplated.
e. When in doubt as to the exact status, do not use " I n press." In general,
"Official records," or " I n preparation," are preferable or some other term t h a t is
indefinite as to the assured future existence of the unpublished volume.

Sec. 6-D. Reference Symbols: Type, Placement, and Sequence (631-640)
631. Reference symbol defined.-A reference symbol or reference
mark is an identifying sign used to key together two or more physically
distinct portions of a table or report. In tabular presentation it is
typically used to refer the reader to a footnote, although occasionally
a reference symbol (other than numerical) may be explained in a
headnote.
*791
Examples: 6,452
38,512
80

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

SEC.

6-D

REFERENCE SYMBOLS

PAR.

632

632. Function and purpose.-The reference symbol refers the
reader from one part of the table to another. I t s presence in the stub,
boxhead, or field warns the reader that an explanatory or informative
statement bearing the same symbol may be found among the footnotes
or in the headnote. Contrariwise, the presence of a footnote or headnote bearing a reference symbol indicates that the same symbol will
be found elsewhere in the table, affixed to the data being qualified.
633. Types of symbols.-Three principal types of reference symbols
are in common use in the general field of tabular presentation: Arabic
numerals ('); capital or small letters (*) (°); and arbitrary symbols
such as the asterisk (*). Census Bureau practice is restricted almost
entirely to Arabic numerals except for "recurrent reference" situations where the asterisk (*) is frequently used (see par. 640).
a. Arbitrary s y m b o l s . - T h e use of various arbitrary symbols such as t h e asterisk
(*), dagger ( t ) , paragraph mark ( ) , etc., has largely been abandoned in general
material because of their unsightliness and their inherent lack of position significance in terms of a series. T h a t is, if several different footnote references are
used in a given table, the location of a n y one of these arbitrary symbols provides
no easily recognized guide as to where a preceding or succeeding symbol may be
found. Their current use is largely restricted to material involving mathematical
or chemical formulae where superior figures and letters (exponents, superscripts,
etc.) are already used with a different meaning.
b. Alphabetical symbols.-Capital or small letters are rarely used by the Bureau
of the Census for reference symbols. Their one advantage over Arabic numerals
lies in t h e lesser likelihood of confusion when a footnote reference symbol is placed
against a numerical entry. This danger is reduced, however, where superior
figures (see par. 633c) are used for reference symbols or where the reference
symbol is otherwise elevated or is identified by a caret or its typewriter
equivalent ( _ / ) .
c. Arabic numerals.-Superior figures, t h e preferred form of reference symbols,
are available both in letterpress (type-set) and on many statistical typewriters.
"Superior figures" or "superscripts" are figures which are smaller in size t h a n
those used for numerical entries, and are elevated in placement.
Examples:

12 # 4,

etc.

634. Form of symbol for letterpress versus typing.T h e proper
form of symbol for letterpress and typewriter use is as follows:
a. Letterpress.-Use superior figures.
b. Typing:
(1) Reproduction t y p i n g . - U s e superior figures.
(2) Other t y p i n g . - U s e superior figures where available. Otherwise use
full-sized figures set off as shown. In double-spaced typewritten material,
elevate the symbol. T h e typewriter diagonal should always follow t h e underscore regardless of position of symbol in front or behind the item.
2 A suggested sequence of arbitrary reference symbols is as follows: Asterisk (*), dagger ( t ) , double dagger
( t ) , section mark (§), parallels ([), paragraph mark ( ) . Where more symbols are needed, these may be
doubled in the same sequence: **, tt, §§. Ul, Ii.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

81

PAR.

634

THE FOOTNOTE
B. Reproduction
typewriter

A. In letter-

press

Total

Total
1

1

1, 3 7 6

1, 376

SEC.

6-)

C. Standard
typewriter

Total 1 /
l /1,376

Note: When preparing copy for the printer or reproduction typist, always
insert an inverted caret under each symbol, regardless of its form, and repeat
this on each printer's proof including the final one. Otherwise a numerical
symbol may be confused with a numerical entry and an entry of 3 50,553 may be
printed as 350,553. Insertion of the caret should be a routine part of verification of each footnote reference on each proof.
Examples:

150,553

3/50,553

50,553

5ass3

635. New series for each table.-Each table is assigned its own
independent series of footnote reference numbers, beginning with 1
in each case. When tables are run within text, one series of footnote
symbols is used for the text; another series for the first table; another
for the second table; etc. For this purpose, leader work run in text,
and the text tabulation, are treated as tables. That is, each presentation has its own independent series of reference symbols. (See par.
654 for placement of footnotes when a text table falls at the bottom
of the page and footnotes appear for both table and text.)
636. Series treatment for multipage tables.- In a multi page
table (a table starting on one page and ending on another) two
methods are available for footnote numbering. It is recommended
that footnotes be placed at the bottom of each table-page where the
given reference symbol appears.
Method A: Table-page b a s i s . - B e g i n t h e reference numbering over again with
011 each page of the table.
This demands placement of footnotes at t h e
bottom of the affected table-page; they must not be gathered at t h e end of the
table since there will be as many footnotes for reference number l as there are
table-pages with reference numbers on them. Listing them in page groups at
the end of t h e table, preceded by page numbers, as "Footnotes for page
is not a satisfactory solution.
Method B: Table b a s i s . - N u m b e r the references continuously in a single series
for the entire table, irrespective of page. This permits the recommended practice
of placing the footnotes at the bottom of each affected table-page, or they may
be gathered together at the end of t h e table.

"1"

637. Placing symbols in front or behind the item qualified.-It is
sometimes thought that a reference symbol must always be placed
at the end of a reader entry, stub entry, or column-head entry even
though the qualification applies only to a portion of the entry.
This is not the case; place it against the affected word or phrase.
Any cell entry containing a word or phrase instead of a numerical entry.
Note the consequence if footnote 3 had been placed here (end of sentence) instead of against the affected
words ("reader entry"),

82

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

SEC.

6-D

REFERENCE SYMBOLS

PAR.

637

After placement, be sure to review the effect in terms of footnote
wording. Queries: Is the footnote such that the reader might
erroneously assume that it applies only to the word or phrase immediately preceding the symbol when it is intended to apply to some
other portion of the statement? Is he likely to assume that it applies
to a greater portion of the statement than is intended? If confusion
is likely, remove the danger by rewording the footnote or shifting
the symbol, or both.
a. Stub entries and column h e a d s . - P l a c e the symbol after t h e affected word or
phrase, date, or figure.
Examples:

Number in family

1940 1

Total 1

None

2

1

1

Oil-bearing crops 1 __
Supplies and materials
Housing,a fuel, and light.

b. Cells with numbers (except dates) . - P l a c e t h e symbol in front of t h e number.
This insures t h e vertical line-up of all final digits. It reduces t h e column width
except where the symbol is placed against a total entry. (The total entry usually
constitutes the largest entry and determines the column width unless t h e boxhead wording demands additional space.)
Wrong:

Right:
7,654
1

7.654
125 1
1,354

125

1,354

c. Dates.-Place t h e symbol after t h e date, whether it is in the boxhead, stub,
or field. Exception: Where the last (right-hand) column on a page is a d a t e
column (entries in cells are dates), place the symbol in front of t h e affected d a t e
instead of after it. This does not affect placement in the other columns of such a
table.
In stub: 19401-

In cells, except last column: 1940 1

In c o l u m n h e a d : 1940 1

In cells in last c o l u m n :

1

1940

d. "Blank" c e l l s . - I n a figure column or d a t e column a footnote reference
standing alone is placed in parentheses and centered horizontally in the cell. In
a reading column it is placed at the left in parentheses and is followed by leaders,
unless it is in the last column; in the last column (if a reader column) it is followed
by a period, as if it were a word.

Total

163

1940

Figure columns

Coverage

Dato

61

All workers_--

75

In last
column ( i f .
a reader
column)

Yes.

e. Reader c e l l s . - I n s e r t t h e symbol after t h e affected word or phrase. (A
reader cell is a cell which contains a word or phrase instead of a numerical entry.)
Expected
completion
In

77655449

7


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Aug.!_

Month
started
Feb,

83

PAR.

·

638

THE

FOOTNOTE

SEC.

6-.D

638. Numbering sequence: General.-Whether in text or tables,
footnote references are numbered in sequence reading from left to
right on each line, beginning at the top and running down the tablepage. This principle applies irrespective of continuity of numbering
of ref er enc es between table pages.
Note: In a parallel table, each pair of facing pages of the book is considered a
single table-page, hence references are numbered in sequence across the pair of
book pages.

639. Numbering sequence: Boxheads.-The boxhead presents
special problems in placement sequence of reference symbols. In
theory, the line-for-line rule (par. 638) is exchanged for a "level-forlevel" principle, but its application is not self-evident except in the
simplest cases. The primary problem, caused by the varying levels
and depths of boxes, is "which boxes are to be considered as being on
the same 'line' or 'level'?" No completely satisfactory rule is available. Frequently, common sense is the only recourse.
Four distinct placement methods are described below, each differing
in the criteria employed in determining coordinate level of box.
Method A is recommended and may be expected to cover satisfactorily
the vast majority of cases. The remaining methods are illustrated so
as to provide a description of the several alternatives considered.
For illustrative purposes a fairly complex boxhead is presented (see
fig. 12) with reference symbols in all columns. In practice such a
multiplicity of reference symbols would be objectionable.
Method A. (Recommended.) Number in sequence reading from left to right
on each box-level and from top to bottom within spanners"-TOP LINE-UP.T h e top rule of the given column or spanner head is used as a guide to determine
t h e "level" for each individual box or spanner. For numbering purposes t h e
boxhead is divided into as many vertical segments as there are columns or spanners
rising to t h e top rule of t h e table.
T h e columns appearing under each top spanner are numbered downward with
those on t h e same level (according to top line-up) numbered from left to right
within t h e given spanner group. In t h e example shown in figure 12, three levels
appear, comprised as follows:
First level: Column heads e and j, a n d spanners A and C.
Second level: Column heads f, k, a n d I; and spanners B and D.
Third level: Column heads g, h, m, a n d n.
T h e resultant sequence of numbering throws together t h e footnotes pertaining
to related material. Also, t h e numbering m a y be said to follow t h e customary
p a t h of t h e eye when t h e boxhead is being read.
Method B. (Not recommended.) Number reading from left to right in each
box-level across the entire t a b l e - B O T T O M LINE-UP.-Here t h e bottom rule
of t h e given column or spanner is used as a guide to determine which boxes are
on t h e same level. In t h e example in figure 12 t h e levels are as follows:
For definition of "spanner," see par. 1202c.

84

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

SEC.

6-D

FIGURE

FIG.

REFERENCE SYMBOLS
12.-NUMBERING

S Y M B O L S IN

SEQUENCE

FOR

FOOTNOTE

12

REFERENCE

B O X H E A D S OF S T A T I S T I C A L T A B L E S ( S e e p a r .

639)

Method A. Recommended. Number in sequence reading from left to right on
each box-level and from top to bottom within s p a n n e r s - T O P L I N E - U P
Spanner A a
Spanner B

Col-

Stubheadi

umn

head
e3

Spanner C

Column
head
f4

Column
head

Column
head
h

Spanner

Col

umn

head

Col-

umn

head
k

10

12

Col

umn

head
1 11

7

Col

Col-

umn

umn

head
m 13

head
n

14

Method B. Not recommended.

Number reading from left to right in each boxlevel across the entire t a b l e - B O T T O M L I N E - UP
Spanner A 2

Stubbead

l

Column
head
e6

Spanner C a

Spanner B •
Column
head
f1

Col

umn

bead

Column
bead

Column
head
j

10

Spanner D
Column
head
k

II

Column
head
1 12

Column
head

Column
head

13

14

m

h

t

n

Method C. Not recommended.

Same as method A except t h a t t h e coordinate
level of box is determined by the B O T T O M L I N E - U P of individual boxes
Spanner A a
Col-

Stubhead

1

umn

head
e

3

Spanner C •

Spanner B •
Column
head
fl

Col
umn
head

Col

umn

head

J8

Col-

umn

head
k

head
h

Spanner D

Col

umn

11

Column
head
1 12

Ool

head

Column
bead

13

14

umn
m

7

10

n

Method D. Not recommended.

Same as method B except t h a t the coordinate
level of box is determined by the TOP L I N E - U P of individual boxes
Spanner A a
Spanner B 7

Col-

Stubhead1

umn

head


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Column
head
f6

Column
head

Column
head
h 12

Spanner C
Col-

umn

head

Col
umn
head
k&

Spanner D

10

Column
head
m 13

14

Col-

umn

head

85

Col
umn
head
n

PAR.

639

THE FOOTNOTB

SEC.

6-D

First level: Spanners A and C.

Second level: Spanners B a n d D.
Third level: All the individual column heads irrespective of the height to
which they rise.
This might be considered as a logical adaptation of t h e line-for-line principle.
In practice it tends to be confusing since t h e distinction upon which coordinate
level is based is too subtle for t h e reader.
Method C. (Not recommended.) Same as method A except that the coordinate level of box is determined by the BOTTOM LINE-UP of individual
b o x e s . - I n t h e example shown, t h e sequence tends to be illogical in terms of
relationships.
Method D. (Not recommended.) Same as method B except that the coordinate level of box is determined by the TOP LINE-UP of individual boxes.This p a t t e r n tends to be very confusing to the reader.

640. Footnoting a recurrent reference.-Frequently an identical
situation may arise in more than one place within a table which demands qualification by an identical footnote. The same footnote
reference symbol may be assigned at each appearance. Either a
reference number or an arbitrary symbol such as an asterisk (*) may
be used.*
This device helps the user, who quickly discovers that footnote
reference (3) (for example) always means the same thing. Also, it
reduces the number of footnotes since the recurrent footnote statement is listed only once at the bottom of the affected page, or at the
end of the table if the notes are gathered there. (For placement of
recurren t-footnote statement, see par. 655.)
a. Use of common reference n u m b e r . - I f a common reference number is
assigned, t h e number is determined by t h e position of its first appearance.
Example:
Color:
White.

Nonwhite__

1

629
185

2

352
178

110

4 15

1

Incomplete.

b. Use of a s t e r i s k . - T h e use of a common reference number for a recurrent
footnote is necessarily confined to those cases where t h e reference numbering is
on a table basis (see par. 636). Obviously, the use of a common reference number
throughout t h e entire table is impracticable where numbering is on a table-page
basis; t h a t is, where symbols begin again with 1 on each page of table. Instead,
an arbitrary symbol such as an asterisk (*) may be assigned to the recurrent item.
Footnotes keyed with an arbitrary symbol are listed above those in a numbered
series (see par. 656).
If desired, t h e qualification may be placed in a headnote, such as "[Asterisk (*)
denotes . . . . ]," or "[ . . . . . is denoted by an asterisk (*) J." Use of t h e headnote is not desirable if a common reference numeral is employed.

86

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

8EC.

6-E

RANGE OF SYMBOLS

PAR.

641

Sec. 6-E. Reference Range of Symbols (641-643)
641. General.-Correct placement of the reference symbol is imperative. Serious errors may be caused where the range of influence
of the reference symbol is greater or less than that intended by the
table-designer. Unless otherwise specified in the footnote, the reference range is as indicated below. See also fig. 13, p. 88. (For special
problems relating to total cells, columns, and lines, see pars. 642-3.)
a. The c e l l . - A footnote reference placed within a given cell qualifies or explains
t h a t particular cell entry only. If entries in two cells are to be qualified similarly,
a reference symbol must be placed in each, unless t h e table construction permits
placement of t h e symbol against a stub entry or column head under t h e conditions
outlined below.
b. Column h e a d . - A reference number placed against a column head applies
t h e footnote qualification to all entries in t h a t particular column unless t h e footnote otherwise specifies. It does not apply to adjoining columns. If t h e qualification applies to some, but not all, of t h e entries in t h e given column it is important that this be made clear. T h e same principle applies to t h e spanner head;
t h a t is, a reference placed against it qualifies everything under i t .
c. Stub e n t r y . - S a m e principle as for t h e column head. A reference symbol
placed against a s t u b entry applies t h e qualification to all entries on t h a t particular line unless t h e footnote specifies otherwise.
d. Subhead in stub.-Theoretically, t h e same principle applies as for a spanner
head in the box. However, an important practical difference must be noted. Jn
the box, t h e spanner head is immediately above both the individual column heads
a n d t h e columns of data. Hence it is easy for t h e user, while reading t h e figures,
to note t h e presence of t h e reference symbol against t h e spanner. In t h e stub,
however, the subhead is above t h e individual s t u b entries and, particularly where
t h e listing is long, may be far removed from t h e cells where t h e figures appear.
For t h i s reason, special caution should be observed when keying references to
subheads of t h e stub. T h e least danger is involved when t h e footnote statement
relates exclusively to a classification level in t h e stub, or a characteristic of t h e
subhead itself. The greatest danger arises when t h e reader is expected to note
a n d to remember t h a t t h e qualification affects vitally the cell entries most remote
from it.
Solution: No specific solution can be offered. In general, however, it is well to
explore such possibilities as (1) use of t h e headnote to s t a t e t h e case; (2) placement
of references in column heads; or, in extreme cases, (3) use of spanner heads within
t h e field (see sec. 14-D), even though such heads are, in themselves, undesirable.

642. Footnoting "Total" cells, columns, and lines.-Because of the
difference in viewpoint of the designer and user, confusion may result
from incautious placement of reference symbols against cells, columns,
or lines containing total entries.
a. From designer's viewpoint.-The designer, who has built up t h e statistics,
is likely to think of t h e total entry in its mathematical sense; t h a t is, as t h e summation of its component parts, wherever found in t h e table. Hence, he may feel t h a t
a n y qualification of all component parts is made clear by a single reference number
placed against the total.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

87

FIG.

13

FIGURE

THE FOOTNOTE

13.- R A N G E

SEC.

OF R E F E R E N C E OF F O O T N O T E S Y M B O L S

6-E

(Pars.

641-643)
[ T h e range of reference of footnote symbols is as indicated by the dotted lines]

Example A. Ref ere nee symbols in boxhead:
White a

All classes

Nonwhite

Area and age
Tota'[

Male

2, 385

1,197

1,188

2,065

1,030

646
551

630
558

1, 105

556
474

549 }
486 i

1,867 :

935

932

1, 642

816

993
874

499
436

494
438

872
770

435
381

Female : T o t a l

Male

Femalei T o t a l

Male

Female

UNITED STATES
All ages__
U n d e r 45 years__-45 years a n d over_--

1,276 :
1, 109

320

167

153

171
149

90
77

81
72

826 {

225

119

106

437
389 }

121
104

64
555}

57
49
47

1,00s

URBAN
All ages_.U n d e r 45 years ___
45 years and over_-RURAL
All ages_
Under 45 years__45 years a n d o v e r - -

518

262

256

423

214

209

95

48

283 }
235

147
115

136
120

233
190

121

112
97

50
45

26

24

22

23

1 Column bead, see par. 641b.
2 Spanner bead, see par. 641b.
• Cell, see par. 641a.

Example B. Reference symbols in stub:
White

All classes

Nonwhite

Area a n d age
Total

Male

Female

Total

Male

2 ,385

1,197

1,188

2,065

1,030

Female T o t a l

Male

Female

UNITED STATES
All ages __

{Under 45years i
455 e a 7 s

anover.

1,035

556 I
474

1,1o0

646 I
551

558

1,867

935

932

1,642

s16

826

874

499
436

494
438

872
770

435
381

437
389

320

167

i7iI
i49

77

153

si I
72

URBAN

I All

ages .

U n d e r 45 years___
45 years a n d o v e r - -

All ages__
: Under 45 years_-45 years a n d over_--

!


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

121

57
49

104

518

262

256

423

214

209

95

48

47

283
235

147
115

136
120

233
190

121
93

112
97

50
45

26
22

24
23 {

1 S t u b e n t r y , see par. 641c.
2 Total line, see par. 643.
• Cell, see par. 641a.
4 Subhead in s t u b , see par. 641d.

88

io6

SEC.

6-E

PAR.

RANGE OF SYMBOLS

642

b. From user's viewpoint.-The user, who finds t h e mass of d a t a neatly
arranged in front of him, is far more likely to interest himself in a particular item
of detail. He expects to find qualifications indicated in a footnote keyed closely
to his particular item, if specific to t h a t item or class. In terms of footnote ref erence he is likely to think of t h e total in mechanical terms; t h a t is, as just another
cell, column, or line.
c. Solution.-Obviously both viewpoints cannot be completely satisfied. But
the designer already knows t h e d a t a ; t h e user does not. T h e purpose of table
design is to make t h e material available to t h e user. Hence, t h e user's viewpoint must be given t h e preference.
Rule 1: Where possible, restrict placement of reference symbols against a
total cell, column, or line to those cases where t h e footnote is intended to cover
only t h a t specific area. Example: A footnote explaining t h a t t h e total includes
figures for a subclass not shown separately.
Rule 2: Where a qualification extends beyond the usual range of the reference
symbol, be sure to specify accordingly in the footnote. This applies with added
force where symbols are placed against totals.

643. Effect of substituting classification term for "Total." Frequently, in order to save space or to eliminate the need for centerheads in the stub, a classification term is substituted for the word
"Total" on the total line; that isInstead of this expanded form

This simplified form may be used

URBAN
Total

Urban __

Under 45 years_.
45 years and over_

Under 45 years __
45 years and over_-

This practice is perfectly sound but questions as to cJarity of reference range may arise if it becomes necessary to qualify (a) the classification term as such; (b) the entire block of data with respect to an
inclusion or exclusion affecting all figures in the block (especially
where the group included or excluded is not of a class listed separately in box or stub); or (c) the total line only.
For illustrative purposes assume the following footnotes as representative of these three types of qualifications, and that there are
sound reasons why the first is not placed in a headnote.
1 Outside metropolitan districts of 100,000 inhabitants or more.
2 Excludes persons of unknown marital status.
a Includes persons of unknown age, not shown separately.

a. Expanded form of s t u b . - H e r e the problem is fairly simple since there are
two separate pegs on which to hang t h e reference symbols.
Example A-1:
URBAN 1
Total Under 45 years__
45 years and over-

(Employment status)
_ 1,000

40

16

750

32

9
5

1 Outside metropolitan districts1 of 100,000 inhabitants or more.
Excludes persons of unknown marital status.
2 Includes persons of unknown age, not shown separately.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

89

PAR.

643

THE FOOTNOTE

SEO.

6-E

b. Simplified f o r m . - T h i s may be confusing.
(Employment status)

Example A-2:
Urban !Under 45 years_.
45 years and over_-

_ 1,000

40

16

750
200

32
6

9

5

1 Outside metropolitan districts of 100,000 inhabitants or more.
Excludes persons of unknown marital status. Includes persons of
unknown age, not shown separately.

In example A-2, how is t h e reader to know what portions of t h e footnote
relate to all figures in t h e block and what portions relate to only t h e total line?
Rephrasing would help in this particular case, but t h e basic problem remains.
Note also t h a t in this simple example t h e reader can tell at a glance t h a t t h e
two age groups do not equal t h e total. However, if the stub classification were
m a n y lines long, perhaps extending over to another page, this would not be
practicable.
c. S o l u t i o n . - I n t h e above case, t h e obvious solution is to change back to t h e
expanded form of stub a n d to explore t h e possibility of placing one or both of t h e
first two sentences in a headnote. Each instance must be analyzed on its merits.
Rule 1: Avoid this situation by adjustment of s t u b construction where
possible, a n d by making use of headnotes to relieve t h e burden.
Rule 2: Expand t h e footnote statements to make clear t h e range of
reference in each case. (See par. 663.)

Sec. 6 - F . Placement and Arrangement of Footnotes (651-656)

651. General.-Footnotes are placed at the bottom of the tablepage or at the end of the table.
a. One-page table.-Place footnotes at end of table. If more t h a n one table
(or portion thereof) appears on t h e same page, arrange their respective
footnotes under each table; do not gather them at t h e bottom of t h e page.
b. Multipage t a b l e . - F o o t n o t e s always should be placed at the bottom of t h e
page ori which t h e reference appears. If t h e footnotes are numbered on a table
basis (a continuous series for t h e entire table) t h e y may be gathered at t h e bottom
of t h e last page but this is not recommended because they then do not appear
where needed.

652. Cross-reference to footnotes placed at end of table. Where
footnotes are gathered at the end of the table, the statement "See
footnotes at end of table" must appear at the bottom of each tablepage containing one or more footnote reference symbols. This
statement is unique in that it is never keyed by use of a symbol nor
is it preceded by the word "Note."
Example:
Color:
White_Nonwhite __

1

645
152

318

2 51

336
161

See footnotes at end of table.

653. Arrangement of notes.-Footnotes appear in order of reference
number. To facilitate reference, each is usually placed on a separate
line. Where a number of short footnotes appear, they may be
90

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

SEC.

6-F

PLACEMENT AND ARRANGEMENT

PAR.

653

arranged across the page. Bureau of the Census practice is to arrange
them in order of reference number, reading from left to right on each
line; they are not ordinarily arranged in columnar form with reference
numbers reading down in each column.
a. Brief notes:
Example A-1. Normal:
1 For the original registration States.
2 For the death-registration States of 1920.
a Values for Negroes prior to 1929 are not included.
Example A-2. Permissible:
1 No data.
# Automotive starter motors only.
a Includes parts and supplies.
' Revised.
Example A - 3 . Wrong sequence:
1 No data.
3 Includes parts and supplies.
Automotive starter motors only.
Revised.
Example A-4. Not recommended:
a Data incomplete.
1 Discontinued.
2 Ginnings.
' N o t available for publication.

b. Long notes involving overruns:
Example B-1. Normal:
1 Based on acreage that works were capable of supplying with water.
2 Based on acreage assessed by enterprises serving 5 units or more.
a Relates to enterprises reporting no irrigation in 1939 but having works capable of supplying
water in 1940.
Example B-2. Permissible:
1 Based on acreage that works were capable of supplying with water.
Based on acreage
assessed by enterprises serving 5 units or more. s Relates to enterprises reporting no irrigation
in 1939 but having works capable of supplying water in 1940.

654. Arrangement where text tables fall at bottom of p a g e . - I n
text material containing tabular data, the tabular presentation may fall
at the bottom of the text column or page. If both text and table
include footnotes, a problem of arrangement arises since tabular
presentations in text are not footnoted in the same series of reference
numbers as the text footnotes.
Solution: The normal solution 1 is illustrated here where the following table is assumed to be found at the bottom of a page (or column)
of text.
TABLE

A. - D I S T R I B U T I O N

OF

"X"

BY

"Y"

Classes of " Y "
Classes of "X"

Total
2

A1

110
165

B

22
33

3
22
33

1

Tabular notes appear first, placed in normal position under the table.

1

Text notes appear below, separated from tabular notes by a cut-off rule.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

22

22
33

33

91

PAR.

655

THE FOOTNOTE

SEC.

6-F

655. Arrangement for recurrent footnote.-Two questions arise:
a. Frequency of appearance.-Even though t h e recurrent symbol appears more
t h a n once on a given table-page, t h e footnote statement must appear only once
at t h e bottom of that page. Thus, in the following example, t h e footnote statement for t h e recurrent symbol (1) appears only once, even though it appears
twice on t h e table-page.
Example A:
Color:
White_--

. I

Nonwhite__1 Incomplete.

352
78

629

185

1

J

110

a 15

a

b. On continued p a g e s . - I n a multipage table where the footnotes are numbered
consecutively throughout t h e table (table basis), and the footnotes appear at t h e
bottom of the affected table-page; t h e recurrent footnote statement must appear
on every page carrying the recurrent symbol, not merely on t h e first page.
Example B (Represents p. 2 of 2 pp. of the table represented as example A):
Urban_-Rural-nonfarm_Rural-farm __
1

4

520
241
181

318
109
98

l

62

Incomplete.

4

656. Placement of footnotes bearing arbitrary symbols such as an
asterisk (*).-Where both arbitrary symbols and numbered references
appear, the footnotes for the arbitrary symbols are listed first.
Example:
White_Nonwhite_• Not reported.
1 Incomplete.

1

624

3

412

4

152

369

158

2

Sec. 6-G. Wording of the Footnote (661-665)
661. Need for brevity.-Since footnotes are hard to read, they
should be as brief as possible, but clarity must never be sacrificed for
brevity. The following factors should be kept in mind:
a. T h e smallness of type and great length of line make footnotes hard to read;
andb. Footnotes tend to be parenthetical in nature with t h e subject-noun frequently
omitted. This makes them hard to comprehend.

662. Style.-Footnotes are written in telegraphic style. Even
though not a complete sentence, the footnote always ends in a period.
Articles and other parts of speech not essential to understanding are
commonly omitted. Standard conventionalized phrasing should be
used wherever possible. Use the same abbreviations and phrasings
in the footnote as in the remainder of the table.
663. Frame of reference.-Footnotes can be shortened, and phrasing simplified, by taking into account the basic frame of reference

92

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

SEC.

6-G

WORDING

PAR.

663

established by the "keying" of the note to the material qualified.
T h e footnote is an integral part of the statistics, or class description,
qualified and should be handled accordingly. Superfluous identifying
phrases and words should be omitted.
a. Basic reference is to material identified by s y m b o l . - A footnote statement,
by definition, refers solely to t h e information to which it is" keyed" by the reference
symbol. (See pars. 632-634.)
(1) Intended coverage same as normal reference r a n g e . - T h e r e is no
need to specify " T h e s e statistics" or " T h i s item" or to describe or name t h e
data or entries qualified. Furthermore, note that if the same type of qualification (such as "Incomplete") is to appear for quite different groups, unnecessarily exact specification of the category covered would make it impossible to
use a recurrent symbol and footnote.
Example A:

White

1 42
Female_(Terse)
1 Includes _
(Verbose) 1 Figures for white female include_-

Example B:
Operatives 1 _ (Terse)
1 Includes_-(Verbose) 1 These statistics include_-1 Comprises _
(Terse)
(Verbose) 1 The category "Operatives" comprises__

(2) Intended coverage LESS t h a n normal reference r a n g e . - D e m a n d s
careful specification of categories or items qualified when intended reference
range is less t h a n normal.
Example C (Assume qualification intended only for 1930 figures):
1940
Operatives 1_
(Right) 1 Figures for 1930 exclude_(Wrong) 1 Exclude_.

75

1930

1940

1930

15

45

30

(3) Intended coverage MORE than normal reference r a n g e . - T h i s
practice is NOT R E C O M M E N D E D . When it occurs, however, it is imperative t h a t the footnote provide clear warning.
b. Order of wording.-As in the instance of t h e headnote, a footnote should
start with what the reader can see, or is likely to assume, and move to t h a t which
is not apparent or not known to him. (See par. 521 b.)
Example:

Cotton manufactures 1__
Silk and rayon manufactures 1__
Accurate statement: 1 Because of a tabulating error discovered too late for correction, the figures for
"Cotton manufactures" include about 2,200 employed females who should have
been tabulated as in "Silk and rayon manufactures."
1 Figures for "Cotton manufactures" include about 2,200 employed females who
But this is better:
should have been tabulated as in "Silk and rayon manufactures"; tabulation error
discovered too late for correction.

664. Numbers and fractions.-In footnotes, numbers are expressed
in figures, even at the beginning of a note or sentence. Fractions
standing alone are spelled out at the beginning of footnotes.
665. Introduction required to tabular footnotes.-If the footnote
consists entirely or partly of a table or leader work, the footnote table
is indented on the left. It must be preceded by introductory matter
carrying the reference symbol. If no other introductory matter
seems required, insert an introductory line, such as "1 See the following
table;",


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Chapter 7

ALLOCATION TO STUB AND BOXHEAD (701-713)
Sec. 7-A. General (701-703)
701. General relationship.-The relationship between stub and
box may be stated in general terms as follows:
a. Identity of basic functions.-The purpose of both t h e box and t h e s t u b is to
describe the specific d a t a and to make clear t h e relationship among t h e d a t a .
b. Basis of allocation of i t e m s . - T h i s depends largely upon several factors which
vary in influence from table to table. T h e final allocation usually represents a
compromise. (The first two factors listed are discussed briefly here. T h e
second two are discussed in detail in sec. 7-B.)
(1) Table purpose or presentation intent.
(2) Relationship between tables in a series.
(3) Comparative mechanical advantages and limitations on ability of each
to indicate intra-table relationships.
(4) Comparative space requirements.

702. Influence of table purpose.-The primary influence of table
purpose is manifested by the need for juxtaposition of certain data.
Thus, if a fundamental condition of presentation is to make possible
a comparison of frequencies of classes A and B in terms of variable
C, it is apparent that both A and B should appear on the same axis,
and that C should appear on the other axis.
Theoretically, it may be argued that it is easier to compare data
by having them appear directly under one another in successive lines,
rather than side-by-side in adjacent columns. Practically, the
question tends to be academic. In only a few instances is it possible
to place all of the important comparisons on successive lines or in
successive columns. The more important practical consideration is
"In which axis will it be most feasible to present the two (or more)
classifications which must appear on the same axis?"
Similarly, when designing tables of the Bureau of the Census, the
question of relative prominence provided by the stub and boxhead
tends to be academic. Again, the problem of making maximum use
of all available space takes precedence.
703. Relationship between tables in a series.-Frequently, series
of tables are shown where all classifications except one remain constant.

94

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

•

sEC.

7-A

GENERAL

PAR.

703

Here it is a decided convenience to the user if a constant pattern in
content of box and stub is maintained from table to table.
The question as to where the variable (the classification that is new
to each table) should be placed is again largely academic. No matter
where it is, if it is properly handled, the user will spot the point of
difference readily in a series where the table pattern is otherwise held
constant. Shifting of the common classifications from one axis to
the other, table to table, will do more to obscure the location of the
classification that is new to each table than will its constant appearance in either the box or the stub, whichever may be argued as giving
it greater prominence.
Sec. 7-B. Allocation Principles and Problems (711-713)

711. Summary of allocation principles.-Certain specific considerations should enter into the question of allocating classifications to
boxhead or stub. The summary listings given here should not be
thought of as independent of one another. Each must be qualified in
terms of the particular situation. However, all other things being
equal, the following statements are generally true.
a. Boxhead presentation is more efficient for(1) Classifications confined to comparatively few items.
(2) Classifications with brief captions.
(3) Constant subclassifications comprised of one, two, or three items, such as
total, male, a n d female, repeated for various classes of data.
(4) Classifications involving possible misunderstanding as to unit of presentation, since a spanner can be used to make t h e unit clear.
(5) Classifications which, to be understood, must be read carefully in terms
of superior heads which can be presented in spanners immediately above.
b. Stub presentation is more efficient for(1) Classifications involving numerous lists of items.
(2) Classifications requiring long and detailed captions.
(3) Constant subclassifications under different heads, involving a large
number of entries on each occasion.
(4) Multiple subclassifications so numerous as to pass quickly beyond page
limits for columns.
(5) Classifications requiring more t h a n three levels of boxhead to present.
c. Standard boxes and stubs advantageous.-In a series of tables, t h e advantages of standard boxes, stubs, or both, outweigh many minor sacrifices in presentation practice. (See par. 703.)

712. Comparative mechanical advantages in emphasizing relationships.-The boxhead, as compared with the stub, has one great
advantage; that is, its greater ability to emphasize relationships
among data, particularly those in adjoining columns. The stub, in
contrast, is better equipped to handle multiple levels of classification.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

95

PAR.

712

ALLOCATION TO STUB

A N D BOXHEAD

SEC.

7-B

a. The b o x h e a d . - I n the boxhead, heads of superior level are- shown in t h e
form of spanner heads. These usually are located directly above t h e affected
column heads. Even where several levels of spanners are employed, t h e individual
column heads are very close to t h e uppermost spanner affecting them. This, in
itself, is a signal advantage in terms of clarity of relationship.
Furthermore, the uppermost spanners divide t h e boxhead into a relatively few,
and easily noted, panels. T h e major classifications thus shown are readily noted
by the user.
However, these "advantages" rapidly disappear as the boxhead becomes more
complex either in number of levels or in length of captions.
b. The s t u b . - T h e s t u b excels where t h e boxhead is weakest; t h a t is, in showing
multiple levels of classification. T h e line caption, itself, constitutes one level.
Indent patterns and t h e use of the total-line caption for classification supply
several more levels. Colon lines, centered subheads, and center heads add to
t h e possibilities. A situation too complicated for boxhead presentation can
be handled easily in t h e stub. In addition, captions which would take up several
lines in t h e box m a y occupy only one line in the stub. Overruns in t h e stub
are easier to read t h a n in the box.
c. Conclusion.-Everything else being equal, place involved classifications in
t h e boxhead if they can be presented in not more t h a n three levels of box. This
applies particularly to instances where proper understanding of subclassifications
depends upon keeping the major and minor heads well in mind at every point,
and the caption length and box arrangement are such as to permit ease of reading.
Place in t h e stub, however, classifications involving many levels and sublevels.
T h e boxhead becomes forbidding in aspect when more t h a n three levels are shown.

713. Comparative space requirements.-This question divides itself
into two parts.
a. Fewer columns than lines per page, irrespective of caption length.Obviously, fewer columns t h a n lines are possible on an upright page since a
column occupies more space t h a n a line.
(1) T h e b o x h e a d . - P r a c t i c a l l y , t h e box can rarely carry more t h a n 15 to 17
separate columns on a large size 1 (such as census-size) page of frequency d a t a ,
or 20 to 25 columns of percent data. To accomplish this, the column captions
must be such as to appear to advantage in extremely narrow boxes. The
number of columns on a smaller ! page (such as document-size) will be correspondingly less.
(2) T h e s t u b . - I n contrast, t h e stub will carry from 75 to 90 line captions
on a large-size page, with the total differing in terms of t h e number of centered
heads, space and analytical breaks, overruns, and space required by title,
headnote, and boxhead. The document-size page will carry s t i l l fewer line
captions.
(3) Conclusion.-Everything else being equal, t h e classifications with t h e
fewest number of intervals should be placed in the box. Classifications involving numerous listings should appear in the stub. Long stubs are preferable
to long boxes. Continued stubs are easier to follow and to refer to t h a n continued boxes.
1 See

par. 211, for comparative page sizes.

96

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

SEC.

7-B

PRINCIPLES A N D PROBLEMS

PAR.

713

b. Box demands briefer captions.-By its nature, t h e box cannot handle long
detailed individual captions. The stub, with its greater flexibility as to line
length, is better adapted for this purpose.
(1) T h e b o x h e a d . - W i d t h of individual column heads should be held, where
possible, to two spaces more t h a n required by the maximum cell frequency
which will appear. This conserves horizontal space and insures more columns
per page.
In addition, where possible, no lower box should be run more t h a n three or
four lines deep, at most. Otherwise, the caption tends to be extremely hard
to read.
(2) T h e s t u b . - A l t h o u g h brevity is essential in stub captions, t h e same
restrictions do not apply as for boxhead captions. True, overrun lines are not
desirable, particularly lines overrunning to make more t h a n two or three lines.
However, in the s t u b the restrictions m a y be set aside freely as needed without
too great an impairment of readability.
(3) C o n c l u s i o n . - E v e r y t h i n g else being equal, assign to t h e box those
classifications with briefer captions. Assign to the s t u b those requiring long
captions.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

97

Chapter 8

THE STUB: GENERAL AND DEFINITIONS (801-835)
Sec. 8-A. General (801-803)

801. Organization of discussion of s t u b . - I n this manual the discussion of the stub is distributed among four chapters. The present
chapter defines the stub as a whole and each of its component parts.
In general, each definition is accompanied by a highly condensed
description of the material involved.
Chapters 9 through 11 discuss the various component parts in
considerable detail. Specifically, chapter 9 describes such classifying
and descriptive devices as the center heads, line captions, and unitindicators. Chapter 10 is devoted to construction aids to clarity
such as indention, leaders, and space breaks. Chapter 11 discusses the
standardized arrangement of items for various subjects; arrangements
which, in some cases, are based on scientific evaluation and, in other
cases, rest upon custom.
Two other chapters have a direct bearing on stub treatment; that
is, chapters 7 and 13. Chapter 7 discusses the relationship between
the stub· and boxhead. Chapter 13 describes the use of horizontal
and vertical rules. In respect to horizontal rules, it must be remembered that, although they do not normally extend through the stub,
their presence or absence frequently is crucial to clarification of stub
listings.
802. Definition.-The stub is that portion of the table, usually
comprising the first column on the left, in which is located the listing
of line or row captions or descriptions, together with needed classifying
and qualifying center heads and subheads. For descriptive purposes, the column head (or caption) of the stub is considered a part
of the stub of the table, although in terms of location and physical
requirements it is a part of the boxhead.
803. Function of component parts.-In terms of construction the
stub is made up of a number of easily identifiable component parts.
Those_most frequently used are listed and defined briefly immediately
below. In general, the fewer the number of parts used, the greater
the'simplicity of the stub. Note, however, that these parts are
devices designed to facilitate reader-understanding. No part should

98

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

SEC.

8-A

PAR.

GENERAL

803

be used which will operate against this requirement. A part should
be used without hesitation if clarity will thereby be increased.
Sec. 8-B. Definitions of Component Parts (811-821)

811. Stubhead or b o x . - T h e column h e a d or caption of the stub
which describes the stub listing as a whole in terms of the classification
presented. (See sec. 9-A.)
Examples:
Division and State
Census year
Type of enterprise
Tenure and race of operator

812. Center head and s u b h e a d . - A classifying, descriptive, or
qualifying statement applying to all subheads and line captions following it until the next center head or subhead is reached which
has a coordinate or superior classification. (See sec. 9-B.)
Examples:
Cotton _
Rayon__

TEXTILES

AGE

Under 45 years_45 years and over __

COUNTIES
Appling
Baker

PERCENT
.
.

January

.

February

.

813. Colon and dash (read-in) l i n e s . - S t u b entries (other than
centered heads and subheads) on lines which contain no data or other
entries within the field of the table. They are subject to the same
rules of indent as are the line captions on data lines. Actually, the
colon line is a heading; the dash line is a fragment, not a line in itself.
They are discussed together because the one is so frequently confused
with the other in table design. (See sec. 9 - C . )
a. Colon l i n e . - A heading within t h e detail of t h e stub. Ending in a colon(:),
it usually (1) indicates t h e nature of the classification presented in its subentries
(lines indented below it) ; (2) indicates the group covered when the terminology
of the subentries relates to a subordinate classification only; or (3) maintains the
identity of a subgroup for which d a t a are presented for t h e component parts but
not for the group total. It does not "read in," as such. (See pars. 922-925.)
Example:
Unglazed structural tile:
Nonload bearing_Load-bearing, back-up tile_-

2, 654
_ 1,827

b. Dash (read-in) line.-Comprises an opening word or phrase which, common
to several consecutive lines, has been removed and placed on a line by itself and
terminated with a dash ( - ) . The dash ( - ) indicates t h a t it is an integral part
of each of its subentries and is to "read in" to each of them; t h a t is, it is a
fragment, not a complete caption in itself. (See pars. 926-933.)
Example;
Produced at refineries fromDomestic petroleum __
Foreign petroleum_-

776554

49

8


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

645
129

99

PAR.

814

THE

STUB : GENERAL A N D DEFINITIONS

sE0.

8-B

814. Line caption: Data line. Basic unit of stub. A descriptive
title of the data appearing on the given line. A data line is any line
containing any cell entry whatsoever, even though the sole cell entry
be leaders, a footnote reference symbol, or any symbol which has a
standard meaning or is explained in the headnote. The data-line
caption is always followed by leaders in the stub. (See sec. 9 - D . )
Examples:
White_Negro ___
Other races-

Urban_
Rural-nonfarm_Rural-farm_-

467

31

547
362

815. Line caption: Total or subtotal l i n e s . - A descriptive title of a
data line comprised of grand totals, group totals, or subtotals; a
special case of the line entry for the data line. (See sec. 9-E.) It
may specify the unit of presentation where necessary. Where it is
obvious that the line is a total line, the word "Total" may be omitted
from the line caption and the name of the area or major group inserted
in its place, as in the following example. (Also, see par. 956b.)
This tends to reduce the number of center heads or subheads in the
stub. (See par. 956f.)
Examples:
Total_-

l 000

Urban_Rural..

750
250

The State __

(total lines)
(other data lines)
(other data lines)

-1 000

Urban __
Rural_

750
250

816. Unit-indicator.-A device for indicating the presentation unit
for a specific line. The unit-indicator is used when the unit varies
from line to line and is not made clear by the normal wording of the
line caption. The unit-indicator is placed on the right-hand side of
the stub. On occasion, it may be used to specify the effective timeperiod or other varying factor. (See sec. 9-F.)
Examples:
Letterpress
1929
__1939 __
- -1929-- 1939--

Offset
--1,000 gals_--1,000 lbs_--1,000 oz
_$1,000--

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

..
..
..
..

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

..
..
..
..

. 1 , 0 0 0 g a l s ..
. . l , 0 0 0 l b s ..
. . . l , 0 0 0 oz ..
. . . . .$1,000..

N O T E . - I n letterpress, unit-indicators are lined up at the right; in offset, it is permissible to line them
up at the left. (See par. 963a.)

817. Stub leaders.-In letterpress, a row of small and closely set
"dashes," or in typewriter offset, a row of typewriter periods, extending from the last letter of a stub entry to the column rule at the
right of the stub, lining up at the bottom of the given line. The
purpose of leaders is to maintain horizontal legibility by helping the
eye maintain the same level between the end of the stub entry and
the point where the figures begin. (See sec. 10-B.)
Examples:
Letterpress_
654
Typewriter o f f s e t . . . . . . . . . . . . . . . . . . . . . . . . . 654

100

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

329

152

329

152

87

87

39

39

SEC.

8-B

DEFINITIONS OF COMPONENT PARTS

PAR.

818

818. Space break.-An open space or blank "line" inserted at
periodic intervals to break up solid blocks of entry lines. It improves
horizontal legibility by grouping the entries. The space break extends the full width of the table and thereby provides "guide '' lines
across the page. (See sec. 10-C.)
Examples:
A. 10-line break
Alabama_Arizona_-Arkansas.
California_-Colorado_-Connecticut..
Delaware_District of Columbia_-Florida_Georgia__
Idaho_-IllinoisIndiana_-Iowa.--

B. 6-line break
1940--1930-1920--

1910.
1900

1890--1880
1870--1860
1850
1840
18301820

819. Continued l i n e . - A line (or lines) appearing at the top of continued columns of the stub which indicates the nature of the continued
classification (or classifications) carried over from the preceding
column. Each line carried over is given the same indent as on the page
where it originally appeared.
Examples:
GROUP O.-ANIMALS, ETC., INEDIBLE-Continued
Patent upper leather:
Patent side upper leather_Goat and kid_Other. . .
Fruits and vegetables-Continued.
Salad dressings-Continued.
Fruits other, grapes:
Liquors, distilled.Wines ..

820. Tracer numbers.-A guide number placed at the beginning,
and again at the end, of each data line. In Bureau of the Census
practice they appear only on parallel tables where the stub is not repeated on the right-hand (facing) page. Tracer numbers are placed
at the left of the stub on the left-hand page and to the right of the last
column on the right-hand page. They are set off from the rest of the
table by ver.tical rules. (See example at end of par. 821, and fig. 9,
example 9-A, pp. 22-23.)
Tracer numbers aid horizontal legibility by identifying that portion
of the line which appears on the right-hand page (which has no stub)
with the portion of the same line.which, accompanied by the line
caption, appears on the left-hand page. The presence of tracer
numbers reduces the danger of confusion caused by possible bad
line-up of the two pages when printed. The series of tracer numbers


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

101

PAR.

820

THE STUB: GENERAL A N D DEFINITIONS

sEC.

8-B

is continuous down the page but starts again with number 1 at the
top of each pair of pages. This device is a function of mechanics
and not of content. (Compare with par. 821, "Group or line numhers," below.)
821. Group or line numbers.-A series of numbers or symbols,
usually presented at the left of, but sometimes within, the stub, which
identify categories or subcategories in terms of a particular classification scheme or code. A function of content and not of mechanics,
they are often confused with tracer numbers, described above in
par. 820.
In. a given table the group or line numbers may or may not start
with 1 and they may not necessarily run consecutively in a continuous series. Each caption or heading assigned such a number
always carries the same number whether the listing is complete or
incomplete, expanded or condensed; or whether it appears on an
office worksheet or a machine sheet or as a component of a formal
printed presentation. Hence, in the printed report there is no
mechanical repetition of the series with each new page. The same
numbers may repeat within a given page if the categories which they
identify are presented in repeated blocks, or the numbers may run
in a series extending over many pages if there is no repetition of a
given category.
Group or line numbers are not confined to parallel tables. When
they appear in parallel tables, they may or may not appear on the
right-hand page; if they do appear on the right-hand page, they may
be inserted. either on the left-hand or the right-hand side of that page.
(In contrast, note that tracer numbers always appear on the righthand side of the right-hand page of parallel tables.)
The example given below, illustrating both tracer and group
numbers, comprises the first six lines of the stub of a parallel table.
(See Sixteenth Census, Manufactures, 1939, vol. I, p. 22, table 5,
for complete table.)
Group
No.

Industry group and industry

All industries, total_.

1

GROUP

2
3

1
2
3

6

Food and kindred products_Tobacco manufactures.
Textile-mill products and other fiber manufactures _
Apparel and other finished products made from fabrics and similar materials_Lumber and timber basic products __

102

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEC.

8-C

STANDARD STUB

ARRANGEMENTS

PAR.

831

Sec. 8-C. Standard Stub Arrangements (831-835)

831. S u m m a r y . - T h e arrangement of materials in the stub depends
upon (a) the nature of the classification and the purpose of presentation; and (b) the limitations of space, typography, and typing.
(See secs. 11-A and 11-B.)
Stub preparation of tables of the Bureau of the Census must first
take into account the high degree of standardization of presentation
developed for many subjects over a long period of years. However,
in a report dealing with a particular subject the normal rule is to
place first the data comprising the subject of the report, even though
in a standard listing that factor may be subordinate.
The following general rules on stub arrangement therefore must be
qualified in terms of (a) the standardized practice, if any, for the
subject covered, or of the Subject Division 1 involved (within the
Bureau of the Census); and (b) the specialized purpose of the given
presentation.
832. Areas.-The following statements are general only. For
detailed discussion, see section 11-A.
a. Regions and divisions.-Usually presented geographically.
b. States.-Geographically under region and division headings if t h e data
display marked regional differences. Alphabetically if t h e geographic factor is
not significant in analysis.
c. Smaller a r e a s . - U s u a l l y alphabetically within States. However, at times,
places may be arranged in groups classified according to size, listed alphabetically
within each separate size-group. Example: An alphabetical listing of cities under
t h e heading: "Cities of 25,000 Inhabitants or More." Another common practice
is to list cities, towns, etc., alphabetically within counties; with t h e counties, in
turn, alphabetically within States.

833. Time s e r i e s . - T h e most recent date may be shown either
first or last, depending upon the presentation purpose and the nature
of the material. (See sec. 11-B. See also pars. 365-366.)
a. Decennial or quinquennial census r e p o r t s . - S h o w t h e most recent date first
with remainder of listing receding in time (down the stub or across to the right).
b. Future t i m e . - S t a r t with the earliest time and carry through t h e present
into the future.
c. Historical development or survey t a b l e s . - N o r m a l l y , start with the earliest
date and lead to t h e present. T h a t is, place t h e earliest date at t h e top of the
list or in t h e first column of t h e table.
d. Current continuing s u r v e y s . - N o r m a l l y , start with the earliest date and
lead to the present.
1 That is, the Agriculture Division, Business Division, Foreign Trade Division, etc., each of which is
a Division of the Bureau of the Census in charge of censuses, statistics, etc., relating to its assigned subject field.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

103

PAR.

834

THE STUB: GENERAL AND DEFINITIONS

SEC.

8-C

834. Quantitative classifications.-May be arranged according to
physical properties, such as size, value, weight, or number, in increasing or decreasing order according to presentation purpose and the
nature of the data. In general, the practice of the Bureau of the
Census tends toward arrangement in increasing order of size, value,
etc.
835. Qualitative classifications.-The many qualitative classifications employed by the Bureau of the Census are usually subject to
standard methods of listing which have been worked out for each
field or subfield. Such standard listings are usually thought of in
terms of scores, or even hundreds, of lines. However, even simple
two-line classifications are ordinarily subject to customary order in
listing. The user is accustomed to this order, and it is reflected in
worksheets and machine sheets of all kinds. No departure should
be made from it without ample and demonstrable cause.
It is impossible to list these materials here. However, two examples
may be given to illustrate the point that even simple listings may
have their standard arrangement. Thus, age data are almost invariably arranged from the youngest to the oldest; listings for race are
normally arranged, "Native white, Foreign-born white, Negro," and
"Other races," in that order.
Many detailed stubs have been developed for classifications in
varying detail which should be followed rigorously as to order of
listing and terminology, spelling, punctuation, abbreviations, etc. A
few of the fields thus covered are "Kind of business," "Occupation,"
"Industry," "Commodities," and "Crops." In many of these fields
various standard stubs are existent to serve various requirements of
analytical content and mechanical spacing. Standard listings may
run from two lines to many pages.

104


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Chapter 9
THE STUB: HEADS AND CAPTIONS (901-966)
Sec. 9-A. The Stubhead or Box (901-909)
901. Definition and purpose.-The column head or caption of the
stub which describes the stub listing as a whole in terms of the classification represented by the stub entries. I t s function is threefold,·
as follows:
a. To describe t h e s t u b listing in terms of classification and subject.
b. To serve as a reference guide. Because of this function, it should be kept
as brief as possible to permit t h e reader to note, at a glance,(1) The principal content of t h e stub.
(2) T h e allocation of entries to t h e s t u b as compared with t h e boxhead.
c. On occasion, to make clear by amplification a classification whose description
has been overcondensed in the title and stub.

902. Coverage description.-In general, only major classifications
should be listed in the stubhead. Too great detail in indication of
cross-classification should be a voided.
a. Indicating major classifications.-By "major classifications" is meant those
of fundamental importance in terms of table-purpose, irrespective of heading or
indent levels. Thus, in a table designed primarily to present "age in single years"
by race, sex, and urban-rural residence, the age classification (if presented in t h e
stub) would be most important in terms of presentation objective, even though
the age entries may be listed under various subheads.
b. Indicating cross-classification.-In general, indication of cross-classification
should be avoided in t h e s t u b heading, except where it can be stated simply or
where it is peculiar to t h e specific problem involved. Thus, if a s t u b shows
listings for age, race, and sex, with each crossed with the other two, it usually
is sufficient to label it in t h e stubhead as "Age, race, and sex," rather t h a n to
attempt to spell it out as "Age by race and sex," or "Age, by race and by sex."
A useful comparison may be made between treatment for the stubhead and t h e
table title. T h e table title not only appears directly above the materials which
it describes; it may also be listed elsewhere, as in the table of contents of the
report. This places a greater premium upon exact specification in the title t h a n
in t h e stubhead since t h e stubhead never appears except immediately above t h e
listings it describes. In spite of this, it is recommended t h a t even t h e table title
should not include too highly involved descriptions of cross-classification. (See
par. 306.) This applies to t h e stubhead with added force.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

105

PAR.

903

THE STUB: HEADS A N D CAPTIONS

SEC.

9-A

903. Use of "Subject," "Class," or " I t e m . " - W h e r e the stub
listings are comparatively unrelated and cannot be generalized in a
few words, it may be necessary (but rarely desirable) to confine the
stubhead description to such a term as "Subject," "Class," "Group,"
or "Item," standing alone. Although in practice these terms frequently are used interchangeably, they may be distinguished roughly
as follows:
a. " S u b j e c t . " - " S u b j e c t " is suggested where t h e s t u b presents listings in
several broad fields with a comparatively slight amount of detail presented for
each.
b. "Class." "Class" is suggested where t h e s t u b listings relate to a single
common field and t h a t field is being divided into a variety of subgroupings with
minimum detail shown for each. "Group," generally speaking, may be used
interchangeably with "Class."
c. 'Item. - " I t e m " is suggested where t h e emphasis is upon detail in classification or where the stub is composed of a long list, or lists, of coordinate captions,
particularly of quantitative material.

904. Order of listing: General.-The necessity of a consistent
pattern is the most important consideration in determining the order
in which classification terms should be listed in the stubhead, particularly for all tables within any given report.
a. Dangers of inconsistency.-If identical stubs appear in several tables
(table differences being confined to t h e boxhead), t h e user m a y be confused if t h e
several stubs are described differently, such as "Age, race, and sex, "Sex, race,
and age," or "Age by race and sex."
Frequently, stubs which look alike at first glance are radically different in
content. In such cases, variation in stubhead description must be relied upon to
warn the reader. Improper and "meaningless" variations of equal, or greater,
magnitude appearing within the same set of tables will effectively destroy t h e
likelihood t h a t t h e reader will note t h e shifts t h a t have a real significance.
b. Methods of insuring consistency.-The need for a standard rule is greatest
where the details of table design for a large number of tables must be worked on by
a number of people over a period of time. Three methods are outlined below in
par. 905. One of these should be selected at the outset and t h e decision m a d e
known to all.

905. Order of listing: Specific.-Of the following three systems,
either of the first two is normally acceptable. The third is recommended particularly for tables comprising a series. However, in
Bureau of Census practice the term "Area," or the area designation
"County," "City," "State," etc., is normally listed first in the stub
heading if the stub includes an area classification.
Method A. Descending order of stub levels (satisfactory).-The most commonly used method. List first (after "Area") t h a t classification which describes
the main centered headings. Then follow with the remaining classifications in

106

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

SEO.

9-A

THE STUBHEAD OR BOX

PAR.905

descending order, such as secondary center heads, classifications described on
total lines, and classifications described in individual line captions.
Example A-1:
County, color, marital
status, and age
ALLEN COUNTY
Single__

White

Under 45 years_-

45 years and over __

Example A-2:
County, color, age, and
marital status
WHITE
Allen County
Under 45 years __

Single _
Married _

M ethod B. Ascending order of stub levels (satisfactory).-Same principle as
for descending order except that the listing is reversed. T h a t is, of the items
chosen for listing in the stubhead, list first (after " A r e a " ) the least significant in
terms of indent or level of head. The others follow accordingly, with the classification shown in the main head listed last.
Example B-1:

County, age, marital
status, and color
ALLEN COUNTY

White
Single_.

Under 45 years_-

45 years and over_-

Example B-2:

County, marital status, age,
and color
WHITE
Allen County
Under 45 years_Single_.

Married __


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

107

PAR.

905

T H E STUB : HEADS A N D CAPTIONS

SEC.

9-A

Method C. Emphasizing table or stub purpose (recommended for series
tables) . - P l a c e first (after "Area") t h e most important classification in terms of
table purpose if t h a t classification appears in t h e stub. Otherwise, place first
(after " A r e a " ) that s t u b classification which most strongly differentiates t h e s t u b
listing in t h e given table from t h e s t u b listings in other tables in t h e same series.
In each case, list t h e remaining classifications as in method A, or B, above.

Example: Assume a series of three tables, the s t u b of each showing citizens
and aliens classified by sex, with t h e following purposes of presentation:
Table 1: An age distribution;
Table 2: A color distribution; a n d
Table 3: A marital status distribution.
Following are hypothetical stubs of such a set of tables captioned according to
method C. Below each, for comparative purposes are shown captions according
to methods A and B. Note t h a t methods A and B stress features common to
all three stubs. Method C stresses the difference between t h e m ; thereby applying
to s t u b "titles" the same principle of "difference" applied to table titles in paragraphs 311 and 332.

Example C. Stub captions according to method C:
Table 1

Table 2

Table 8

Age, citizenship, and sex

Color, citizenship, and sex

Marital status, citizenship, and sex

CITIZENS
All ages _

CITIZENS
Total_.

CITIZENS
Total

Male_
Female_

Male_Female_--

Male_Female _

Under 45 years_.
Male_
Female __

White_Male_Female-

Single
Male_Female _

45 and over_
Male_
Female_

Nonwhite_Male_Female __

Married _
Male-Female _

ALIENS
All ages__

Male

ALIENS
Total_-

_

ALIENS
Total_..

Male_-

Male_-

S t u b captions of above according to methods A and B:
Table 1

Method A: Citizenship, age, and
sex
M e t h o d B: S e x , age, a n d
citizenship

Table 2

Citizenship, color, and sex
Sex, color, and citizenship

Table 8

Citizenship, marital status, and
sex
Sex, marital status, and citizenship

906. Disadvantages of mechanical listing. Method C (par. 905,
above) overcomes the principal objections to listing mechanically
on the basis of stub level, whether descending (method A) or ascending
(method B). In either case (methods A and B), the mechanical listing may result, particularly in a series of standard tables, in stressing
(by placing first) that classification which is common to all tables in
the series.1 In consequence, the differentiating characteristic may be
buried in the middle or at the end of the stub description.
1 T h e Bureau of Census practice of giving "Area" precedence in the stub box (see par. 905, 3rd sentence)
constitutes an exception to this rule. This is because the reader's first problem, when using Census Bureau
reports, is to locate the area in which he is interested. This problem does not arise when a series of tables is
presented for a single area, or where the area classification is shown elsewhere than in the stub.

108

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

SEC.

9-A

THE

S T U B H E A D OR BOX

PAR.

906

Example: In a report containing 25 tables devoted to a detailed age
distribution of various groups of the population, a stubhead listing
based on method B might well result in "Age" appearing first in the
stub description of all tables. Such a stress is of dubious value to the
reader who knows from the report title that age data are the primary
subject of the report. Actually, the differentiating feature between
table stubs might be that described in a secondary level of center head
in the stub. In this case, use of either methods A or B will tend to
bury it. By use of method 0, the differentiating factor can be given
prominence irrespective of its placement level within the stub proper.
907. Amplifying title description.-On occasion, a classification
description may be highly condensed in the title because of space
considerations or because inclusion of the full description would make
the table title difficult to read. In such cases, assuming that the
classification involved is presented in the stub, the stub heading
might well carry the description in full.
Example: Assume the existence of three different standard stubs
comprising industry descriptions, the difference between the stubs
being confined to the number of classifications shown; that is, to the
difference in stub length or number of items. In such cases, normal
practice requires that a distinctive description be assigned to each,
such as "Major industry group," "Intermediate industry group,"
and "Detailed industry group."
a. All listings represented.-If, in a series of tables, all three types of listings
are represented, then t h e table titles and the stub boxes should carry the full
description; that is, " M a j o r industry group," "Intermediate industry group," etc.
b. One listing represented.-If, in a given series, only one table appears with
an industry classification, or if t h e same industry classification is used in all tables,
it is permissible, and frequently wise, merely to carry I N p u s r R Y in t h e table title,
and to rely upon the stubhead to carry the full formal description, such as "Major
industry group.

908. Stubhead captions to be avoided.-The stubhead is sometimes
used as a depository for descriptive or other statements which belong
elsewhere in the table or for which no other "convenient" location
can be found. Use of the stub box as a miscellaneous file interferes
with its true function. At times, this practice may lead to confusion.
Common practices to be avoided are as follows:
a. Improper indication of presentation u n i t . - U n i t indication in t h e s t u b box
should be restricted to t h a t which refers specifically to the s t u b classification.
Even then it might better appear(1) As a part of the headnote if the unit indicated is common to the entire
stub;
(2) As a parenthetical expression under a descriptive center head if it applies
only to a given stub block; or
(3) At the right-hand side of t h e s t u b (or in a unit-indicator column) if it
varies from line to line.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

109

PAR.

908

THE

S T U B : HEADS AND CAPTIONS

SEC.

9-A

However, if the unit indication applies to the entire stub, it is acceptable to
place it in t h e stubhead. For detailed discussion of handling unit-indicators, see
paragraphs 961 to 966.
Example A. Reference is to stub classification.
Acceptable, but better to insert "years" after
line captions where space permits.
Sex and age (years)

Total

Male, 14 and over_14 to 19.-

20 to 2 4 -

25 to 44.-

Example B. Reference is to figures appearing In
table field. Not acceptable.
Amount (number of sales)

Total
sales

Total amount_-

Under $50,000-$50,000 to $100,000 ..
$100,000 to $150,000 ..

b. Headnote-type of s t a t e m e n t . - A v o i d use of 'the s t u b box for insertion of
qualifying statements properly placed in the headnote. Ordinarily such statements are permissible in t h e s t u b heading in only those cases where the headnote
is already so extensive t h a t t h e qualifications, although of imperative importance,
would tend to be buried. However, if a qualification of this t y p e is at all a p p r o priate for placement in the stubhead, it is usually such t h a t it would better be
placed in a footnote, if headnoting is impracticable.
In the example which follows, t h e original tables carried no headnotes in many
cases and an exceedingly brief headnote in others. All tables carried the stub-box
statement " F o r definitions and explanations, see text" even though (1) in many
of t h e tables t h e statement applied also to t h e boxhead classifications, (2) footnoting of some types of explanations was extensive, and (3) there was no place
other t h a n t h e text where the user could be expected to look for such definitions a n d
notes. Also, the composite map numbers, referred to in the example, appeared
within the boxhead; not in t h e stub.
Example C-1. Original presentation: Headnote reading " F o r t h e 17
Western States and Arkansas a n d Louisiana"; s t u b head as follows:
Item
[For definitions and explanations, see text. Major basin totals from 1930 and 1920
Censuses include figures for unidentified tributary b a s i n s . A composite map index-number
is shown in parentheses for each drainage basin]

Example C-2. Revised presentation: Confine stubhead to " I t e m , " omit
reference to text, and shift remainder to headnote, revised as follows:
(For 17 Western States, Arkansas, and Louisiana. Major basin totals from 1930 and 1920
Censuses include figures for unidentified tributary basins. Figures in parentheses in box2 a t & 3 . T M > o- s, constitute composite map mdex-numbers assigned to speciied drain-

110

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

8EC.

9-A

THE STUBHEAD OR BOX

PAR.

908

c. Footnote-type of s t a t e m e n t . - A v o i d use of t h e s t u b box for insertion of
statements which could as well be placed in footnotes. Thus, in the above
example, if headnote space had not been available, the reference to major basin
totals might well have been placed in a footnote with the reference number
attached to the stubhead, as " I t e m ."
T h e footnote would read "1 Major basin
totals . . . .

909. Cross-references.-Cross-references in the stubhead should
be held to a minimum, particularly generalized cross-references to
text. Obviously, it would be easy to include in every boxhead such
a statement as "(For definitions, see text)." Furthermore, since
such a cross-reference should apply to the stub as a whole, it might
more reasonably appear as a footnote to the stubhead; or, better
still, it might appear as a headnote.
However, on rare occasions a cross-reference to another table may
be inserted parentheticallya. To remind the reader of an important relationship between this s t u b and
a classification presented elsewhere in the same report; or
b. To indicate the location of some figure, listing, or fact which is important
to the understanding of the classification shown.
Example A:
Labor force status
(includes institutional population omitted in table 6)

Example B:
Owner-occupied units by mortgage status
(see table 2 for total dwelling units)

Sec. 9 - B . Center Heads and Subheads (911-920)

911. Definition.-Center heads or subheads, as their names
imply, are descriptive or classifying headings centered horizontally
in the stub.
912. Purpose and function.-The purpose and function of the
center head or subhead is threefold:
a. To define, describe, or qualify s t u b entries following it.
b. To separate the table into readily identifiable blocks in terms of major
aspects of the data.
c. To state or imply relationships among the various sections of the table, or
within t h a t section to which the given head or subhead applies.

913. Comparison with colon and dash lines. The center head
and subhead and the colon line perform identical functions. However, the colon line is subject to normal rules of data-line indent;


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

111

PAR.913

THE

STUB : HEADS A N D CAPTIONS

SEC.

9-B

it is not centered in the stub. In contrast, the dash (read-in) line
is not a heading. It is an integral part of each data-line caption
indented under it, and it is intended to be read as such.
The use of colon and dash lines is frequently confused in table
design. Therefore, in order to bring out the essential differences in
purpose and treatment, they are discussed together in section 9-C,
below.
Example B. Dash, or read-in, line:
Number of families havingNo children_One child __
T w o children _

Example A. Colon line:
By race:
White_Negro__Other races_-

Typical instances of the misuse of the colon and dash are (a) the
colon is used where no end punctuation is needed for the caption, and
(b) the dash is mistakenly substituted for the colon.
914. Range of influence.- A given center head or subhead qualifies all succeeding subheads and line captions until another center
head or subhead of equal or higher level is reached. No exception to
this rule is permissible.
a. Basic questions.-The principle involved here seems so obvious t h a t it
tends to be taken too much for granted. Over-concentration on the answer to
t h e first of t h e two questions listed below is the most frequent cause of error.
Both questions are equally important.
Question 1: Is this particular segment of d a t a of sufficient importance to
warrant emphasis by insertion of a centered head or subhead? If "Yes,"
insert i t ; if "No," be sure to take question 2 into account.
Question 2: (Assuming center heads or subheads have already been
inserted in other portions of t h e stub) Is it necessary to insert a centered
head or subhead here to make clear t h a t these d a t a are not a part of t h e d a t a
described by the immediately preceding center head or subhead?
b. Conclusion.-A center head or subhead must frequently be inserted, not
because t h e d a t a affected are important in themselves, but merely to prevent their
erroneous inclusion within t h e range of influence of the immediately preceding
head.
Example A. Omission of beading for initial group:
A-1. Undesirable because time period for first
block not stated:

A-2. Preferable:

Year or quarter

Year or quarter

YEAR

1942
1941
1940

1942
1941
1940
QUARTER

1942: 1st quarter_.
2d quarter_-

3d quarter_.

112


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

QUARTER
1942: 1st quarter_-

2d quarter __
3d quarter_.

SEC.

9-B

CENTER READS A N D SUBHEADS

PAR.

914

Example B. Omission of heading for terminal group:
B-2. Right:

B-1. Wrong, because region for last block not

stated:

Region and country

Region and country

AMERICA

Canada-French_Canada-Other_-Newfoundland_Mexico ..
Other America_-

AMERICA

Canada-French_-Canada-Other_-Newfoundland_-Mexico_Other America_--

Australia.
Azores_Other and not reported.-

ALL OTHER

Australia_Azores_Other and not reported_-

915. Levels of center heads.-The coordinate or subordinate
relationships between centered heads and subheads must be made clear
by difference in mechanical presentation, such as differences of type
size, face, or boldness, or by use of capitalization schemes. In offset
work prepared with the standard typewriter, reliance must be placed
on capitalization only since variation in type size, face, and boldness
cannot readily be achieved when typing. (See par. 916 and fig. 14.)
a. General principles.-Irrespective of reproduction method, the following
general principles are applicable:
(1) Center heads of superior classification are shown in larger or bolder
type, or with greater use of capital letters, than heads of inferior classification.
(2) W i t h i n a given t a b l e , center heads of coordinate construction-level
should be in the same presentation form (that is, the same type size, face,
boldness, and capitalization). Thus, in example A - 2 , the heading "Ruraln o n f a r m " s h o u l d be in bold c a p s since in c o n s t r u c t i o n - l e v e l it is c o o r d i n a t e w i t h

"URBAN."
Example A-1. Right:

Ezample A-2. Wrong:

URBAN

URBAN

Male

Male

RURAL-NONFARM

Male

(x)

Rural-nonfarm
Male

(3) B e t w e e n t a b l e s , particularly within t h e same series, all heads of comparable construction-level should be identical in presentation form even though
this may mean a difference between tables for headings consisting of identical
terms.
2 When the Varitype machine Is used, boldface and italic type and various printing-style type faces are
available for photo-offset work. The effect produced, and the flexibility provided, is intermediate between
that achieved by use of the standard typewriter equipped for typing statistical tables and the results achieved
by the use of letterpress (type-set) composition, whether it be by means or the monotype, lino type, or intertype, or is hand-set. The Electromatic Proportional-Spacer typewriter also offers improved legibility for
typewriter offset material; high-speed typing production is combined with a printing-style type-face.
Finally, the use of standard typewriters can be combined with typewriters constructed with special sizes
and faces of type to provide italic and boldface lines and larger type for stub centered heads and table titles.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

113

PAR.

t

915

THE STUB: HEADS A N D CAPTIONS

SEC.

9-B

Example: In table 1, age d a t a are classified by urban-rural residence a n d
then by sex; in table 2, age d a t a for males and females are classified by color.
T h e primary heads are in bold caps, as "URBAN" in table 1; "MALE" in
table 2. The two subordinate heads are also treated alike but are in lower case
(small letters), as "Male" in table 1; "White" in table 2. (See example B.)
Example B. The capitalization shown here is correct in each case; also, the two presentations are consistent with each other:
Table 2
Table 1
MALE
URBAN
Male
White
All ages_.

Under 45 years_.

45 years and over_

All ages__

Under 45 years_-45 years and over __

T h e fact that the term "male" is a center head in both tables does not mean
it must appear with the same capitalization both times. On the contrary,
the presentation-form depends entirely on t h e given construction-level at
each appearance.
b. Restrictions on number of levels.-Not more t h a n three levels of center heads
should be used in either letterpress or offset work. If t h e number of levels to be
presented numbers more t h a n three, t h e table plan should be re-examined to determine t h e possibility of splitting the table so as to make two or three tables.
(1) Mechanical l i m i t s . - I n letterpress, t h e number of type sizes and faces
is virtually unlimited in theory but is definitely limited in practice. In typewriter-offset, the limit exists both in theory and in practice.
(2) Reason for practicable l i m i t s even in l e t t e r p r e s s . - N o reader can
be expected to note, or to remember, subtle distinctions in t y p e size, boldness,
a n d face. Thus, few readers can distinguish between short lines of 6-point and
8-point t y p e of identical face, boldness, and capitalization unless the lines involved are placed fairly closely together. T h e same is true for isolated words
of typewriter pica and typewriter elite when separated by a mass of text
or figures.
To be of value, level-distinction. in headings must be such t h a t t h e reader
cannot only grasp it quickly when looking at a given page, but also can carry
t h a t impression from one page to another in a multipage table. Distinctions
t h a t are not obvious distract t h e reader from t h e subject matter of t h e table
and may even mislead him.

916. Methods of level-distinction.-Although exact methods of
level-distinction are necessarily different in letterpress and offset,
in each instance established methods exist in Bureau of the Census
practice which take the various limitations into account.
a. Letterpress methods.-In letterpress work (type-set composition) t h e distinction among various levels of center heads may rest upon various combinations of boldness (blackness) of t h e t y p e and t h e capitalization. Any one of
t h e three combinations shown in figure 14 is acceptable. However, combination
3 is not suitable in cases where two or three levels of heads are used with one level
consisting of a specification of a year (1940) standing alone. (See par. 916d,
below.) Other combinations are possible but should be approached cautiously.
b. Offset m e t h o d s . - E x c e p t where special typewriters are employed, t h e distinction between center-head levels in materials typed for offset must be confined
to capitalization a n d underscoring schemes. An acceptable combination is
illustrated in figure 14.

114

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

SEC. 9 - B
FIGURE

CENTER HEADS AND SUBHEADS

14 - L E T T E R P R E S S

AND TYPEWRITER-OFFSET COMBINATIONS

F O R C E N T E R H E A D S IN
A.

F G . 14

STUB (See p a r . 915a)

Letterpress Examples

[Combination 1 is used throughout this manual]

Single level only:

Combination 1

Combination 2

Combination 3

FOREIGN BORN

Foreign Born

F O R E I G N BORN

Foreign Born

F O R E I G N BORN

Widowed and divorced

Widowed and divorced

FOREIGN BORN

Two levels of heads:

Widowed and Divorced
FOREIGN BORN

FOREIGN BORN

F O R E I G N BORN

W I D O W E D AND DIVORCED

Widowed and Divorced

Widowed and Divorced

Three levels of heads:

First papers

First papers

B.

First papers

T y p e Specifications for Letterpress E x a m p l e s

Combination 2

Combination 1

Combination 3

Single level only:

(1) 6-point antique
bold, caps

Or (1) 6-point antique or (1) 6-point caps
bold, caps and lower
case

Two levels of heads:

(1) 6-point antique
bold, caps

or (1) 6-point antique or (1) 6-point caps
bold, caps and lower
case

(2) 6-point caps and lower (2) 6-point italics, lower (2) 6-point italics, lower
case
case
case
Three levels of heads:

(1) 6-point antique
bold, caps

or (1) 6-point antique
bold, caps

(2) 6-point caps

(2) 6-point caps and lower (2) 6-point caps
case
lower case

(3) 6-point italics, lower

(3) 6-point italics, lower

case

c.

and

(3) 6-point italics, l o w e r

case

case

Typewriter-Offset E x a m p l e s W i t h Specifications

Examples
Single level

or (1) 6-point caps

only:

Two l e v e l s of heads: 1

Specifications

FOREIGN BORN

( 1 ) All

caps

FOREIGN BORN

( l ) All

caps

Widowed and D i v o r c e d

T h r e e l e v e l s of heads:

FOREIGN BORN

Widowed andDivorced
F i r s t papers

( 2 ) I n i t i a l caps ( c a p s and l o w e r
case)
( l ) All

caps

( 2 ) I n i t i a l caps (caps and l o w e r
case), underscored
( 3 ) Small

l e t t e r s (lower case)

1 Where two levels only appear, and one level is composed of year-specification standing alone, underscore
the uppermost level as:

1940

FOREIGN BORN

Foreign born

1940

776554
49


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

115

PAR.

916

THE

S T U B : HEADS A N D CAPTIONS

sEC.

9-B

c. Total block given same level as components.-In a table composed of repeated blocks of data of coordinate classification, preceded by a totality or summary block, t h e center head for the summary block, is not distinguished from
those for t h e subordinate blocks. This is in accordance with the principle t h a t
no more head levels should be established t h a n is absolutely necessary.
Example: Where a table is comprised solely of blocks for "Total," "Male,"
and "Female," assign the centered head for "Total" to the same level as for
" M a l e " and "Female." (Letterpress head combination No. 1 is used for
purposes of the example; see fig. 14.)
Example D-1. Customary:

Example D-2. Unnecessary distinction:

TOTAL
MALE
FEMALE

TOTAL

Male
Female

d. Lack of distinction between lower case, and caps and lower c a s e . - N o
a t t e m p t should be made to indicate difference in t h e level of center heads by
relying exclusively upon the difference between caps and lower case (This Is
in Caps and Lower Case) and lower case (This is in lower case). Since t h e distinction rests entirely upon t h e use of initial caps for the important words, a singleword heading (such as " W h i t e " ) or a specification of a year (such as "1940")
will appear exactly t h e same with both methods of capitalization. Combinations
1 and 2 suggested in figure 14, p. 115, avoid these difficulties. Combination 3
avoids it for single words, but not for years.
Example E.

Comparisons:
Caps and lower case
Widowed and Divorced

Lower case
Widowed and divorced

White

White

1940

1940

Solution: See combinations offered in fig. 14.

917. Omission of center heads in special c a s e s . - I n some cases it
m a y be worth while to omit center heads, particularly where a satisfactory substitute can be found or where t h e omission will not be
misleading.
a. Substitute colon lines, or use group total lines, in short tables where (1)
only one level of heads would be required and (2) only a few lines of d a t a are to
appear under each heading.
Example A-1. Top-heavy:
AGE

All ages _
Under 45 years__45 years and over_RACE
All classes_White_Negro_-Other r a c e s SEX

Total_-Male __
Female_-

Example A-2: Satisfactory:
Age:
All ages_Under 45 years_45 years and over__Race:
All classes_-White _
Negro__Other races. __
Sex:

Total_Male_-Female_-

a The latter problem ("1940" standing alone as a centered head) also arises between caps, on the one band,
and small caps, on the other, since small-cap numerals rarely are available. Combinations 1 and 2 in fig.
1 4 , have been designed specifically to solve all problems of this type. On each level, in these combinations,
the type selection is distinctive for both letters and figures.

116

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

SEC.

9-B

PAR.917

C E N T E R HEADS AND SUBHEADS
Example A - 3 . Satisfactory:
All ages_Under 45 years_.45 years and over_All races __
White_Negro_-Other races-Both sexes_Male-Female_-

b. Omit "NUMBER" where only " N U M B E R " and " P E R C E N T " heads
appear if there is no likelihood of misunderstanding. It is usually sufficient to
run the remaining head in lower case if it is the only one in the table.
Example B-1. " N U M B E R " not essential:

Example B-S. Simpler and clear:

Age

A ge

NUMBER

All ages_.Under 45 years.-45 years and over_-

429

All ages_Under 45 years_-45 years and over_-

225

PERCENT
All ages_--

100.0
65. 6

Under 45 years__45 years and over_-

34. 4

654
429

225

PERCENT

All ages_Under 45 years_-45 years and over_-

100.0
65.6
34. 4

c. Omit the first heading if it describes summary lines inserted merely to
establish the framework of the distribution which follows.
Example C-1. First heading superfluous:
Major occupation group and industry group
POPULATION AND EMPLOYMENT STATUS
Total population (all ages) __

All persons 14 years old and over_In labor force_--

Employed workers (except on public emergency work)_
MAJOR OCCUPATION GROUP
Employed (except on public emergency work) __

Professional workers_Occupation not reported_INDUSTRY GROUP
Employed (except on public emergency work) __

Agriculture, forestry, and fishery__Agriculture_-

Example C-2. Simpler and clear:
Major occupation group and industry group
Total population (all ages) __

All persons 14 years old and o v e r In labor force_-Employed workers (except on public emergency work)_-MAJOR OCCUPATION GROUP
Employed (except on public emergency work)._

Professional workers_-


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

117

PAR.

918

THE

S T U B : HEADS A N D CAPTIONS

SEC.

9-B

918. Combination of head levels: General.-The question of possible combinations of head levels arises in two different types of cases.
a. To avoid multiple levels of h e a d s . - O n t h e one hand, it is desirable t h a t
center heads be as brief as possible so t h a t t h e content may be grasped quickly
a n d remembered readily. On t h e other hand, single-word heads, particularly
in typewriter-offset work, do not stand out well in themselves, even though one
level is in caps or is underscored, and distinctions as to comparative bead-level
tend to be overlooked.
Combination of two levels of heads may lead to clarity in such cases. In effect,
this means repeating t h e upper level whenever t h e lower level is to be inserted.
T h e following examples are in typewriter-offset style (see fig. 14, p. 115) but the
same situation may occur in letterpress, particularly if the upper of t h e two levels
is not in a contrasting type face.
In example A-1 t h e "head level" of the numerals is not particularly hard for
the reader to grasp readily because each block contains only a few lines. However,
the inherent defect may be serious where t h e blocks of d a t a include a large number
of lines each, particularly in multipage tables.
Example A-1. Without combination:

1940
Total
A l l ages

Under 45 years
45 years and over

.
.
.

White
All ages
Under 45 years
45 years and over
Nonwhite

All ages
Under 45 years
45 years and over

.
.
.
.
.
.

..
.

White
All ages
Under 45 years
45 years and over
Nonwhite

All ages
Under 45 years
45 years and over

1940: TOTAL
All ages
Under 45 years
45 years and over

.
.
.

1940: WHITE
All ages
Under 45 years
45 years and over

.
.
.

1940: NONWHITE
All ages

1930
Total
Al1 ages..
Under 45 years
45 years and over

Example A-2: With combination:

.
.
.

Under 45 years
45 years and over

.
.
.

1930: TOTAL
All ages
Under 45 years
45 years and over

.
.
.

1930: WHITE
All ages
Under 45 years
45 years and over

.
.
.

1 9 3 0 : NONWHITE

All ages
Under 45 years
45 years and over

.
.
.

.
.
.

b. To make possible an essential fourth l e v e l . - I n typewriter-offset work t h e
mechanical limitations on methods of level-distinction make more than three
levels of center heads or subheads impracticable. In letterpress work this is
not true to t h e same extent but it is still hard for the reader to keep in mind more
t h a n three level-distinctions in terms of type size, face, boldness, etc.
Where four levels are essential, two main choices appear, of which the first is
preferable.
(1) Use the group total line to indicate the lowest level, or
(2) Combine two levels of heads by use of colons or dashes. However, if
this be done, care must be exercised in making the combination, as indicated
below in par. 919.

118


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEc. 9-B

CENTER HEADS A N D SUBHEADS

PAR.

919

919. Selection of head-levels when combining.- T h e following
general principles should be observed:
a. Combine consecutive h e a d s . - T h u s , if four levels are involved, combine t h e
first a n d second, or the second and third, or the third and fourth. Obviously, if
nonconsecutive heads (such as the first and third) are combined, table rearrangement will be necessary.
b. Maintain same order in combination as when not combined.-Reversal of
order in combination is confusing unless the order of appearance of the data is
also reversed.
e. Combine the shorter heads, clarity permitting.-This rule merely conforms
to the need for brevity in headings.
Example A-1. Easily grasped:

Example A-2. H a r d to grasp:

URBAN-MALE

NOT IN LABOR FORCE-INSIDE
METROPOLITAN DISTRICTS

d. When combining heads of unequal length, place t h e briefer head first, if
possible. Otherwise, the briefer head is likely to become lost. However, do not
reverse order to achieve this. (See par. 919b, above.)
Example B-2. H a r d to grasp:

Example B-1. Easily grasped:

INSIDE METROPOLITAN DISTRICTSURBAN

URBAN-INSIDE METROPOLITAN
DISTRICTS

e. Examples.-The first listing below (example C-1) shows a four-level problem
in typewriter-offset. Both the double and single underscore are used here to
differentiate head-level. Since in practice, the double underscore should not be
used, t h e problem is how to get rid of it. Note applications of rules c and d.
Example C-2. Combination A (Acceptable):

Example C-1. A 4-level listing (Not acceptable):
URBAN

URBAN-INSIDE METROPOLITAN DISTRICTS

Inside Metropolitan Districts

Married
Years of school completed

Married
Years of school completed

Example C-4. Combination C (Acceptable):

Example C-3. Combination B (Acceptable):

URBAN--INSIDE METROPOLITAN DISTRICTS

URBAN
Inside

Married-Years of School Completed

Metropolitan Districts

Married-Years of school completed
Example C-5. Combination D (Undesirable):

URBAN
Inside Metropolitan Districts-_Married
Years of school completed

920. Combination of head levels: Limitations on u s e . - H e a d combinations should be avoided unless they overcome a specific difficulty.
In general, the combinations tend to be hard to read and to grasp.
Their only merit is the avoidance of even worse situations. As long
as only three levels of headings will be required if no combination is
made, do not combine except to take care of the situation described
in par. 918a, above; that is, to avoid single-word heads in a context
where they do not stand out well in themselves.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

119

PAR.

921

THE

S T U B : HEADS A N D CAPTIONS

SEC.

9- C

Sec. 9-C. Colon and Dash (Read-in) Lines (921-933)

921. General.-These two types of stub entries have little in
common except (a) strong similarity of appearance and placement and
(b) the fact that cell entries never appear opposite them. Errors in
their usage are almost exclusively due to a misunderstanding as to
their difference in function. Typically, the dash tends to be used
(erroneously) in place of the colon; rarely, the other way around.
a. The colon line is a heading and should be handled as such. T h e only real
difference between it and a center head or subhead is t h a t the colon line, running
within the s t u b detail, is subject to indent rules instead of being centered. T h e
colon line does not "read in" 4 to its subentries. Each subentry of a colon line
is a complete line caption in itself.
Example:
Nonwhite:
Negro_Other_-

b. The dash (read-in) line is a fragment; it is never complete in itself. A
device to avoid repetition of an identical word or phrase at the beginning of a
group of consecutive coordinate entries, it is an integral part of each of its subentries taken individually. As such, it always "reads in" smoothly to each
such subentry.
Example:
Needing major repairs, withWhite occupants_Nonwhite occupants _
T H E C O L O N L I N E (922-925)

922. Colon line defined.-A descriptive, classifying, or qualifying
heading run within the detailed stub, conforming to data-line indent
and ending in a colon (:). No cell entries appear opposite it in the
field of the table.
Example:
By month:
January_
February _
March_--

46
39
51

923. Colon line compared with dash line.T h e use of the colon
line arises from the need for classification or description of its subentries. The use of the dash (read-in) line arises from the desire to
avoid repetition of an opening word or phrase common to a number of
successive lines. The colon line is not designed to "read in" to each
of its subentries; if it is capable of "reading in" it is a coincidence.
T h e dash (read-in) line is designed specifically to "read in" to each of
its individual subentries taken separately (not as a series); if it does
not "read in" properly, the use of the dash is incorrect. For comparative use of colon and dash lines, see fig. 15.
4

For definition of "read in," see par. 927.

120

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

SEC.

9-C

FIG.

COLON A N D DASH ( R E A D - I N ) L I N E S

15

F I G U R E 1 5 . - C O L O N AND D A S H L I N E S : T H E I R U S E AND M I S U S E

(See p a r . 923)
[Arrow ( ) signifies lines with incorrect or faulty use of dash or colon]

Example F-1. Original s t u b
(abridged)
STEEL WORKS AND ROLLING MILLS
All products, total value __

Example F-2. Corrected s t u b
STEEL WORKS AND ROLLING MILLS
All products, total value_-

Unrolled steel, total_IngotsProduced and consumed in works_For sale and interplant transfer.Direct steel castingsProduced and consumed in works_-For sale and interplant transfer_-

Unrolled steel, totaL
IngotsProduced and consumed in works_For sale and interplant transfer _
Direct steel castingsProduced and consumed in works.For sale and interplant transfer_--

Finished hot-rolled products and forgings,
total_Produced and consumed in works_-For sale and interplant transfer_Concrete reinforcing bars_Merchant bars, mill shafting, etc.Steel..
Iron_Bolt and nut rods, spike and chain
rods, etc_Wire rods __

Finished bot-rolled products and forgings,
total.-Produced and consumed in works ..
For sale and interplant transfer_-Concrete reinforcing bars_Merchant bars, mill shafting, etc.:
Steel_Iron_Bolt and nut rods, spike and chain
rods, etc_Wire rods.

WIRE AND WIRE PRODUCTS (ALL
METALS)
All industries, total value __

WIRE AND WIRE PRODUCTS (ALL
METALS)
All industries, total value __

Iron and steel wire:
Plain wire, total_For sale and interplant transfer
Produced and consumed in works_-Galvanized wire: For sale and interplant transfer __

Iron and steel wire:
Plain wire, t o t a L .
For sale and interplant transfer . . .
Produced and consumed in w o r k s Galvanized wire: For sale and interplant transfer_-

Copper wire:
For sale and interplant transfer_
Produced and consumed in works_-

Copper wireFor sale and interplant transfer _
Produced and consumed in works_-

Fabricated wire products for sale:
Iron and steelNails, brads, and spikes_Tacks_Staples_Other fabricated iron and steel wire
products. . .

Fabricated wire products for sale:
Iron and steelNails, brads, and spikes_Tacks_Staples_Other fabricated iron and steel wire
products. . .

Nonferrous wire productsFly screeningCopper and bronze_Other metals_Fourdrinier and cylinder wire cloth . . .
Other wire clothCopper and bronze_Other metals_Copper strand, except insulated-

Nonferrous wire products:
Fly screening:
Copper and bronze_Other m e t a l s Fourdrinier and cylinder wire cloth_-Other wire cloth:
Copper and bronze_Other metals_Copper strand, except insulated


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

121

PAR.

924

THE

S T U B : HEADS A N D CAPTIONS

SEC.

9-C

924. Single subentry.-Where colon lines are' used, and one (or
more) of the categories has only a single subentry, the subentry should
be run in on the same line as the colon line, preserving the colon.
(Note: For dash lines, the same rule applies except that the dash is
then omitted, see par. 929.)
Undesirable:
Increase:
1930 to 1935--

Right:
Increase: 1930 to 1 9 3 5 -

925. Typical u s e s . - T h e colon line is used for many purposes.
few of the more common uses are illustrated below.

A

a. To indicate the nature of the classification employed in its subentries, particularly when such classification differs from t h a t utilized in t h e data lines in the
preceding group.
Example A-2:
By kind:
Anthracite coal_-Bituminous coaL _
Coke purchased _
By use:
Generator and boiler fuels_-Bituminous coal carbonized_--

Example A-1:
Race:
White_Negro_-Other races __
Sex:
Male __
Female_-

b. To indicate the kind of data classified by t h e subentries, or t h e area or
group covered, when the subentries do not make it clear in themselves.
Example B-1:
Diphtheria:
Both sexes_Male __
Female_-

18
11
7

Example B-2:
Birmingham:
Immediate rate_Objective rate_--

4. 98
4. 92

c. To maintain the identity of a group for which data are presented for t h e
component parts but not for t h e group total.
Example C-1:
White_Nonwhite:
Negro_.Other__-

40
8
2

Example C-2:
Automotive dealers:
Automobiles and trailers __
Accessories, parts, etc_Filling stations_--

d. To indicate the presentation unit for t h e d a t a classified by the subentries
when, otherwise, t h e presentation unit might be in doubt. (Note t h a t this is a.
means of avoiding use of a field spanner above t h e figures in t h e field. See par
1434c.)
Example D-1:
Thousands of dollars expended:
1927-1928 _
1929--1930--

Thousands of units produced:
1927--1928--1929 _
1930---

Example D-2:
Number of families:
One person __
Two persons __
Three or more _
Example D-3:
Per 1,000 of the population:
Under 45 years old-45 years old and over __

e. To indicate a common effective date or portion of a year.
Example E-1:
First quarter:
Under 4 5 - 45 and over_(or)
July 1, 1941:
Under 4 5 45 and over_

122

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Example E-2:
First quarter:
1941-1942--

July 1-

(but)

1941
1942

sEc. 9-C

COLON A N D DASH

(READ- I N ) L I N E S

PAR.

926

T H E DASH (READ-IN) LINE (926-933)

926. Dash (read-in) line defined.-A stub entry, terminated by a
dash, designed to be read as an integral part of each individual line
indented under it. That is, it is a part of each subentry taken separately, not of the group of subentries as a series. It is a device
designed primarily to avoid repetition of the same opening phrase for
a number of consecutive data lines of coordinate significance.
a. Basic principle.-The basic principle involved is thatA and b __

B andThis:

is another way of saying this:

b_
C

C and a--

A a n d b __
B and a - B and bB and c_-C and a_-

b. Construction m e t h o d . - T h e common opening phrase is removed from all
members of t h e group, is inserted by itself above t h e first affected data line, and is
terminated by a dash ( - ) , not a colon. T h e data-line captions thus shortened
are then indented under it.
Example A - 1 (Before insertion of dash line):
Cropland harvested on farms operated by owners_-Cropland harvested on farms operated by managers
Cropland harvested on farms operated by tenants
Crop failure
Example A - 2 (After insertion of dash line):
Cropland harvested on farms operated b y Owners_Managers __
Tenants_Crop failure--

_acres_- 218
acres _ 451
_acres- 762
acres
63

_acres_- 218
_acres_ 451
_acres- 762
_acres_- 63

927. "To read in" defined.-Most errors in construction of nondata lines arise from the use of the dash where the colon is required.
Correct usage of the dash rests upon proper understanding of the
term " t o read in."
A line caption "reads in" to each of its sub entries if each separate
combination will read smoothly when the two portions (the dash
portion and the subentry portion) are run together without the dash.
The line does not "read in" if smooth reading requiresa.
b.
c.
d.

Insertion of a colon at the joining point;
Addition of any word or phrase to the combination;
Rearrangement of wording; or
Running in the subentries as a series with the read-in portion at the beginning.

Exception: The "residual" type of subentry (such as " A l l other")
is a special case discussed in paragraph 928, below.
928. Residual subentries: A special case. Residual 5 subentries
may, at times, be exempted from the rule that the dash line must
"read in" smoothly to each of its subentries. Since by custom the
wording of residuals tends to be condensed heavily, they frequently
cannot be "read into" properly. The following is faulty in this
1

For discussion of the residual, see par. 1143a.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

123

P.AR.928

T H E S T U B : HEADS A N D CAPTIONS

sEC.

9-C

respect but is permissible since the defect is limited to t h e residual
("Owned homes with other races").
Faulty:

Owned homes withWhite occupants _
Negro occupants.Other races __

Alternatives: Several are possible. T h e dash may be retained, or a
colon substituted, as in t h e examples which follow, but note t h a t each
has its own limitations. In particular, caution is necessary when
transposing words in such headings lest t h e meaning be changed.
Thus, in "Alternative D" below, a faulty transposition has changed
t h e unit of presentation from "Owned homes" to "Occupants of
owned homes." Frequently, it is better to accept a residual line t h a t
is faulty in terms o f " reading in" than to attempt to reword it. However, this exception to the "read-in" requirement applies solely to
residual entries; not to " l a s t lines of a group," as such.
Alternative B

Alternative A

Owned homes occupied byWhites_Negroes__Others_-Alternative C

Owned homes by race of occupants:
White_-

Negro
Other races_-

Owned homes withWhite occupants_Negro occupants_Occupants of other races __ --(Lengthy)
D. Wrong (meaning changed)

Occupants, owned homes, by race:
White_-Negro_-Other r a c e s -

929. Single subentry not permissible.-The use of the dash line
with only one subentry is incorrect by definition. R u n t h e subentry
in with t h e dash line and delete the dash, thereby forming a unified
line caption. (Note: For colon lines t h e same mechanical procedure
is followed, except t h a t t h e colon is preserved.)
Right:

Having one child or more_-

Wrong:

HavingOne child or more_-

930. Confusion of terminal with internal d a s h . - T h e sole function
of t h e terminal dash is to indicate that the line "reads in" with each
of its subentries. Do not confuse this with other uses of t h e dash
when it appears within a given line.
If a beginning word or phrase, common to two or more entries,
is separated by a dash from each, do not lift it and use it as a read-in
line, retaining the dash. In such cases, the dash should be changed
to a colon when the phrase is placed above the group. (In the following example, the use of the dash in t h e " original form" is questionable;
commas would have been more appropriate.)
Example A - 1 . Original form:
Softwoods-balsam fir

Softwoods-cedar_Softwoods-cypress_
Softwoods- Douglas f i r -

124

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Example A - 2 . Wrong:

Example A - 3 . Right:

SoftwoodsBalsam f i r - Cedar_Cypress ___
Douglas fir ___

Softwoods:
Balsam f i r Cedar_-Cypress ___
Douglas fir_--

SEC.

9-C

COLON AND DASH

(READ-IN) LINES

PAR.

931

931. Consecutive read-in lines permissible.-If table requirements
make it desirable, one read-in line may follow immediately upon
another as long as the read-in principle is maintained throughout.
That is, the first dash line must read smoothly into the second dash
line, and the second, in turn, must read smoothly into each of its
subentries, taken separately.
In this form, the line caption for each of the data line subentries is
split into three parts instead of into two.
In general, this usage should be discouraged since it presents an
additional complication in the table. In any case, this technique is
undesirable unless at least three levels of classification of data are
involved.
Example A-1. Preferred:
Number or homes:
Not needing major repairs, withWhite occupants__Nonwhite occupants _
Needing major repairs, withWhite occupants__Nonwhite occuoants __

Example A-2. Acceptable:
Number of homesNot needing major repairs, withWhite occupants_Non white occupants_Needing major repairs, withWhite occupants_Nonwhite occupants_-

Example A-3. Unnecessary (Combine the dash lines and overrun):
Cropland harvestedOn farms operated byOwners_Managers__Tenants __
Crop failure_--

932. Effect on space requirements of stub.-Over and beyond the
question of clarity or of greater ease in reading, the use of the dash
(read-in) line may increase or decrease stub depth or width. Typewriter-offset examples are used to illustrate this point.
a. Stub depth.-In general, if t h e original entries (before subtracting the
opening word or phrase) are so short t h a t overruns are not necessary, t h e addition of a dash (read-in) line will increase t h e depth of t h e stub. (See example A.)
If t h e original entries are long and require overruns, t h e read-in line will decrease
t h e depth of stub by N - 1 lines, where " N " is t h e number of overrun lines saved.
(See example B.)
Example A. Stub increased in depth:
A-1. Original:

Revenue,
From
From
From
From
From

total.
.
timber
.
water power....
special uses...
grazing
.
all other
.

DDDDDDDDDDDDDDDDDDDDDDDD

Example B. Stub decreased in depth:
B - 1 . Original:

Cropland harvested on farms
operated by owners
.
Cropland harvested on farms
operated by managers
.
Cropland harvested on farms
operated by tenants
.
Crop failure
.

" D " denotes line added if "read-in" is used.
A - . With read-in:

Revenue, total
WWW
FromTimber
WWW
Water power.... WWW
Special uses... WWW
Grazing
WWW
All other
WWW

" w "denotes spaces saved in width.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

B-8. With read-in:

Cropland harvested on
farms operated byI WWWWW
Owners•••••••••••. W
Managers......•••. W
[ WWWWW
Ten a n ts........... W
I WWWWW
Crop failure
. WWWWww

DDDDDDDDDDDDDDDDDDDDDD IWWWww
" D " denotes line saved in depth.
" W " denotes spaces saved in width.

125

THE

PAR.932

S T U B : HEADS A N D CAPTIONS

sEC.

9-C

b. Stub width.-Obviously, if t h e length of t h e original lines controls t h e s t u b
width, a decrease in maximum line width will pull in t h e s t u b width accordingly.
This may be seen in examples A a n d B, immediately above, where t h e space
saved in stub width is indicated by t h e number of W's appearing at t h e right.

933. Test of read-in application.-Errors in use of the terminal
dash can best be illustrated by examining the result when the read-in
line is coupled with each of its subentries. Note the result in the
following typical cases.
Correct usage in examples B to E, inclusive, requires the use of a:
colon; the dash is correct only if the wording is rearranged. Thus,
in example C, the wording might be changed to "Owned homes
w i t h - , " and the dash retained accordingly. (See also fig 15, p. 121.)
Effect when combined

Example A. Right:
Families havingOne child _
T w o children.. _
More than two children--

Families having one child_Families having two children_-Families having more than two children_--

Example B. Wrong:
By number of itemsOne item_T w o items_More than two items_-

By number of items one item_-By number of items two items_By number of items more than two items_--

Example C. Wrong:
Owned homesWhite occupants_Negro occupants._Occupants of other races_--

Owned homes white occupants_Owned homes Negro occupants__Owned homes occupants of other races_--

Example D. Wrong:
1940

January_February _
March_-

Example E. Wrong:
W ells, pumped and flowingArea irrigated._
_acres_
Area works were capable of supplying
with water
_acres_-

1940 January_
1940 February_1940 March_-Wells, pumped and flowing area irrigated
acresWells, pumped and flowing area works were
capable of supplying with water.-- _acres

Sec. 9-D. The Line Caption: The D a t a Line (941-948)

941. "Line caption" defined.-The descriptive title or stub entry
for any line of the table. A data line is any line containing any cell
entry whatsoever in the field of the table. The cell entry may be
informational (figure, word, phrase, reference symbol, etc.) or noninformational (leaders). Even though all cell entries on a given
line are noninformational (leaders), the line is to be considered a
data line.6
6 "All-leader" lines are undesirable. They should be omitted from copy except (a) where, in letterpress
work, retention seems desirable in order not to destroy standard presentation of repeated blocks or where
the lack of data is significant in itself; or (b) where, in typewriter-off'set work, standard table forms with
preprinted stubs are being used.

126


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

SEC.

9.-)

T H E L I N E CAPTION : THE

DATA L I N E

PAR.

941

T h e d a t a line is the basic unit of the stub since it is on these lines
t h a t t h e information appears. Dash and colon lines are not data
lines. T h e caption of t h e data line is invariably followed by stub
leaders, except where space prevents.
Eramples:

Wood manufactures_1939-Total

_ 12,623
2, 492

_ 39,752
100. 0

Paper and pulp__Land and drainage enterprises __
50 to 499 acres.-Denmark
Iceland
Chicago, Ill_-_dollars_AmountFull owners--1930-

155
623
754

942. General classes: Prime and nonprime.-Prime lines are those
rows or lines of data constituting the hard core of the table. From
them, all other, or nonprime, data lines in t h e table can be derived
by mathematical processes.
A line that is prime in one table may be nonprime in another, and
vice versa. Also, a line may be prime to a given table in a series or
report; but if it can be reconstructed from lines appearing in some
other table, or in a combination of tables, it will be nonprime to the
series or report as a whole.
These distinctions are useful in table-planning in thata. Inclusion of a prime line increases, and its deletion or omission decreases,
t h e amount of basic data made available to t h e user; whereasb. Inclusion of a nonprime line increases, and its deletion decreases, facility
of table-use by the reader.

Thus, derived figures, or group totals, are. strong interpretative
aids; their presence makes it unnecessary for t h e interested reader to
compute them himself. However, the deletion of such a line, if
nonprime, does not deprive the reader of basic data since t h e statistical
ingredients remain, and t h e user can reconstruct the deleted material.
943. Types of prime lines.-In combination, the information
shown on prime lines constitutes the total body of basic information
presented to the user. In terms of the information presented in a
given table, prime lines may be classified as follows:
a. Lines of frequency data not further classified in t h e same table.
b. Lines of derived figures (percentages, averages, means, medians, etc.) which
cannot be derived from data presented in t h e same table.
c. N onadditive total and subtotal lines; t h a t is, totals or subtotals for which
incomplete distributions are shown. If such a total line is removed, t h e reader
cannot reconstruct it from t h e data remaining.
d. "Special-group" d a t a lines presenting group figures for special-purpose or
nonstandard intervals which, if deleted, cannot be reconstructed from the data


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

127

PAR.

943

T H E STUB : HEADS A N D C A P T I O N S

SEC.

9.-D

remaining. Example: A line giving total population for those "21 years old
and over" in a table otherwise confined to standard 5-year age groups is a prime
line, since the standard interval " 2 0 to 24" does not provide a classification break
at 21 years.

944. Functions and types of nonprime lines.-Non prime lines
perform three important functions and are of several types.
a. Functions.-In general terms, nonprime lines(1) Provide significant and important direct information to t h e user, such
as grand totals and group totals. Major totals, in most instances, are virtually essential to effective use of t h e table.
(2) Provide measures of significance, such as percentages, averages, medians,
etc., which t h e user may be unable to compute for himself, or t h e need for which
he m a y not recognize.
(3) Aid in establishing t h e analytical framework which needs to be understood if t h e meaning a n d significance of t h e d a t a are to be grasped readily.
b. T y p e s . - I n terms of a given table, nonprime lines may be classified into two
main types:
(1) All total a n d subtotal lines which can be obtained by addition of t h e
data appearing in t h e same table. Where space demands a sacrifice of lineage,
it is usually necessary to delete prime lines before sacrificing major totals.
(2) Lines of other derived figures (percentages, averages, medians, etc.)
which can be derived from data presented in t h e same table.

945. Additive and nonadditive l i n e s . - A classification of data
lines useful in placement of horizontal rules. (See sec. 13-B, below,
for placement and use of horizontal rules.)
a. Additive l i n e s . - A n y one of two or more lines of entries which, taken together,
will add to a total or subtotal line presented in t h e table. T h e individual lines in
such a " g r o u p " may appear consecutively or m a y be dispersed. They may, in
themselves, be prime or nonprime; thus, one of t h e "group" may be a prime d a t a
line and one or more of the others m a y be a subtotal line.
b. Nonadditive l i n e s . - A line not a member of a n y group of lines which
together add to a total or subtotal shown in t h e table; or a n y line not subject, or
not intended to be subjected, to t h e additive process, such as means, medians, etc.

946. Capitalization.-Line captions are normally set in lower case
(This is in lower case), never in caps and lower case (This Is in Caps
and Lower Case), and rarely in caps (THIS IS IN CAPS). Captions
for grand totals, or major group totals, are sometimes, but rarely, set
in caps in typewriter-offset work. In letterpress work, the use of
boldface type for group totals makes the use of caps unnecessary.
Example A-1. Right:
All classses_White occupants __
Negro occupants__
Occupants of other races_

Example A - 2 . Not acceptable:
All Classes _
White Occupants.Negro Occupants._Occupants of Other Races_Example B-2. Rarely acceptable:

Example B-1. Right:

All classes
White occupants
Negro occupants
Occupants of other races

128

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

.
.
.
.

ALL CLASSES . • • . . . . . . . . . •

White occupants
Negro occupants
Occupants of other races

.
.
.

SEC.

9-D

T H E L I N E CAPTION : THE

Example B-4. Wrong:

Example B-3. Not acceptable:
A l l Classes
W h i t e Occupants
Negro Occupants
O c c u p a n t s of O t h e r R a c e s

946

PAR.

DATA L I N E

.
.
.
.

ALL CLASSES
WHITE OCCUPANTS
NEGRO OCCUPANTS . . . . .
OCCUPANTS OF OTHER RACES

.
.
.

947. Punctuation.-Normal punctuation is used throughout except
as follows:
a. Periods are omitted at the end of abbreviations which are followed by leaders. However, if the abbreviation is followed by a footnote reference symbol
(even though t h a t symbol is followed by leaders) t h e period is inserted preceding
the reference symbol.
Example A:

Letterpress

Right: Teachers, n. e. c __
Wrong: Teachers, n. e. C . - - -

Typewriter
Teachers, n . e . c
Teachers, n . e . c .

.
.

.

Example B:

Right: Teachers, n. e. c.1
Wrong: Teachers, n. e. c l _ -

Teachers, n . e . c . ! . . . . . . . . . . . . . . . . . . .
Teachers, n . e . c l . . . . . . . . . . . . . . . . . . . .

b. Where a line caption overruns in a table with tracer numbers, or with two
or more reading columns, t h e cell entries are placed against t h e top line of the
caption, leaders are omitted, and the overrun ends with a period. (See par. 1014.)

948. Use of roman, boldface, and italic.-In letterpress, line
captions are normally set· in roman (ordinary) type. The use of
boldface and italic is normally restricted to the specific situations
outlined in section 10-D.

Sec. 9-E. The Line Caption: Total and Subtotal Lines (951-960)
951. Definition.-The descriptive title (or caption) of any data line
composed of grand totals, group totals, or subtotals. A special
case of the line caption for the data line.
952. Various meanings of term "Total."-In tabular presentation
the term "Total" is subject to a number of different interpretations,
a fact which may lead to confusion. Several of the more important
meanings which may be assigned to the term are:
a. The result of a summation process applied to t h e line's subordinate entries;
the sum of the parts where the parts are actually shown separately. This is the
most commonly understood meaning of the term, similar to t h a t used in accounting. It is the normal meaning in such phrases as " t o t a l line," " t o t a l entry," etc.
b. The aggregate, or total number, as an independent item. This usage is
typified by such individual lines (without subentries) as " T o t a l value (thousands
of dollars)."
c. The total, or the whole, as a class in itself, or as representing a class of
classes. T h i s is typified by the "Total" class in the area classification: "TOTAL,

URBAN, RURAL."


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

129

PAR.

952

T H E S T U B : H E A D S AND

CAPTIONS

SEC.

9-E

In most instances, these differences tend to be academic in that a
given total entry may reflect two, or even all three, of these meanings
at the same time.
At times, however, it is important that the reader understand which
meaning is intended, or that one of the meanings does not apply.
Here, the table designer should keep the above-mentioned distinctions in mind and should govern his treatment of center heads, the
footnotes keyed to them, and the line captions for total and subtotal
lines, accordingly.
953. "Total"-Signifying result of summation.-Where standing
alone on a line having subentries, "Total" implies that the entries
shown against it represent the summation of the figures shown for
the subentries where the subentries are additive. Special problems
area. Inclusion in "Total" of a subentry not shown separately.-The most
common case is where t h e "unknown" category is not shown but is included in
t h e total figures. Footnote the total line (or t h e appropriate entry) and specify
" n o t shown separately."
Example A-2. Wrong (footnote needed):

Example A - 1 . Right:
Total !-

300
180
100

500
300
195

Total_

Under 45 years_-Under 45 years_.45 a n d over_-45 and over_1 Includes "Age not reported," not shown separately.

300
180
100

500
300
195

b. Presentation of selected items of a distribution.-Frequently it is desirable
to show a total with only one or two members (not all members) of a distribution.
In letterpress, italic is· used (for both caption and entries) for such subentries.
In typewriting, where italic is not available, no distinction is made if the situation
is self-evident. Where not self-evident, footnote t h e omission or make the situation clear in t h e caption.
Example B - 2 . Right (typewriter):

Example B-1. Right (letterpress):
All ages__Under 45 years.--

300

180

All ages
U n d e r 45 y e a r s

300
180

c. Failure of percentages to add to 100.0.-Since normally t h e Bureau of t h e
Census rounds, but does not force, percentages, a full percentage distribution
may not add to 100.0. This does not affect presentation practice. If t h e total
line is shown, t h e entry of "100.0" appears in each appropriate cell even though
t h e actual summation in t h a t column may be 99.9, 100.1, etc. No special treatment is needed for t h e total line.
d. Nonadditive derived f i g u r e s . - T o t a l line treatment is t h e same for nonadditive derived figures as it is for additive frequencies; t h a t is, t h e treatment
as a total line depends upon t h e relationship of t h e frequencies of t h e classes into
which t h e distribution is being made, even though t h e derived figures do not add
to t h e entry on the total line. In examples C to F, t h e " t o t a l lines" are treated
exactly alike. In example C, t h e percentages add to t h e total entry. In example
D, they do not add vertically since t h e distribution is horizontal. In example E,
each cell percentage reflects t h e cell-to-cell relationship of two different frequency
tables. In example F, medians are shown.

130

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

SEC.

9-E

THE

Example C:
Total_-

PAR.953

L I N E CAPTION : TOTAL L I N E S
Example D:

_ 100.0

100 .0

100 .0

Under 45 years ___ 55. 5
44. 5
45 and over_-

65. 0
35. 0

80. 5

Total_-

Under 45 years.
45 and over_-

19.5

Example E (Percent white):
Total_-

Under 45 years __
45 and over_-

- 100.0

75.0

25.0

_ 100.0
_ 100.0

65. 0
80. 5

35.0

- 26.5

26.7

26.2

-26. 9

27.1
23.7

26.6

19. 5

Example F (Medians):

- 44.9

65.1

24.6

- 45. 0
- 44.8

65. 2
65.0

24. 8
24.5

Total_-

Under 45 years __
45 and over_-

-23.5

23. 3

954. "Total"-Signifying the aggregate as an independent item.Commonly, this may be expressed in two ways as indicated below.
See paragraph 957 for discussion of conditions under which "Total"
should be placed first or last in the caption.
a. Stressing the totality aspect, by placing the word "Total" first:
Total value (thousands of dollars).
Total acreage (thousands of acres)_

b. Stressing the nature of the item, by placing "Total" last:
Value, t o t a l Acreage, total__-

or
or

Value (total)_-Acreage (total)_-

955. "Total" The whole as a class in itself.-Typically, this
. represents a case where a centered head reading "TOTAL" has been
moved down to, and combined with, a total-line caption reading
"Total . . . .
This operation, although frequently desirable, may
give rise to confusion in the mind of the reader unless performed
carefully.
a. Center heads must be shifted to total lines consistently, or not at a l l . - I n
t h e example below, t h e heads "TOTAL," "URBAN," and " R U R A L " are all
handled as coordinate, with " T O T A L " representing a class comprised of t h e
combined urban and rural areas. If one of these is shifted to t h e total line, all
must be. If t h e shift is made, then all total-line captions serve a dual purpose,
indicating both summation and class.
Example A - 1 . Original:

Example A - 2 .
shifted:

TOTAL

Total_-White ___
Nonwhite_-

Total ___
White_-Nonwhite_-

Urban_-White ___
Nonwhite_--

URBAN

Total__White_-Nonwhite_--

Rural_White_-Nonwhite_--

RURAL

Right.

All

Example A - 3 .
shifted only:

Wrong.

First

Total_.White_-Nonwhite_-URBAN

TotaL _
White_-Nonwhite_-RURAL

Total_-White_-Nonwhite_--

Total_-White ___
Nonwhite_--

b. Class use of "Total" as line caption where total-line entries represent summation of subentries.-Shifting of center heads to total-line captions should not
be done mechanically. The dual function of t h e "Total" caption under such
circumstances must be kept in mind. Qualifications inherent in terms of either
function must be indicated against the single line-caption.

77655449

10


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

131

PAR.

955

T H E STUB : HEADS A N D CAPTIONS

SEC.

9-E

At times, this may make the combination undesirable. Thus, if a qualification
of an entire block of d a t a is indicated by footnoting a center head, and the "Total"
caption is also qualified on a point pertaining to the summation aspect only, the
shift of t h e center head to the total line is undesirable. Not only will two reference symbols be necessary against t h e single word "Total," but one reference
will refer to t h e entire block a n d the other to the single line only. This is likely
to confuse the reader. (See pars. 642-643.)
Example B-1. Clear:
URBAN

Example B-2. Confusing:
Urban 12

I

Total _
Under 45 years_-

500
280

210
45 years and o v e r 1 Outside metropolitan districts.
'Includes "Age not reported," not shown
separately.

500

Under 45 years_280
45 years and over_210
1 Outside metropolitan districts.
Includes "Age not reported," not shown
separately.

956. Multiple uses of total-line captions.- Within the limitations
expressed above, the word "Total" may be omitted completely where
it is obvious t h a t a given line of entries comprises totals or subtotals.
Common uses of total-line captions which reduce t h e number of
center heads required area. To make clear the exact group being classified:
EMPLOYED

Total__White persons_Non white persons_-

can be listed

Employed persons_White_-Nonwhite_--

b. To make clear the unit of presentation:
FAMILIES
Total_

Having no children_Having one child_--

can be listed

Number of families _
Having no children_
Having one c h i l d - -

c. To clarify the nature of the classification:
Total_-Under 45 years old_-45 years and over_-

can be listed

All ages_Under 45 years_-

4 years and over_--

d. To specify area covered:
THE STATE

Total_-Urban_Rural_-

can be listed

T h e State _
Urban_
Rural_--

e. To avoid designating a grand total cell as "Total-Total" when the description
of the grand total column is too long to be made clear in the box:
Example C (Assume detailed caption for total column would read "Total population, 15 to 45 years old,
except persons in institutions"):
Laber force status
Occupation

Total

Total 1__
Musicians and teachers of music,
etc.
1 Clearer if it read "All occupations .

132

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

or "All occupations, total .

SEC.

9-E

THE

PAR.

LINE CAPTION: TOTAL LINES

956

f. To provide an additional head-level where additional levels of center heads
are not desired.-Ordinarily, head combinations (par. 918) should not be made
u n t i l the total-line captions have been used for head purposes.
Example D-1. Top-heavy:

Example D-2. Improved:
THE STATE

THE STATE

Total_-

WHITE
Married

Married__

White

Spouse present . . .
Spouse absent_-

Spouse present_-Spouse absent_-

957. Placement of "Total" in combination with other wording.Frequently, where a wording other than "Total" is used for a total or
subtotal line, it is desirable to include the word "Total," also. Here
the question arises whether the word "Total" should be placed first
(Total male) or last [Male, total, or Male (total)] in the given caption.
This depends largely upon the sense in which the term is used and the
position of the line in the general pattern of the table. Everything
else being equal, the following rules may be applied.
a. Place "Total" first (in the caption)(1) If the basic purpose of the caption is to indicate t h e result of summation
of subentries.
(2) If t h e word " T o t a l " represents a class in itself, particularly if there are
a number of total lines of coordinate value:
Example A - 1 . Right:
Total males . . .
Urban males _
Rural males_--

Example A - 2 . Wrong:
Males, total.._
Males, urban_-Males, ruraL.

(3) If there are no other total or subtotal lines of coordinate value which
will start with "Total."
(4) If t h e aggregate, or total number, as such, is to be stressed.
b. Place "Total" last (in the caption)-

(1) If t h e basic purpose of the caption is to indicate the nature or class of
data presented, rather t h a n t h e fact that it represents a summation.
(2) If the reason for its insertion is to avoid a misapprehension as to whether
t h e given category is entirely represented. If definitely parenthetical, parentheses are appropriate.
(3) In the case of minor group totals where the line caption of t h e subtotal
is a standard item in a coordinate listing and is appearing as such.

958. "Total" vs. "All." At times the word "All" is used on total
and subtotal lines. Although "All" may be used in place of "Total"
at times, it is not synonymous in table design and should not be considered as such.
a. "Total" may be used standing by itself.always be followed by another word or phrase.
Example A - 1 . Right:
Total_-

White-Negro. . .
Other_-


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Example A - 2 . Wrong:
AU

White_-Negro_-Other_-

All

should not be; it should
Example A - 3 . Right:
All races_-

White-Negro__Other_--

133

PAR.958

•

T H E S T U B : HEADS AND CAPTIONS

SEC.

9-E

b. Both may be used in indicating presentation u n i t . - H o w e v e r , "Total,"
when so used, implies nothing more t h a n t h a t t h e line entries represent the summation of the units shown in the subentries. "All" implies t h a t t h e line contains
figures for the entire universe comprised of units "x," irrespective of type.
Th us, " T o t a l families" means merely " T o t a l families (of the t y p e shown)."
"All families" means "All families (of every type)." Therefore, if "All" is used,
the caption should specify in full the group covered; if "Total" is used, complete
specification in the line caption is not necessary, although frequently desirable.
Example B-2. Wrong:

Example B-1. Right:
TotaL

All a g e s _

75

Under 45 years old_-45 to 84 years old_-Example C-1. Right:

Under 45 years old_-45 to 84 years old_-Example C-2. Right:

25

All ages__

All ages!__

90

Under 45 years_-45 to 84 years old-85 years and over_Age not reported_-

75

50

90

Under 45 years old_-45 to 84 years old_25
85 years and over __
11
1 Includes "Age not reported,"
not
shown separately.

25
11
4

959. Capitalization, boldface, and use of r u l e s . - T o t a l and subtotal lines are normally set or typed in lower case, thus:
Right: Number of stations_-Right: Number of s t a t i o n s

Wrong: Number of Stations_-Undesirable: NUMBER OF STATIONS • • . • . . .

.

a. Use of boldface.-In letterpress, both t h e captions of grand total and group
t o t a l lines, and t h e data appearing in the lines, m a y be set in boldface. (See also
par. 1032.)
Example:
All ages __

Under 45 years old_-45 years and over_-

100
65
35

47
29

18

38
27

11

15
9

6

b. Caps.- I n offset (typewriter), the use of caps should be reserved, if used at
all, for(1) Grand total line captions where many group subtotal captions are also
shown.
(2) Group subtotal captions t h a t are extremely short (single words) and
which may tend to be hidden by t h e longer subentries. However, t h e use of
caps should be avoided even here except where absolutely necessary. Where
used, all lines of coordinate value in construction must be treated similarly.
c. Horizontal rules and underscore.- Insert in the field only under the gran<1total entries and those major group totals which carry subentries under them,
which subentries add to the total in question. Do not carry the horizontal rule
or underscore under the captions in the stub.

960. Indention.-See section 10-A.
Sec. 9-F. The Unit-Indicator 7 (961-966)

961. Definition.-A device designed to make clear the exact unit of
measurement to which the given statistics refer. The unit-indicator
is sometimes referred to, as in the G. P. 0. Style Manual, as the
"unit of quantity."
7 For

discussion of unit-indicators in the boxhead, see par, 1225a; in the field of the table, see sec. 14-C.

134

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

SEC.

9-F

THE

U N I T - INDICATOR

PAR.

961

a. May appear in stub, box, or field.-Unit-indicators m a y appear in t h e stub,
in t h e box, or in t h e field of the table. T h e discussion below relates specifically
to their appearance in the stub, but the general principles involved apply whereever they appear.
b. Similar auxiliary d e v i c e s . - F o r t h e purposes of this manual, t h e term "unitindicator" includes all similar devices which, serving as caption auxiliaries, specify
t h e time reference ( __
1 9 2 1 - - ) ; the nature of the measure employed
_median__); the magnitude of the unit (_thousands__);
etc. The general mechanical rules are identical for all such devices, and the circumstances of proper use are the same. The discussion which follows, therefore, makes no differentiation between these classes of indicators, as such.
Examples:
_number_. . . . . . . . . . . .dollars
.
. . . . . . . . . . . . w e e k s ..
. percent ..
_median_. . . . . . . . . . . . number of s a l e s
.
. . . . . . . . . . . . p e r 1 , 0 0 0 of t h e p o p u l a t i o n ..
_acres.bushels __
. . . . . . . . . . . .1929
.
_pounds_. . . . . . . . . . . . farms r e p o r t i n g . . . . . . . . . . . . . .
Note: In typewriter-offset work, unit-indicators may be lined up either at the left or the right, as long as
the line-up is consistent in the given report.

962. Restrictions on u s e . - T h e unit-indicator is used only when
t h e presentation unit, etc., would otherwise be in doubt, particularly
where the unit varies from line to line. It is an auxiliary device,
essential in some tables, not needed in most.
a. Needed where unit varies.-Typically, it is used when t h e presentation unit
differs from one part of the table to another, particularly from line to line.
b. Not needed where unit is constant.-Where an identical presentation unit
prevails throughout the table, it should be made clear in the title or in the headnote. (See pars. 325 and 512a.) In such cases, the unit-indicator should be used
in stub, box, or field in only those cases where the wording of line captions or
boxhead captions are such as to confuse the issue.

963. Placement and use of l e a d e r s . - W h e r e u n i t - i n d i c a t o r s a r e

r u n in t h e stub, they are placed at t h e right of the given stub line.
Leaders are run solidly between the line caption and indicator; that
is, no open space precedes or follows the leaders. This rule applies in
both letterpress and typewriter offset.
Example A-1. Right: Cotton__Example A - 2 . Wrong: Cotton_-Example A-3. Wrong: Cotton

_pounds_- 78, 524
_ pounds __ 78, 524
_pounds_- 78, 524

a. Line-up of unit-indicators of different l e n g t h . - F o r mechanical reasons, t h e
line-up may vary for letterpress and offset, as follows:
(1) In letterpress, unit-indicators invariably line up at t h e right with
exactly two leader dashes (1-em leader) between each unit-indicator and the
first column rule.
ExampleB:
Cotton ..
Wheat_-

running ele-_] 2
_bushels_- 444

(2) In typewriter-offset, unit-indicators may line up at either t h e left or
the right. If they are lined up at the left, two typewriter periods should be
placed between the longest indicator on the page and the first column rule.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

135

,

PAR.

963

THE

S T U B : HEADS AND CAPTIONS

Example C-1. Lined up at t h e left:

SEC.

9-F

Example C-2. Lined up at t h e right:
Cotton
r u n n i n g b a l e s .. f 3 3 3
W h e a t . . . . . . . . . . . . . . . . . . . b u s h e l s . . ] 444

Cotton
r u n n i n g b a l e s .. , 3 3 3
Wheat . . . • . . . . . . . . . b u s h e l s . . . . . . . . 1 4 4 4

b. Overruns.-Avoid overruns of unit-indicators; t h a t is, do not break them
over to t h e next line. Carry them over entirely or not at all.
(1) If there is not enough room for the entire unit-indicator on a given line,
leave the remainder of t h e line blank; place the entire unit-indicator at the
right of the next line, with leaders following but not preceding it.
(2) T h e informational entries appear against the line on which the unitindicator falls, except in tables with tracer numbers or with more t h a n one
reader column. (See par. 1014.)
Right: W h e a t flour
1,000 sacks of 100 lb_- 14, 741

Right: Cocoa and chocolate
pounds_- 65, 224
Wrong: Cocoa and chocolate_65, 224
(pounds)
Wrong: Cocoa and chocolate
_pounds_- 65, 224

Wrong: Wheat flour__1,000 sacks
of 100 l b - 14, 741

c. "Do." for "ditto" not used by Census B u r e a u . - T h e abbreviation Do.
(meaning " d i t t o " ) is not normally used by the Bureau of the Census. When
used with unit-indicators, G. P. 0. practice should be observed as follows:
" ' D o . , ' followed by 2 ems of leaders (4 leader dashes), is used under a unit of
quantity in a s t u b if the unit is spelled; if t h e unit is abbreviated, the abbreviation is repeated."

964. Reference range.-The unit-indicator, when inserted within
the stub, is applicable to all items appearing in the cells on that line
until a unit-indicator column, another stub column, or the right-hand
edge of the table is reached. This is the same rule as for the line
caption itself.
Example A. Not clear; what is t h e relationship of t h e unit-indicator in t h e
s t u b to t h e "Value" spanner in t h e box? A s t u d y of original table showed that
total value was meant, not value per indicated unit.
Value (thousands
of dollars)

Quantity
Item
1936

1937

1938

1936

1937

1938

-_number_-1,000 l b - -1,000 gal_-

Cattle_B u t t e r ___
Milk_-

Example B. Preferred; unit in stub is clearly related to quantity figures only.
T h e "Value" spanner is not qualified by the unit column.
Value (thousands of
dollars)

Quantity
Item
Unit
Cattle ___
Butter_M i l k __

136

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Number_1,000 lb ___
1,000 gal__-

1936

1937

1938

1936

1937

1938

SEC.

9-F

THE

U N I T - INDICATOR

PAR.

965

965. Specify units of measure clearly.-Avoid popular terminology
which is not scientifically accurate. Thus, such terms as bushel,
barrel, gallon, ton, etc., may be confusing or unsatisfactory when
standing alone since these measures vary by weight according to
product or commercial practice. T h e standard practices of t h e
various subject-matter divisions of the Bureau of t h e Census, as
exemplified in their published materials, reflect the importance of
this factor and the proper method of treatment.
Where t h e exact nature of a given measure is constant for an entire
report, it may be explained in t h e text. Where it differs from table
to table, particularly when differences appear within a given table,
it should be qualified in the tabular material.
T h e following excerpt is from an export table.
Example:
Barley:
Grain
MaltBuckwheat __
Corn:
Grain_-Meal_-Hominy a n d corn grits_Kafir and milo_
Corn cereal foods, ready to eat__
Oats: G r a i n Wheat:
Grain
Flour wholly of domestic wheat__-

---1,000 b u . of 48 lb_---1,000 bu. of 34 l b - ---1,000 bu. of 48 lb--1,000 bu. of 56 l b - --1,000 bbl. of 196 l b .
-1,000 l b - --1,000 bu. of 56 lb_-1,000 lb----1,000 bu. of 32 lb_1,000 bu. of 60 lb_--1,000 bbl. of 196 lb __

966. Unit and date columns.-A unit column is advisable where
every line, or a vast majority of t h e lines, in a table requires a unitindicator. Such a column usually appears as t h e first column in the
field, headed by such a term as "Unit." (See example B, par. 964.)
Unit columns may appear elsewhere in t h e table as needed.
Similarly, where necessary to indicate the effective d a t e against
each line caption, a date. column may be added in a similar position.
A unit column counts as a reader column (see par. 1014b); a date
column does not.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

137

Chapter 10

THE STUB: CONSTRUCTION AIDS TO CLARITY
(1001-1033)
Sec. 10-A. Stub Indention (1001-1010)

1001. Indention defined.-The degree of inset assigned a given
line-caption, measured from t h e left-hand margin of the s t u b ; or the
a m o u n t of space left blank between t h e left-hand margin and the
beginning of the given line-caption.
"Indent pattern" refers to the organized method, or scheme, employed to allocate indent space from line to line, and to the over-all
effect produced thereby.
1002. Purpose and function.-To emphasize and clarify the interline relationships in terms of comparative level of classification;
t h a t is, superior, coordinate, or subordinate.
"Classification level," in this instance, refers to the construction
aspects of the table, not to absolute position in a purely conceptual
scheme based on content. T h a t is, items of coordinate ranking in a
conceptual scheme may not carry t h e same indent within a table, or
among tables, particularly where a subtotal line, colon line, or dash
line intervenes.
Thus, in t h e following example, each individual race is coordinate
in t h e conceptual ranking, and bears the same indent in table 1.
However, t h e interposition of the colon line "Other races" in table 2,
and of t h e subtotal line "Nonwhite" in table 3, results in shifting the
minor race groups to a subordinate classification level for purposes
of indent. Note, also, t h e use of t h e term "Other races" in tables 1
and 2. This is a residual, and as such has no identity of its own.
Therefore, its value will vary in terms of t h e detail of classification
shown above it.
Conceptual ranking

White
Negro
Indian
Chinese
Japanese
Filipino
Korean
Other races

138

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Table 1
White_-Negro_-Indian_Chinese __
Japanese_Filipino_Korean_Other races_--

Table 2
White ___
Negro__Other races:
Indian_Chinese_Japanese_Filipino_.etc.

Table
White_-Nonwhite ___
Negro__Indian_
Chinese_Japanese __
Filipino ___
etc.

SEC.

10-A

STUB INDENTION

PAR.

1003

1003. Basic principles of comparative indent.-With the exception
of the "Total" or "reverse" indent (see par. 1004), t h e amount of
indent given a particular line is in inverse order to its level in the
classification as presented. Lines of superior classification are generally presented (in Bureau of the Census tables) above lines of inferior
classification, with the totals and subtotals appearing at t h e top of
the listing of component parts.
a. Lines of inferior classification are indented more t h a n lines of superior classification; a component part is normally indented under its total, except where
"reverse" indent (see par. 1004) is used for line captions of group totals.
b. Lines of coordinate classification are given identical indent within a given
table block.
c. Lines of superior classification are given less indent t h a n lines of inferior
classification, except in case of "reverse" indent used for line captions of group
totals.

1004. "Reverse" versus "flush" indent for group totals and subtotals.-Either of these two types of indent may be used for group
totals and subtotals. Although t h e "flush total" (flush to the left
margin) conforms strictly to t h e principle of comparative indent, a
common practice is to indent t h e total line and to run the first sublevel
of detail flush. Advantages of each method and factors determining
t h e selection are indicated in paragraphs 1005 and 1006, below.
Total with ''reverse" indent:
Total_-White_
Nonwhite_--

Total with "flush" indent:
TotaL
White_-Nonwhite _

1005. Indented group totals.-For normal use, indention of group
totals is recommended. Several advantages justify this departure
from the normal rules for comparative indent. Note, however, t h a t
a space break is essential above each such caption. (See par. 1005d,
below.)
a. Permits use of flush position for description of first level of subentries.The user is more interested in these t h a n in any emphasis upon t h e presence of
t h e group totals since he tends to assume t h a t t h e major group totals will be there
if he wants them.
b. Permits better definition of the left-hand margin of the table.-Improved
margin definition is achieved by placing t h e more frequently recurring classification-level in t h e flush position. This is more apparent in long listings t h a n in
short ones. Obviously, if t h e grand total caption is run flush, then only t h a t one
line caption on the entire page will be flush under normal indent rules; t h e remainder must be indented. A "ragged edge" at t h e left is not conducive to
easy reading or to grasping of comparative indent.
c. Permits better use of stub width when subentry captions are l o n g . - U s u a l l y
t h e line captions for subentries are longer t h a n t h e line caption for t h e total line.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

139

PAR.

1005

T H E STUB : CONSTRUCTION AIDS

SEC.

1)-A

By indenting t h e total line and running t h e subentries flush, t h e number of overruns for subentries is decreased, particularly in detailed classifications involving
multiple levels.
Example A-1. Wastes space and defines
left edge poorly:
Exports-

Crude materials_Foodstuffs-Semimanufactures_-Finished manufactures-

Imports __

Crude materials_Foodstuffs-Semimanufactures_-Finished manufactures __

Example A-2. Saves 1 pica (4 spaces) and
defines left edge better:
Exports__

Crude materials_Foodstuffs-Semimanufactures_-Finished manufactures_Imports_-

Crude materials __
Foodstuffs-Semimanufactures_-Finished manufactures_-

d. Space break essential to avoid appearance of a s u b e n t r y . - A space break,
such as a blank line, is desirable above a n y total-line caption given a reverse indent.
Otherwise t h e entry is likely to be mistaken for a subentry under t h e last line of
t h e preceding group. The space break is essential in typewriter-offset work,
and in letterpress composition where t h e subtotal lines are not in boldface.
(Even where boldface subtotals are used, t h e space break is helpful in this situation.) If no such space break is provided, t h e reverse indent should not be used.
Example B-1. Right (space above "Chinese"):
All classes_-White_Negro_Other races_Chinese, total_-Native_
Foreign born_-

Example B-2. Wrong (no space break):
All classes _
White_-Negro __
Other races __
Chinese, totaL __
Native_Foreign born_-

1006. The flush total.-This method has distinct advantages in
certain instances, as follows:
a. May clarify relationship between total and its immediate subentries.Use the flush total where serious uncertainty may arise in the reader's mind concerning this relationship. Paragraph 1006c illustrates such an instance. However, where this relationship is obvious, no real advantage accruing from this
factor is found in using the flush total.
b. Permits better use of stub width when caption is l o n g . - S t u b width is saved
by t h e flush total if t h e caption for t h e total line is longer t h a n t h a t for the subentries. Even here, however, overrunning t h e total caption may be preferable to
use of t h e flush total if t h e increased depth is no object. The following examples
are numbered from t h e widest to the narrowest in s t u b width.
Example A-1:
Persons 14 years old and over_

14 years __
15 years_16 years __

Example A-2:
Persons 14 years old and over_.
14 years __
15 years_16 years_-

Example A-3:
Persons 14 years
old and over_
14 years
15 years __
16 years_-

Example A-4:
Persons 14 years
old and over.
14 years_15 years_16 years_-

c. Useful in clarifying relationship where only a few items have subentries.Here t h e margin is more closely defined by use of the flush total t h a n where t h e
indented total is used. In example B - 1 the danger t h a t t h e last subentry of t h e

140

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

SEC.10-A

STUB

INDENTION

PAR.

1006

previous block will be mistaken for a subtotal for t h e following group of major
entries is lessened considerably, as compared with example B-2. In example
B-3, space breaks are used, coupled with all cap lines in typewriter-offset. T h e
result is still confusing and, in addition, wastes vertical space.
Example B-1. Preferred:
Example B-2. Confusing:
(Letterpress or typewriter)
(Letterpress or typewriter)
Dauphin_
Dauphin_White_White_-Nonwhite 1__
Nonwhite.Davis_
Davis_Dayton__
Dayton__Ebert_Ebert_Ecorah_Ecorah_White_White_-Nonwhite 1 _ _
Nonwhite_Evanston _
Evanston_-Farrell __
Farrell_1 T h i s looks like a subtotal for the following indented lines.
Ezample B-3. Confusing and wastes vertical space

a. Letterpress:
Dauphin_White_Nonwhite_Davis-Dayton_Ebert_Ecorah_White_-Nonwhite_Evanston_-Farrell_-

b. Typewriter (caps undesirable):
DAUPHIN . . . . . • .
White
.
Nonwhite. . . . . . . . . •

DAVIS . . . . . . . • ,
DAYTON. . . . . . • •
EBERT . • . • . • • • ,
ECORAH . . • . • • ,.
White
Nonwhite

,,,.
,.

EVANSTON.. ,, .•
FARRELL. . . . . ,,

1007. Criteria for selection of indent method. Following are rules
of thumb which may be used in selecting the method to be employed.
In each case, the entire picture needs to be taken into account. With
everything else equal, however, adherence to these rules will result in
a satisfactory, though not necessarily the best, result.
a. Use indented total(1) Where only a grand total is shown (no group totals appear).
(2) Where the group-total designations are consistently shorter t h a n their
subentries.
(3) Where a great many group totals appear of coordinate classification, each
of which carries subentries.
(4) With a series of coordinate blocks of three or more lines each.
b. Use flush total(1) Where a great many group totals appear of coordinate classification but
only a few carry subentries.
(2) Where the group-total designations are consistently longer t h a n their
subentries.
(3) With a series of coordinate blocks of less t h a n three lines each.
c. Grand total l i n e s . - A grand total entry may be given a " t o t a l indent" even
though t h e major group totals under it are run flush. T h e contrary is also true.
d. Consistency of total i n d e n t . - E v e n for " t o t a l lines" representing coordinate
items, it is not essential that all such total-line captions be indented equally on t h e
page. It is desirable, however, when other construction aspects permit.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

141

PAR.

1008

THE

STUB : CONSTRUCTION AIDS

sEc.

10-A

1008. Effect of vertical spacing on indent s c h e m e . - A n y kind of
indent scheme leans heavily for effectiveness upon proper vertical
spacing of the line captions and heads. In particular, an indented
total must always carry a space break above it to avoid confusion.
Where space is left below an indented total for insertion of a horizontal
rule across the field of the table, at least the same space must be left
clear above it!
Headings and total-line captions should always tend to "sit down"
on the material to which they relate. T h e following rules should be
adhered to rigidly.
a. An indented line caption for a total line must always have clear space above
it.
b. If space is provided below a total-line caption, at least the same space must be
left clear above i t ; preferably the space above should be greater t h a n t h e space
below.
Example A - 2 . Confusing:

Example A-1. Acceptable:
Urban, t o t a l
White
Negro
Other

.
.
.
Rural, total

White
Negro
Other

Urban,

.

.
.
.

Rural,
White
Negro
Other

races
White
Negro
Other

.
.
.

Example B-2. Confusing:
All

races
White
Negro
Other

All

races
White
Negro
Other

.
.
.
.

RURAL
All

total

.
.
.
.

URBAN

URBAN
races
White
Negro
Other

.

.

Example B-1. Acceptable:
All

total

White
Negro
Other

.
.
.
.

.
.
.
.

RURAI
.
.
.
.

1009. Indent spacing: Basic rules.-The following indent rules
are listed in order of comparative level. The "flush" position is
included in order to make a complete picture. All indents are figured
from the "flush" position. For illustration of the various indent
possibilities, see fig. 16, page 143.
Rule 1. Total or "reverse" indent (where u s e d ) . - I n d e n t 3 ems,
letterpress; 6 typewriter spaces, offset. (In a table tight in width, the
total indent may be decreased to 2 ems, letterpress; 4 typewriter
spaces, offset.) Two subsidiary rules are basic:
a. Where several total-line captions appear, each comprised of "Total,"
standing alone, it is not essential t h a t they be given uniform indentions throughout
t h e page. Specifically-

142

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

SEC.

10-A

STUB INDENTION

FIG.

16

F I G U R E 1 6 . - I N D E N T I O N S , W I T H A N D W I T H O U T C O N F L I C T S AT V A R I O U S

LEvvLs (See pars. 1009-1010)
[Each square ( ) represents 1 em of indent in letterpress, 2 typewriter spaces in typewriter-offset]

Rule
No.

Line
No.
1
2
3
4

5
6

7
8
9
10
11
12

13
14

15
16

17
18
19
20

21
22
23
24
25
26

27
28

29

30
31

Total_-

1

Flush caption_
Flush caption: I t s overrun causes no conflict if followed by anflush caption__
Flush caption __

other

Flush caption: Where followed by a caption with a normal 2-em
indent, no overrun conflict will occur_
indent_-Primary indent: I t s overrun will not conflict unless it is
by a caption with secondary indent_-

primary
Primary
followed
Primaryindent: Where followed by a subentry with secondaryindent, avoid conflict by indenting the overrun 1 em
more_-Secondaryindent_-Secondary i n d e n t _

Primary indent: Where followed by a total caption,
overrunwill conflict unless set in 1 em more_
Total_Flush caption_

2
5
3
3
5

3
6
4
6

1

Primary
Secondary
followed
Secondary
Where followed by a subentry, avoid
conflictindent:
by indenting t h e overrun 1 em more_
Subordinate
indent_-Subordinate indent_

to the same rules as other

Flush caption __
Flush caption_
indent_-_
Primary indent_indent__
Secondary indent: An overrun here, followed by a total
presents a dual complicaton. First, since
t h e basic indent conflicts with the total indent, the
total caption (below) is moved 1 em to the right
(rule 1-b). In the new position, however, it conflicts
with t h e overrun indent on the line above; hence, the
overrun also is moved to the right (rule 6)

Primary
Secondary
caption,

O Total_-

2
3
4

4
5

4
1

5
2
2
3

4
6
6
6
6
6

6

lb
2

Flush caption_
Flush caption_


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

2
5
2

the

Flush caption_
indent_Primary indent_indent__
Secondary indent: Conflict does not occur except when
by a subentry_

Total:
I t s overrun is subject
Oline captions __

2

2

143

PAR.

1009

THE

S T U B : CONSTRUCTION AIDS

SEC.

10-A

b. If t h e basic indent of t h e preceding line caption (not its overrun) interferes,
"Total" is indented an additional 1 em, letterpress (2 spaces, offset), to avoid
conflict.

Note: Where the indent of an overrun of the preceding line caption interferes,
t h e overrun is indented 1 em (2 spaces) more; t h e "Total" indent is not changed.
(See rule 6, par. 1010.)

Rule 2. Flush position.- No indent, flush at left margin.
a. If a line or tracer number, or a group number, appears at t h e left of t h e stub,
allow 1 em in letterpress (2 typewriter spaces in offset) between t h e column rule
and t h e flush position in s t u b , not including t h e space for t h e column rule.

Rule 3. Primary indent.- Indent 2 ems, letterpress; 4 typewriter
spaces, offset. In a tight table, this may be reduced to 1 em, letterpress; 2 typewriter spaces, offset.
Rule 4. Secondary and following subordinate indents.-For each
level, indent an additional 1 em, letterpress; 2 typewriter spaces,
offset.
Example A-2. Table tight in width:
Total indent _
Flush position _
Primary indent__Secondary i n d e n t - Subordinate indent _

Example A-1. Normal indent:
Total indent_-Flush position_-Primary indent_-Secondary indent_-Subordinate indent _

1010. Indent for overruns.-The above rules are for basic indent
and are based on the assumption that each line caption occupies one
line only. This indent scheme must be maintained rigidly if the
pattern is to fulfill its objective. Where overruns "conflict" with
the basic indent of the following caption, the adjustment is always
made in the indent of the overrun, never in the basic indent.
Overrun defined.-That portion of a caption that extends beyond the
first line.
Example of cverrun:
This caption is not overrun_-

overun

This caption is long enough to require
this overrun_-

Observance of the following rules will eliminate most indent problems arising from the presence of overruns. These rules are supplementary to rules 1 through 4, listed in par. 1009, above.
Rule 5.-Overruns are normally given a hanging indention of 1 em,
letterpress (2 typewriter spaces, offset), more than the basic indention
of the caption of which they are a part.
Example A-1. Right:
Insecticides, fungicides, and related
industrial and household chemical
compounds_-

Example A-2. Wrong:
Insecticides, fungicides, and related
industrial and household chemical
compounds--

Rule 6 . - I f an indent conflict results; that is, if the indent of an
overrun will be identical with the basic indent of the line caption

144

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEC.

10-A

STUB INDENTION

PAR.

1010

immediately following, the overrun is assigned an additional indent of
1 em, letterpress; 2 typewriter spaces, offset.
Note: Compare with rule 1 - b , where conflict of basic indent is involved and
note t h a t conflicts are always resolved by a method which provides minimum
disturbance to t h e classification function of the basic indent pattern.

Sec. 10-B. Leaders (1011-1018)

1011. General definition.-A row of tiny and closely placed dashes
(dash leaders), or a row of periods (typewriter-offset) lined up at
the bottom of the data-line caption and extending from it across the
entire width of t h e table, except where interrupted by a column rule
or an informational cell-entry.
-I

Example: White_--

Yes [ 1930 [---

--I

25

l

()

1---

--1

6

I

10. 6

1012. Purpose and function.-The purpose and function of t h e
leader line is twofold:
a. It carries the eye from one point to another along a given line; that is, it
maintains horizontal legibility.
b. (In the field of the table) By filling otherwise empty cells, it makes it clear
that the lack of informational entries is intentional and not caused by an oversight.
Examples:

A-1. With leaders

Total_
12

9. 7

1925

No

Other races__-

2

10

Yes

White_Negro_-

]. 8

A-2. Without leaders
Total

10

White

Yes

Other races

No

Negro

12
1925

(2)

9. 7
2

I. 8

1013. Stub leaders defined.-That portion of a leader line which
appears in the stub. It bridges the gap between t h e data-line caption
and the field of t h e table. T h e stub leader extends from the end of
t h e caption to the column rule which defines t h e right-hand edge of
the stub. Exception: For extension of leaders from t h e top line of an
overrun caption, see paragraph 1014, below.
For mechanical reasons, leader practice differs in letterpress and
offset work, as follows:
a. Letterpress.-Normal rules prevail. T h e dash leader is used, extended the
full width of the table except where interrupted by column rules or superseded
(in cells) by informational entries. In the cells, the leaders extend across the
full cell width.
Example A:
White_Negro __

25


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

16

2

145

sEc. 10-B

THE STUB : CONSTRUCTION AIDS

PAR.1013

b. Offset (typewriter).-A row of periods is used as t h e substitute for t h e letterpress leader dashes. In the cells, three periods (. . . ) are used to reduce the
number of typing strokes for filling otherwise blank cells. They are lined up
at the right, not centered.
Example B:
White
Negro

25

.

J

23.78

16

2

4

2,378

1014. Leaders and overruns.-Normally, leaders are placed against
the end of the line caption, irrespective of the presence of overruns.
"Overrun" defined.-That portion of a caption which extends beyond
the first line. That is, if a line-caption takes up three lines in the stub,
the second and third lines constitute the "overrun."
Example:
overrun

Musicians and teachers
25
of music_--

15

Two special cases arise, however, where the leaders are omitted
from the stub (though not from the field) because the informational
entries are run against the top line of the caption, even though an
overrun is present.
a. Tables with tracer numbers.-Where tracer numbers appear on t h e left and
right side of each page (or on left and right of parallel pages) and a line caption
overruns, omit the s t u b leaders for that caption. R u n the information against
its top line, and end the overrun with a period.
Reason: Tracer numbers should line up horizontally. For better line identification, the left tracer number should always appear at the beginning of a line caption (first or top line). Placement of the cell entries also on this line keeps the
tracer numbers lined up properly across the page (or facing pages) and simplifies
reading across. In such a case stub leaders are inappropriate.
Example A-1. Standard practice:
("X" denotes line number columns):
X

10
11

Personal services_
Miscellaneous industries
and services.

X

10

15
15

40
60

70

10

15

40

50

10

15

60

70

10

50

10
11

Example A-2. Not recommended:
X
10
11

Personal services_Miscellaneous industries
and services_-

X
10
11

b. Tables with two or more reader c o l u m n s . - I f only one "reader" column
(column containing a word or phrase) appears in the field of the table, r u n the
leaders (and informational entries) from the bottom line of an overrun as usual
unless tracer numbers are present. A standard date column is not considered
to be a reader column. If tracer numbers appear, run the information against
the top line and end the overrun with a period.
If two or more reader columns appear in the field, and a line-caption overruns,
omit the s t u b leaders for t h a t line caption, and end the overrun with a period
(same rule as for tracer numbers). T h i s is true irrespective of overruns in either
or all of the reader columns.

146

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEc. 10-B

LEADERS

Right:
Australia, Commonwealth of.
Wrong:
.Australia, Commonwealth of __

PAR.

Governor General_.

Parliament: Senate,
House of Representatives.

Dominion; State.

Governor General_-

Parliament: Senate,
House of Representatives.

Dominion; State.

Right:
Cotton fibers in raw
state_-

June 18

75, 124

81, 251

If overrun in
cell.

15

Wrong (one reader column; line up at bottom, not at top):
Cotton fibers in raw Mar. 17 42, 648 98, 250 Single reader
state.
column.

15

1014

Wrong (two reader columns; line up at top, not at bottom):
Cotton fibers in raw
state_

Yes_-

42, 4es ] os, 280

Second reader
column.

15

1015. Period omitted preceding leaders.-Periods are invariably
omitted preceding leaders, even though the period in question terminates an abbreviation.
a. Letterpress.-Here, t h e error, if any, is immediately apparent since t h e
period is distinctly different from t h e opening dash in t h e leader.
Right: Laborers, n. e. c __

Wrong: Laborers, n. e. c._-

b. Offset.-Since a row of typewriter periods is substituted for dash leaders,
t h e erroneous inclusion of a terminal period would be concealed were it not for
t h e fact t h a t the error tends to be compounded by leaving a space between it and
t h e first leader dot (period).
Right: Laborers, n . e . c

.

Wrong: Laborers, n . e . c .

1016. Leaders run flush to caption.-Leave no space between the
last character of the line caption and the beginning of t h e leader line.
Right: W h i t e

.

Wrong: W h i t e

.

1017. Colon or dash (read-in) captions never take leaders.-If the
caption ends in a colon or a dash, no informational entries should
appear opposite it in t h e cells and the caption does not take leaders.
If informational entries are to appear against such a line, delete t h e
colon or dash and substitute leaders, changing t h e wording as necessary.
Example A:
A-1. Right:
Softwoods:
Balsam_-Cedar_Cypress_Example B:
B-1. Right:
By race:
White_Negro_-

A-3. Wrong:
Softwoods:_
Balsam _
Cedar_Cypress_-

B-2. Right:

B-3. Wrong:

All classes __
White_Negro_
Other __

Other_-

77655449

A-2. Right:
Softwoods_Balsam_-Cedar __
Cypress_-

11


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

By race:White _
Negro
Other __

147

PAR.

1017

THE

STUB : CONSTRUCTION AIDS

sEC.

10-B

Example C:
C-1. Right:

Families havingNo children __
1 child-2 children_--

C-2. Right:

C-8. Wrong:
Families havingNo children_1 child _
2 children--

Total families_-Having no children.
Having 1 child __
Having 2 children_-

1018. Stub leaders and unit-indicators.-This problem is discussed in paragraph 963 in the material on unit-indicators.
Sec. 10-C. Space Breaks: Analytical and Reader (1021-1030)

1021. Definition.-The space break is an open space or blank
"line" inserted at periodic intervals to break up solid blocks of line
captions (and their data). Extending across the full width of the
page, it should be inserted wherever more than 15 lines (at most)
would otherwise appear in a solid group.
Example A - 1 . Without reader breaks:

Alabama_Arizona_
Arkansas_California_Colorado __
Connecticut_Delaware_-District of Columbia __
Florida_-Georgia__Idaho
Illinois_Indiana_Iowa_-

. 93,787
. 17,600
40, 910
. 97,510
. 51,503
. 37,915
. 11,535
. 23,834
. 71,526
. 40,770
. 16,129
_ 87, 176
. 46,969
. 77,501

Example A - 2 . With reader breaks:

Alabama_Arizona.
Arkansas_California_Colorado_-

. 93,787
17, 600
. 40,910
. 97,510
. 51,503

Connecticut_Delaware_-District of Columbia_Florida_-Georgia_-

. 37,915
. 11,535
23, 834
. 71,526
. 40,770

Idaho_Illinois_-

. 16,129
. 87,176

1022. Purpose and function.-Improves horizontal legibility by
grouping the entries:
a. Provides guide "lines" across the p a g e . - T h e reader's eye, when tracing a
line of entries, can more readily note the relative position of the line in a small
block t h a n in a large block.
b. Relieves monotony and eye-strain caused by unrelieved masses of :figures,
particularly where the table is tight and any kind of white space is at a premium.

1023. Analytical versus reader breaks.-The reader break is often
confused with the analytical space break since each consists of an open
space or blank line extending across the page.
a. The reader break is purely mechanical in nature. I t s insertion depends
strictly upon the need for breaking up large masses of data remaining after analytical design has been completed.
b. The analytical space break is a means of setting off lines or groups of lines
in terms of analytical relationship. A function of original design, it has little to
do with mechanics as such.

148

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEc. 10-C

SPACE BREAKS

Example A-1. Reader breaks:
Apples__
Cherries __
Peaches_Apricots__
Pears _
Plums and prunes_Olives_Grapes_
Almonds_Filberts and hazelnuts_--

PAR.

1023

Example A-2. Analytical breaks:1
Apples__
Cherries_Peaches __
Apricots-Pears_-Plums and prunes __
Olives __
Grapes__
Almonds.
Filberts and hazelnuts.-Pecans__
Tung__Walnuts--

Pecans __
Tung__Walnuts_-Oranges_
Oranges_Grapefruit _
Grapefruit-Lemons __
Lemons_1 If desired, a reader break might be inserted after "Pears"; that is, 5 lines down.

1024. Insertion of analytical breaks.-The erroneous belief that
insertion of all space breaks is t h e duty of the "mark-up man" (person
marking up copy for t h e printer or reproduction typist) leads to complaints when the resultant grouping "doesn't make sense," or where
it might have been improved "if the breaks had been inserted properly."
Usually, t h e difficulty is caused by failure of t h e table-designer to
provide analytical breaks at the proper places.
a. Analytical space b r e a k s . - A n a l y t i c a l grouping of t h e d a t a by means of
space breaks is as much a part of analytical design as determination of line-order.
In general, space breaks of this t y p e should be inserted before t h e table is sent to
reproduction typing or to the printer. T h e table-designer will profit by observing the following procedure:
(1) S p e c i f y g r o u p i n g of lines w h e r e significant a n a l y t i c a l p u r p o s e s
will be s e r v e d . - P r e f e r a b l y , t h e breaks should appear on t h e posting sheets,
since they are helpful in posting, computing, and checking.
(2) Review s t u b for solid m a s s e s r e m a i n i n g . - N o t e , particularly, a n y
instance where 15 or more lines appear without a break. If these are passed
by t h e designer, it is-probable t h a t the mark-up man will break up each mass by
arbitrary insertion of reader breaks.
(3) D e t e r m i n e a c c e p t a b i l i t y of reader b r e a k s . - I f t h e material is such
t h a t no analytical advantage can be gained by organized grouping of lines by
use of minor analytical breaks, t h e material may be let pass; provided, t h a t
t h e designer is willing to accept reader breaks inserted at arbitrary points.
If not, or if a logical grouping in terms of meaning a n d use of t h e data presents
itself, minor analytical breaks should be indicated. See example under paragraph 1023 for improvement achieved by use of planned analytical breaks.
Note: T h e analyst or table-designer should not specify amount of space; he
should merely indicate t h a t a space break is desired, thus:
Example A-1. When marked like this:
Rhode Island _
Connecticut __
#>New Y o r k _ New Jersey __
Pennsylvania_--

Example A-2. It will be set like this:
Rhode Island __
Connecticut_-New York_New Jersey__
Pennsylvania_-

* Exact amount of space may vary.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

149

PAR.

1024

T H E S T U B : CONSTRUCTION AIDS

sEc. 10-C

b. Reader b r e a k s . - T h e mark-up man concerns himself (1) with t h e mass
effect after the analyst has completed his task, and (2) with general page-fitting
requirements which may demand "spacing out" which cannot be anticipated by
t h e analyst. Any large mass of data remaining after insertion of analytical
breaks called for on the copy may be broken up by the mark-up man according
to mechanical rules. (See par. 1025, below.)

1025. Reader-break insertion: General principles.-Following is a
list of more or less standard practices in inserting reader breaks.
a. Required in a n y solid group of more t h a n 15 lines.
b. Subgroups thus created should not be less t h a n five lines each. However,
t h e last subgroup (or last two subgroups) in a column or list may contain as few
as three lines, but not fewer, in order to equalize columns or pages. (Subgroups
caused by analytical breaks do not involve these limitations since the grouping is
based on the meaning of the data. See par. 1027b.)
c. Should not be inserted between an overrun and its parent entry.
d. Should not be inserted where t h e indent shifts from a higher to a lower level.
e. Should not be inserted where they might lead to faulty analytical groupings.
Although this is the primary concern of t h e analyst, the mark-up man avoids
errors of this t y p e where possible within t h e limits of his time and training.
f. Placement should not be influenced by possible use as counting aids; t h a t is,
reader breaks are not designed to aid counting of items by arranging t h e data in
standard groups as such, even though this is often considered a byproduct.
It follows t h a t (1) If a 5- or 10-line basic group is selected, t h e pattern need not be maintained rigidly merely to facilitate item-counting.
(2) If facilitating item-counting is an objective in table design, a line or group
number scheme should be adopted, or the analyst should specify special breaks
accordingly.

1026. Reader-break insertion: In lists of coordinate i t e m s . - I n
lists of independent or coordinate entries, or in blocks of data arranged
alphabetically or in obvious rank order, it is customary to insert a
reader break every 5 or 10 lines. "Counting off" of lines starts over
at the top of each stub column (in fractional-measure tables), at the
top of a page, and at every major group total shown.
a. Consistency of b r e a k . - I n general, if either a 5-line or a 10-line break is used
at t h e beginning of a column, it should be used throughout t h a t column, except
at the bottom for equalizing with other columns. Although it may be desirable,
it is not essential t h a t the same basic grouping be used throughout the table. In
fractional-measure tables, reader breaks in parallel portions of t h e table need not
line up with each other.
b. Left-over g r o u p s . - I f t h e total number of lines in a listing is not a multiple
of t h e number of lines in t h e standard adopted for the reader-break groups, t h e
odd group (or groups) appears at the bottom of t h e column, page, or listing, not at
t h e top. This means that(1) T h e first group at t h e top of a given column or page should be a standard
group, whether or not t h e preceding group (at bottom of preceding column or
page) is "complete."

150

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEc. 10-C

SPACE BREAKS

PAR.

1026

(2) In order that columns (and pages) may be even, t h e printer is at liberty
to take out or insert reader breaks, readjusting the groupings accordingly.
Usually, such adjustment is confined to the last group (or last two groups) in
each column, but not always.
(3) Where less t h a n three lines will fall in the last group, they should be
combined with the preceding group or groups. After such combination, the
consolidated group may be broken in turn into smaller groups with a lesser or
greater line count t h a n the standard groups above.

1027. Reader-break insertion: In lists of noncoordinate items.Placement of reader breaks in lists of noncoordinate items requires
use of considerable judgment to avoid bad grouping. In general, the
table-designer would be well-advised, in such cases, to insert minor
analytical breaks, drawing upon his knowledge of the meaning of the
data to obtain maximum clarity and usefulness of the result.
Basic problem: When line captions are given differential indent, at
what point will insertion of a space break enhance meaning and avoid
confused groupings?
The following list of general rules may prove helpful.
a. Standard groups not necessary.-Where t h e items are not coordinate, no
standard grouping of the 5-line or 10-line type is necessary; frequently it is not
practicable. However, wherever any portion of the stub facilitates such groupings,
their use is advisable for that portion.
b. Maximum and minimum lines per g r o u p . - I n general, t h e same rules hold
as for coordinate listings. T h a t is, break up all masses of 15 line captions or more.
Do not show groups of less t h a n 3 lines.
In practice, if the line relationships are sufficiently distinctive, two lines, or
even one line, may be set off. However, such small groupings should be reserved
for such cases as(1) Tables with blocks of frequency data which end with one or two lines
of derived figures, such as medians or averages; or
(2) Tables with blocks of standard intervals followed by one or two overlapping intervals. Example: The age break "21 years old and over" appearing
at the end of a standard 5-year group listing.
Example A. Right:
Under 4.0 percent__4.0 percent_-4.1 to 4.4 percent__
4.5 percent __
4.6 to 4.9 percent__
5.0 percent_.

Example B. Right:
15 to 19 years__20 to 24 years _
25 to 29 years__30 to 34 years.35 to 39 years_-40 to 44 years__-

5.1 to 5.4 percent_-5.5 percent_-5.6 to 5.9 percent__6.0 percent__6.1 or more_-

45 to 49 years__50 to 54 years _
55 to 59 years__60 to 64 years_-65 and over__-

Average rate (percent)_

21 years and over__-

c. Insert where indent shifts to a higher, not to a lower l e v e l . - T h e end of a
block of subentries is usually a sound location for a space break; the beginning of
such a block is usually a bad location.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

151

;

PAR.

1027

sEc. 10-C

T H E STUB : CONSTRUCTION AIDS

Example C-1. Right:
15 to 19 years__-

White_-Negro_--

20 to 24 years_--

White_-Negro _

25 to 34 years.-

White_-Negro_--

Example C-2. Wastes space:

E i a m p l e C-3. Wrong:

15 to 19 years __

15 to 19 years

_

White _
Negro_-20 to 24 years_-

White _
Negro_-20 to 24 years__-

White_-Negro_--

White_-Negro_--

25 to 34 years_.

25 to 34 years_-

White_-Negro_--

White _
Negro _

d. Insert above, not below, subtotal e n t r i e s . - A special case of 1027c, above.
Space breaks, if inserted, are placed above, not below, a subtotal entry. T h a t is,
a space break should not separate a subtotal entry from its first three subentries.
This rule bolds good even though space breaks also may be inserted within the
subentry block. Furthermore, the fact t h a t a space break appears below the last
subentry is no reason to insert one just above the first subentry.
E x a m p l e D-1. Acceptable:

Under 5 years__-

5 to 14 years. _
15 to 24 years_25 to 34 years_-

35 to 44 years_35 years_-36 years_-37 years_-38 years__39 years__40
41
42
43
44

years _
years-years _
years _
years__-

45 to 54 years.55 to 64 years __

65 to 74 years_-

75 to 84 years_85 years and over_-

Example D-2. Not acceptable:

Under 5 years_5 to 14 years_-15 to 24 years_-25 to 34 years _
35 to 44 years__35
36
37
38
39

yearsyears_years_years_years_-

40 years_41 years_42 years_43 years_44 years_45 to 54 years_--

55 to 64 years _
65 to 74 years_75 to 84 years _
85 years and over_-

e. Avoid space breaks after colon or dash l i n e s . - T h e insertion of a space
break after a colon or dash line leads to confusion. This is particularly true of the
dash line since the "read in" is thereby interrupted. If a space break is needed,
place it above, not below, such a caption.

1028. Inclusion of total and subtotal lines when counting off lines
for reader-break groups.-This is largely a matter of choice. In
generala. Include in count if no horizontal total or subtotal rule is u s e d . - H e r e , the
total or subtotal caption is running solid with its subentries and groups naturally
with them.
b. Exclude from count if horizontal total or subtotal (underscore) rule app e a r s . - T h e slight gap left in t h e s t u b by the use of total rules in the field tends to
separate t h e total or subtotal caption from t h e first subentry. If t h e total or
s u b t o t a l caption is included as a member of t h e first group, when counting off,
t h e reader's eye tends to note t h a t t h e first group seems short. To t h a t extent
t h e user is distracted from his purpose which is t h e meaning of t h e data, not its
mechanics.
c. Disregard question of boldface captions.-Use of boldface captions, as
such, has no effect on the above suggestions.

152

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEc. 10-C

SPACE BREAKS

PAR.

1029

1029. Overrun captions versus reader breaks.-Overrun captions
result in one or more "blank lines" appearing across t h e f i e l d - b u t
not in t h e stub.
Basic questions: Cannot such lines be considered as equivalent to
reader breaks? If not, how is the count made in establishing readerbreak groups where overruns occur? Should a break be inserted at
every "so-many" captions or at every "so-many" lines?
Solution: No completely satisfactory solution is available. In
general, where overruns occur, the need for reader breaks is reduced
since the overruns serve t h e purpose in the field. However, reader
breaks are still needed in the stub to aid the reader in keeping his
place there or in locating a given line caption.
General principles.-The following general principles may serve as a
guide, keeping in mind t h a t reader breaks are inserted as a positive
reading aid to the user, not to establish a physical pattern, as such.
a. Where a few overruns occur.-Insert reader breaks as usual, ignoring t h e
overruns when counting off. Count captions, not lines.
b. Where overruns are frequent.-The basic objective is to avoid solid blocks
in either stub or field. The following steps are, in effect, performed simultaneously:
(1) E x a m i n e s t u b . - I n s e r t reader breaks at least every 20 lines. Count
lines occupied, not captions. Twenty per group is acceptable where many lines
are overrun with consequent additional "breaks" in t h e field.
(2) Examine f i e l d . - A p p l y same rules as where overruns are few. T h a t is,
solid blocks of 15 entry lines or more should be relieved by inserting reader
breaks in t h e stub, extending through the field.
(3) A d j u s t r e s u l t a n t c o m b i n a t i o n of breaks in s t u b to sensible p r o p o r t i o n s . - T h a t is, where rule 2 demands a break to relieve t h e field, a break
inserted according to rule 1 (to relieve the stub) may become unnecessary, or
might be better placed; and vice versa. In other words, apply common sense.
c. Where space is at a prerniurn.-Rely heavily upon t h e overrun lines to
break up the rows of figures. Insert reader breaks sparingly in the s t u b to allevia t e the worst of the otherwise solid blocks.
d. Caution: Never insert a reader break between an overrun and its parent
entry.

1030. Tie-in of stubs by standard placement of space breaks.Two general situations arise where the space break may be used to
ad vantage in helping t h e reader who is working from table to table.
a. Identical s t u b s . - W h e r e a series of tables is run with identical stubs, sound
presentation practice demands consistent placement of space breaks, stub to
stub. Two major reasons are: (1) Consistent placement is of distinct service
to the user; and (2) their presence on the initial worksheets will facilitate transcription and verification of the data.
It follows that placement of space breaks of this type are an integral part of
table design. Within limits, it should be possible, by good planning, to establish


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

153

PAR.1030

THE

STUB : CONSTRUCTION AIDS

10-C

sEc.

standard breaks for standard classifications which may be used from report to
report, as well as within a given report.
Example A. Variation in placement of s p a c e breaks between tables can be a serious nuisance:
Table 1

Table 2

All ages__Under 5 - 5 to 910 to 14
15 to 19-20 to
25 to
30 to
35 to
40 to

All ages __
Under 5 - 5 to 9 - -

10 to IL
15 to 1 9 - 20 to 2 4 - -

2429-34-39-44.

25 to
30 to
35 to
40 to
45 to

45 to 49-50 to 54-55 to 59-60 to 64-65 and over __

29 _
34 _
39--44-49.-

50 to 5 4 - 55 to 59 _
60 to 64 _
65 and over_-

In practice, two major limitations must be n o t e d :
(1) Standardized space breaks must be planned by t h e table-designer, not
left to t h e mark-up man who cannot take t h e time to analyze all similar s t u b s
before marking up the first; and
(2) Publication space requirements may make it necessary f o r the mark-up
man to add additional space breaks, or to reduce the number requested.
However, where standardized breaks are indicated, they will be followed
wherever mechanically possible. • Also, the mark-up man, and the reproduction-typing planner, will frequently contribute a little extra thought and effort
to well-planned and properly prepared "copy."
b. Interrelated stubs (not identical).-Even where a standard classification
scheme is involved, stubs may vary from table to table (sometimes within a table)
in terms of condensation or expansion. Mechanical counting off for reader breaks
may result in t h e space breaks falling at one set of points in s t u b A and at another
set in s t u b B. If they fall at t h e saine points in both stubs, t h e user (and t h e
transcribers and verifiers in t h e production work) will be aided in working from
one to t h e other.
For reasons of space, age is used in t h e examples. However, need for standardization increases with t h e use of stubs with long captions which are less easy to
read at a glance.
Example B-1. Nonstandardized breaks:.
Table 1
Table 2
Under 5 _
15 to 2 4 5 to 14
25 to 34-3 5 t o 4 4__
15 to 24.
25 to 3 4 - 45 to 54-55 to 6 4 65 to 74-75 to 8 4 85 and over_-

35 to 4445 to 54
55 to 6 4 - -

65 and over

Table S
15 to 19--20 to 2 4 - 25 to 29--30 to 3 4 - 35 to 39--40 to 44--45 to 4 9 - 50 to 5 4 - 55 to 59---

60 to 6 4 - 65 and o v e r - -

Table 4

Under 5 _
5 to 9 _
10 to 14 _
15 to 19 _
20 to 2 4 - 25 to
30 to
35 to
40 to
45 to

29--34-39-44.-49--

50 to 5 4 - 55 to 59--60 to 64--

65 and o v e r -

154

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEc. 10-C

SP ACE B R E A KS

Example B - 2 . Standardized breaks:
Table 1
Table 2

15 to 24-25 to 3 4 - 35 to 44-45 to 5 4 55 to 6 4 65 to 7 4 75 to 84 __
85 and over__

Under 5 _
5 to 1 4 - 15 to 2 4 - 25 to 34_-35 to 4 4 - 45 to 5 4 - 55 to 6 4 - 65 and over_-

PAR.

Table 3

15 to 19 _
20 to 2 4 - 25 to 2 9 - 30 to 3 4 - 35 to 39--40 to 4 4 - 45 to 49--50 to 5 4 - 55 to 59--60 to 64 _
65 and over_--

Note: The "standard break" adopted here
precedes age 45. The user not only can
locate corresponding lines or groups of lines
more readily, given the common reference point; also, at a glance,
he can grasp difference in stub detail above and below that point.

1030

Table 4

Under 5 _
5 to 9 _
10 to 1 4 - 15 to 19 _
20 to 2 4 - 25
30
35
40

to 2 9 - to 3 4 - to 39--to 44--

to 49--to 5 4 - to 59 _
to 6 4 - 65 and over_-

45
50
55
60

Sec. 10-D. Use of Boldface and Italic Type (1031-1033)
1031. G e n e r a l . - I n letterpress or varitype operations three kinds
of type are available for use in tabular work. Avoiding typographical
technicalities, these may be described as ordinary (roman) body
type, boldface (black letter), and italic. While theoretically the
italic may be either boldface or lightface, the boldface italic is rarely
used in tabular work of the Bureau of the Census. Materials in the
table are normally set in roman. T h e use of boldface or italic in the
stub is largely restricted to the specific situations outlined below.
In offset work, typed entirely by means of the ordinary typewriter,
neither boldface nor italic is available except in the case of preprinted
table forms. Instead of boldface or italic, reliance must be placed on
clarity of indent.
By "preprinted table forms" is meant forms in which the entire
table form (sometimes merely t h e stub) is letterpressed or varityped
in advance. T h e field (cells) is then filled in on t h e ordinary typewriter. Here, t h e use of boldface or italic, if used, must necessarily
be confined to the stub.
1032. Boldface.-Use of boldface is normally subject to the following restrictions. (See also par. 959.)
a. Limited largely to primary levels of center heads and to additive total
l i n e s . - W h e n used to indicate an additive total, not only the stub entry but also
numerical and other descriptive entries across the table are set in boldface. In
preprinted forms, t h e boldface must, by necessity, be confined to the stub.
b. Boldface not used to indicate coordinate level of e n t r y . - I f a s t u b entry line
of a given classification level is set in boldface, it does not mean t h a t all lines of
the same level also should be set in boldface. Only those lines are set in boldface which constitute the summation of the subordinate entries t h a t are presented
under them. Thus, if in a given classification additive subentries appear for
three out of five members, boldface is used for the three which constitute subtotals. It is not used for the other two.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

155

PAR.

1032

THE

S T U B : CONSTRUCTION AIDS

sEc.

10-D

c. M a y be accompanied by horizontal r u l e . - T h e use of boldface may be combined
with a horizontal rule running under t h e total, but this combination is reserved
for major group totals or for grand totals. For minor totals, t h e use of boldface
alone is usually sufficient. (See also par. 959c.) Note: In offset work where
boldface is not available, group totals are normally set off by a horizontal rule
running across the field but not through t h e stub.
Example A. Right (primary level of center
heads; see also fig. 14, p. 115):

E:ample B. Right (additive total lines):
ALL RACES
All ages__

ALL RACES
Male
All ages.
Under 45 years_45 a n d over_-Female
All ages_Under 45 years_.
45 and over.--

835

392

Under 15 years_15 to 24 years __
25 to 34 years_35 to 44 years __
45 to 64 years__
65 a n d over_WHITE
All ages_-

60
180
240
120
150
85

38
75
110
57
72
40

548

236

WHITE
Male
All ages
Under 45 years_45 and over__

Under 15 years_15 to 24 years __
25 to 34 years__
35 to 44 years __
45 to 64 years__
65 years and o v e r -

48
120
130
85
100
65

22
51
40
38
56
29

Example C-1. Right (used only for summation
ALL RACES
All ages __
U n d e r 15 years __
15 to 24 years_Single __
All other_25 to 34 years __
Single_All other __
35 to'44 years_
Single_All other_45 a n d over_--

Example C-2. Wrong (the " u n d e r 15" a n d
a n d over" lines should not be in boldface):

lines):
835

392

60
180
110
70
240
140

38
75
50
25
110
52

100

120

40
80
235

57
20
37
112

ALL RACES
All ages_Under 15 years__
15 to 24 years_.
Single_All o t h e r 25 to 34 years__
Single __
All other_
35 to 44 years_.
Single_All other_45 and over_-

835

392

60

38
75
50
25
110
52

180
110
70
240
140
100

45

58

120
40
80
235

57
20
37
112

1033. Italic.-The use of italic is limited largely to letterpress and
varitype work. At all times, italic is optional, not required, and
should be confined to cases where there is a genuine increase in clarity.
Within these limits, its use is generally restricted to the following:
a. Nonadditive items appearing within a group of additive i t e m s . - A t times it
is desirable to show, within a group of items adding to a total, one or more selected
single-line subentries. Again, one may show a line of derived entries or a class
overlapping two or more of t h e items presented. Italic is useful to set off such
isolated lines.
Example B. Acceptable:

Example A.

Right:
Colusa County __

Colusa township_-

Colusa City_--

Williams township_-

Williams town-Example C-1. Acceptable:
F a r m families __

Urban-farm--

Nonfarm families_--

156

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

98
50

23
48
8
289
8
423

Nevada_Urban ___

Percent-

R u r a l __

_ 91,058
34, 464
37. 8
_ 56,594

Example C-2. Wrong (full distribution shown):
Farm families_-

Urban-farm_
Rural-farm-

Nonfarm families__-

289
8
281
423

SEC.

10-D

BOLDF ACE A N D ITALIC TYPE

Example D-1. Preferable (space in-

stead of italic):
All ages__

100

Under 15 years_15 to 24 years.

20
50
20

25 to 34 years__
35 to 44 years_45 and over_-

21 and over_-

19

PAR.

Example D - 2 . Acceptable
tight table:
All ages_Under 15 years_15 to 24 years_25 to 34 years_35 to 44 years_45 and over.21 and over.-

1033

n a
100
20
50
20
5

19

b. Deduction or decrease entries.-ltalic is useful to identify an item which is
to be subtracted in the computation to reach the group total; to represent a
minus item, as such; or to identify a decrease item in a table of increases. Where
italic is thus used, t h e meaning must be made clear in t h e stub, headnote, or
footnote. In general, if a minus sign is used in the cell, t h e italic is superfluous.
Where italic is used to denote deductions or decreases, the s t u b caption is usually
roman since t h e presence of italic may vary from column to column.
Example E. Deduction item; table headnote reads "[Italic denotes excess of credits (deduct)]":
Total_

Natural resources_Transportation_Finance and commerceGeneral government_-

288

43

21

46
28
- 218

37

22
19

6
21

8

c. Unit-indicators.-Units of quantity may be set either in roman or italic
where they appear at the end of stub captions, but roman is preferable, especially
where the unit-indicators are brief. (See also pars. 961-966.)
Example F - 1 . Preferred:

__number __
_percent __
_median__acres__bushels__pounds_-


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Example F-2. Acceptable:

_number.-_percent..
_median_
_acres_
_bushels__pounds..

157

Chapter 11

THE STUB: STANDARD LISTINGS AND INTER-STUB
CLARIFICATION (1101-1144)
Sec. 11-A. Listing Areas (1101-1118)

1101. Standard census areas.-In the publications of the Bureau
of t h e Census, statistics are presented for virtually every type of areapolitical, social, economic, administrative, demographic, statistical,
etc. T h e method and order of listing of area designations have largely
been established by custom, frequently in considerable detail.
This manual discusses a few of the more important practices in
respect to certain types of areas used by two or more Subject Divisions 1 of the Bureau. These include Bureau of the Census regions
and divisions, States, counties, incorporated places, minor civil
divisions, metropolitan districts, census tracts, and urban-rural
areas.
Specialized types of areas, used largely by single Subject Divisions,
or used only in highly specialized fields, are not discussed in these
materials. These include such types as drainage basins, customs
districts, Forest Service regions, and cotton regions.
1102. United States.T h e term employed in listing the United
States as a center head or line caption varies as follows:
a. United S t a t e s . - T h e expression most commonly employed. Normally, it
signifies continental United S t a t e s 'only. It 'is 'used 'wherever the question of
possible inclusion of d a t a for territories and possessions is not likely to arise. T h e
definite article "The" is not used.
Example A-1. Right:

Example A-2, Wrong:

U N I T E D ST AT E S

T H E U N I T E D ST A T E S

United States_-

The United States_-

or

or

b. Continental United S t a t e s . - U s e d where it is necessary to emphasize t h a t
t h e given d a t a refer solely to continental United States; t h a t is, t h a t figures for
t h e territories and possessions are not included. "Continental United States"
does not include Alaska.
T h e definite article " T h e " is not used. T h e word "continental" is given an
nitial cap only where it begins t h e caption.
1

See footnote, par. 831.

158

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEc. 11-A

AREAS

Example B-1. Right:
C O N T I N E N T A L U N I T E D STATES
or
Continental United States_-or
In continental United States--

PAR.

1102

Example B-2. Wrong:
T H E C O N T I N E N T A L U N I T E D STATES
or
The Continental United States_or
In the Continental United States_-

c. United States, territories, and possessions.-Used where d a t a are included
for territories and possessions as well as for continental United States.
Where only the aggregate is shown, with no subentries indicating geographic
components, t h e full statement (United States, Territories, and possessions) is
essential where data for all these areas are included.
Where total-line captions are involved and (1) space is at a premium, and (2)
the geographic components are shown as subentries so the reader can see clearly
t h a t the Territories and possessions are included, the expression "United States,
aggregate" m a y be substituted to save space.
Where one or more, but not all, of the Territories and possessions are included,
it is better to use merely "Total" and to footnote (or headnote) to indicate clearly
t h e area covered, as "Includes Puerto Rico and Canal Zone" or "Comprises continental United States, Alaska, and Hawaii."

1103. Regions, divisions, and S t a t e s . - I n the Bureau of the Census
the standard stubs showing regions, divisions, and States embody
several fundamental features of geographic arrangement. (See
6 g . 17, p. 160.)
a. Geographic arrangement.-Since many statistics of t h e Bureau reflect strong
geographic differences, a standard arrangement of States is desirable which will
permit comparisons of d a t a for States within t h e same geographic area. T h e
standard arrangement for this purpose is t h a t shown in figure 17. T h e Nation
is first divided into three (or four) regions. The regions are divided, in turn, into
geographic divisions, nine in all, with each division composed of a specific group
of States.
(1) Order of l i s t i n g . - T h e list is arranged according to geographic progression. First, t h e regions are listed, starting with t h e northeast corner of
t h e Nation, and listing from north to south, then listing t h e next region to t h e
west. A similar scheme is used for listing component divisions under each
region, and for listing component States under each division.
(2) A l p h a b e t i c a l s c h e m e n o t e m p l o y e d for area s u b e n t r i e s . - T h e
standard geographic a n d alphabetical listings should not be mixed. Regions
and divisions are always listed geographically, never alphabetically. States
may be listed geographically or alphabetically. Where listed geographically,
States should be presented under divisional headings or subtotal captions.
Where States are listed alphabetically, figures should not be shown for regions
or divisions in that table.
Example A - 1 . Right:
N e w England_.

Maine_.
New Hampshire ___
Vermont.-Massachusetts_Rhode Island_-Connecticut_-

Example A-2. Wrong:i
N e w England_-

Connecticut ___
Maine_
Massachusetts-New Hampshire_Rhode Island_Vermont_-

Example A - 3 .
The North_-The South_The West_-Alabama_-Arizona_Arkansas_-

1 If the figures have a sufficient area pattern to warrant presenting divisional or regional subtotals, the
States also should be arranged geographically, not alphabetically.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

159

FIG.

17

THE

STUB : STANDARD LISTINGS

SEC.

11-A

F I G U R E 1 7 . - S T U B S FOR REGIONS, DIVISIONS, AND STATES

(See par. 1103)
A. 4-region s t u b (censussize page)
United States_-

Regions:
The Northeastern States __
The North Central States_-The South __
The West_-The Northeastern States:

New England _
Middle Atlantic _

The North Central States:

B. 3-region s t u b (census-size page)
United States __

Regions:
The North __
The South __
The West_-The North:

New England_-Middle Atlantic _
East North Central_-West North Central__-

East North Central_-West North Central _

The South:

South Atlantic __
East South Central_West South Central _

The West:

The South:

The West:

Mountain __
Pacific __

N e w England:

Maine
New Hampshire_Vermont_Massachusetts_Rhode Island __
Connecticut _

Middle Atlantic:

New York_New Jersey __
Pennsylvania __

East North Central:

Ohio __
Indiana __
Illinois __
Michigan __
Wisconsin __

W e s t North Central:

Minnesota __
Iowa __
MissourL
North D a k o t a - South Dakota_-Nebraska __
Kansas _

South Atlantic:

Delaware__Maryland_District of Columbia_-VirginiaWest Virginia __
North Carolina__South Carolina_-Georgia_Florida

East South Central:

Kentucky _
Tennessee__
Alabama __
Mississippi__

West South Central:

Arkansas_-Louisiana __
Oklahoma-Texas __
Mountain:

MontanaIdaho_
Wyoming __
Colorado __
New Mexico __
Arizona __
Utah.-Nevada __
Pacific:

Washington _
Oregon _
California _

160

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

South Atlantic_-East South CentraL
West South Central_Mountain __
Pacific_-

N e w England:

Maine.-New Hampshire_Vermont_Massachusetts_Rhode Island_
Connecticut_--

Middle Atlantic:

New York_New Jersey_
Pennsylvania_

East North Central:

Ohio __
Indiana __
Illinois _
Michigan
Wisconsin_--

West North Central:

Minnesota __
Iowa_
MissourL
North Dakota _
South Dakota_-Nebraska __
Kansas _

South Atlantic:

Delaware_Maryland_-District of Columbia_-Virginia _
West VirginiaNorth Carolina_South Carolina_-Georgia __
Florida

East South Central:

Kentucky_.Tennessee_Alabama_Mississipp
W e s t South

ntral:

Arkansas_-Louisiana __
Oklahoma_-Texas __
Mountain:

Montana_-Idaho_
Wyoming_-Colorado_-New Mexico __
Arizona __
Utah--

Nevada __

Pacific:

Washington__Oregon _
California_--

C. 3-region s t u b (document-size page)
United States __

The North_The South __
The West_N e w England_-

Maine_New Hampshire_Vermont_Massachusetts __
Rhode Island_-Connecticut__Middle Atlantic __

New York_New Jersey_
Pennsylvania __

East North CentraL

Ohio __
Indiana_Illinois-Michigan __
Wisconsin _

West North Central_.

Minnesota_Iowa __
MissourL
North Dakota _
South Dakota _
Nebraska_.Kansas__South Atlantic__

Delaware__Maryland_District of Columbia_-Virginia __
West Virginia_-North Carolina _
South Carolina _
Georgia_Florida.-East South CentraL

Kentucky_-Tennessee __
Alabama_Mississippi_-West South Central__

Arkansas _
Louisiana __
Oklahoma__Texas __

Mountain __

Montana __
Idaho_
Wyoming_-Colorado __
New Mexico __
Arizona_-

Utah--

Nevada_-

Pacific __

Washington _
Oregon _
California_--

SEC.

11-A

AREAS

PAR.

1103

b. Planning p a g e s . - S t a n d a r d stubs A and B (fig. 17) are designed to fit on
one full census-size ? page, set in 6-point type, leaded (pronounced ' l e d e d " ) .
Space breaks are inserted above each region and division subhead. (The leading
and space breaks are not shown in fig. 17 since the pages of this manual are document size.) In typewriter-offset work, both where the entire table (including
the stub) is typed and where page forms are prepared with a preprinted stub,
the leading is normally omitted in order to avoid soft-rolling t h e platen on the
typewriter.
Standard stub C is designed to fit on one full document-size 2 page, set in
6-point type, solid.3 Frequently, because of space factors, s t u b C is presented
without t h e region totals. T h e cross-reference convenience of these standard
1-page stubs is obvious, particularly if a series of such tables appears.
In order to insure 1-page tables of this type, the table-designer should keep in
mind that(1) Only one line of data per State is practicable, with no overruns in t h e
stub. In a narrow s t u b some abbreviation will be needed.
(2) T h e table title should be not longer t h a n two lines.
(3) T h e boxhead should be held to a maximum of about six lines.
(4) Headnotes, if any, should be held to not more t h a n two lines.

1104. Regions.-Either a 3-region, or a 4-region, distribution
may be used. T h e difference is that the " T h e North" in the 3-region
grouping becomes " T h e Northeastern States" and " T h e North Central
States" when the 4-region grouping is used. (See fig. 17.) T h e composition of " T h e South" and " T h e West" is identical in both groupings. T h e standard order of listing is as follows:
Example A. 3 regions:
United States_

The North_-The South_-The West_--

Example B. 4 regions:
United States_-

The Northeastern States_The North Central States_-The South_The West_--

1105. Divisions and States.-List divisions geographically within
each region. List States geographically within each division. Alphabetical arrangement of S t a t e data is out of place in a table showing
d a t a for divisions.
a. Position of division subtotals.-Division subtotals may be grouped at the
top of t h e S t a t e list, or they may be interspersed; that is, inserted above t h e
appropriate State subentries. In census-size pages, they are usually grouped at
t h e top; in document-size pages, t h e y are usually interspersed.
b. Division subheads e s s e n t i a l . - I f division totals are grouped at t h e top of t h e
table, division subheads must also appear at appropriate points in t h e State list in
order to make the geographic S t a t e arrangement intelligible. Where only one
line per State is shown, divisional grouping by means of space breaks may seem to
obviate the need for division subheads; this is not the case.
For definition of "census-size" and "document-size," see par. 211.
"Leaded" means additional white space (2 points, or as inch) is left between the lines; a decided aid
in reading small type. "Solid" means no additional white space is thus left. Normally, however, a space
break appears above each subhead or subtotal line, whether the stub is leaded or solid.
2


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

161

PAR.

1105

Example A - 1 . Right (divi-

sion totals at top): 1

Example A-2. Right (division totals interspersed):2

United States_-

United States__

New England . . .
Middle Atlantic..
East North Central_-West North Central_South Atlantic_-East South Central_-West South Central_Mountain_Pacific_--

New England_-Maine ..
New Hampshire_Vermont_Massachusetts_Rhode Island_Connecticut--

New England:
Maine ..
New Hampshire_Vermont__Massachusetts_Rhode Island . . .
Connecticut_etc.
1

SEC.11-A

T H E STUB : STANDARD LISTINGS

Preferred for census-size pages.

Middle Atlantic_-New York_New Jersey_Pennsylvania_East North Central_-Ohio-Indiana_Illinois_-etc.

Example A-3. Wrong (division subheads a n d
grouping needed):
United States__

Maine_New Hampshire-Vermont_Massachusetts_Rhode Island_ConnecticutNew York_New Jersey_Pennsylvania_Ohio-Indiana_Illinois_-Michigan_Wisconsin_-Minnesota_etc.

2 Preferred for document-size
pages.

c. Incomplete listings of States under division heads.-Where not all States
a r e shown under the division subheads (or subtotal lines), t h e reader should be
warned t h a t t h e list is incomplete. This situation commonly arises where t h e
information was not obtained or tabulated for certain States. Here, two choices
a r e available: (1) R u n t h e list of States in full with appropriate footnotes for t h e
States for which information is lacking, as in example B-1. (2) If this is not
desired, omit t h e listing for such States a n d modify t h e divisional head by prefixing with " I n . . . . ," as in example B - 2 .
Example B-1. Right:

Example B-2. Right:

In Middle Atlantic_-Middle Atlantic_New York.
New York_New Jersey 1__
Pennsylvania_Pennsylvania_-In East North Central.
East North Central__1 Information not available (or "Not included in survey").

Example B-8. Undesirable (no warning):
Middle Atlantic__New York.
Pennsylvania_East North Central_--

1106. Divisions.-List the nine census divisions geographically
(unless r a n k order is desired), as follows:
New England_Middle Atlantic_
East North Central_West North Central_South Atlantic_
East South CentraL _
West South Central_-__
Mountain_
Pacific_

1107. S t a t e s . - M a y be listed geographically, alphabetically, or in
rank order in terms of some characteristic. The complete list must
include the District of Columbia; otherwise the State totals will not
add to the total for the United States.
Where States are listed geographically, divisional subheads, or
their equivalent, should appear above each divisional grouping,

162

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

SEC.

11-A

PAR.

AREAS

1107

irrespective of whether division totals are shown in the table. (For
arrangement with division totals, or with totals for regions, divisions,
and States, see fig. 17, p. 160.)
Stub A. Geographical:

United States_

New England:
Maine_New Hampshire_-Vermont.-Massachusetts-Rhode Island_-ConnecticutMiddle Atlantic:
New York_-etc.

Stub B. Alphabetical:

United States_.--

Alabama_Arizona_Arkansas_-California_-Colorado_-Connecticut . . .
Delaware_
Dist. of Columbia_-Florida.-etc.

a. Arrange geographically where(1) Significant geographic differences are reflected in t h e d a t a ;
(2) Comparisons are intended with tables with a geographic arrangement; or
(3) T h e interest of users is most likely to be concentrated on comparisons of
data for neighboring States.
b. Arrange alphabetically where(1) No important geographical differences are reflected;
(2) T h e primary interest is likely to lie in data for individual States, thereby
setting a premium on t h e ability to locate individual States in t h e list; or
(3) Comparisons are most likely to be made of d a t a for noncontiguous States.

1108. Counties.-List alphabetically within each State, except
where rank order is desired. (In Louisiana, the term "parish" is
used instead of county.)
a. Coextensive counties and cities.*-A county is coextensive with a city when
t h e two areas have identical boundaries. Four such cases existed in 1940, involving San Francisco, Denver, Philadelphia, and New Orleans. T h e political divis i o n s t h u s " p a i r e d " m a y or m a y n o t h a v e i d e n t i c a l n a m e s .

In

t h e cases of

San Francisco and Denver, t h e county and city, in each instance, constitute a
single political entity; as, for example, " T h e City and County of San Francisco."
In Philadelphia and New Orleans, county and city governments exist side-by-side.
San Francisco city and County
Philadelphia city and Philadelphia County
New Orleans city and Orleans Parish
Denver city and Denver County
Basic treatment.-Show t h e d a t a for t h e given political unit at the point in
t h e s t u b where it would normally fall, even though this may mean complete duplication of figures within t h e same table or in different tables. Thus, in a county
table show d a t a for Philadelphia County; in a city table show d a t a for Philadelphia
city. In a table presenting d a t a for both counties and cities, show the Philadelphia figures twice. This procedure is most convenient for the user. Since
the number of such paired areas is few, t h e space problem is rarely serious.
(1) Individual tables for each c o u n t y or c i t y . - T h i s is the only instance
where t h e space factor is likely to be serious.
, For t h e special case of New York city, see par. 1108b.

776554
49

12


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

163

PAR.

1108

THE

SEC.11-A

STUB : STANDARD LISTINGS

Where space permits: Repeat in full under t h e appropriate title for each area.
For example, show t h e entire table, appropriately titled as being for Orleans
Parish or for San Francisco County; then show t h e entire table again, appropriately titled as being for t h e city of New Orleans or for t h e city of San Francisco.
This is particularly convenient for t h e reader when a long series of county tables,
arranged alphabetically by county, is followed by a series of city tables, arranged
alphabetically by city.
Where space forbids repetition: In t h e county series, insert t h e table for t h e
area as a county. T h i s will insure t h a t t h e sum of entries in t h e county tables
will equal t h e total entries for t h e State. In t h e city series, insert t h e table
title with a headnote as indicated below; omit t h e table proper. T h e city
area is thereby accounted for at t h e point where t h e reader would tend to look
for city figures.
Example:
TABLE

5. - P O P U L A T I O N

CHARACTERISTICS FOR D E N V E R COUNTY

(Give table in full.)
TABLE 10.-POPULATION

C H A R A C T E R I S T I C S F O R C I T Y OF D E N V E R

[Denver city and county are coextensive; hence figures for Denver city are the same as shown for Denver
County in table 5]
(Omit remainder of table.)

(2) Tables with c o u n t y and c i t y s t u b s . - L i s t with full d a t a as a county;

repeat in full as a city.
Example A. Where cities are listed under county subtotals:
Denver County _
275
164
1, 282
Denver city_.1,282
275
164
Orleans Parish_-New Orleans city_-

629

629

3, 124

469

381

928

3,124

469

381

928

Example B. Where counties and cities are listed in separate groups:
CITIES OF 25,000 OR MORE

COUNTIES

Custer_-Delta_-Denver_-

70
65

_ 1,282

Colorado Springs_Denver_-Pueblo __

862
1, 282

463

b. Counties within a city (New York c i t y ) . - I n 1940 New York city represented t h e sole instance where a city was coextensive with a group of counties.
Furthermore, each included county is itself coextensive with a specific borough.
In three cases, t h e county and borough have t h e same name; in two cases, t h e
names are different, as follows:
Bronx County__
Bronx borough
Kings County__Brooklyn borough
New York County__
M a n h a t t a n borough
Queens County__
Queens borough
Richmond County __
Richmond borough
(1) C o u n t y t a b l e s . - E a c h county should be included individually in t h e
county list. If desired, each may be footnoted " P a r t of New York c i t y " ; or
borough, part of New York city," specifying t h e appropriate
borough name.
Where cities or minor civil divisions are listed by counties, include each
county in its appropriate alphabetical position in t h e full list (not in a special
group) and handle as follows:

164

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEc. 11-A

AREAS
Minor civil division

Albany County__

Albany city_Berne town_-Bethlehem town_-Coeymans town_-

Ravena village__Bronx County_

Barker town __
Binghamton city _
Binghamton town_-Chenango town_
Colesville town_-1

1108

Minor civil division
221, 315
130, 577
1, 325
9, 782
4, 536
1,810

Jefferson County-Con.

Philadelphia town_-Philadelphia village_.Rodman town _
Rutland town_--

1, 372
722
856
1, 622

Theresa town _

1, 675

Black River village (part)_

362

-1,394, 711

Bronx borough (part of New York
city)!__ 1,394, 711
Broome County_

PAR.

165, 749
1, 223
78, 309
1, 576
3, 265
2, 652

Kings County_-

_ 2,698,285

Brooklyn borough (part of New
York city)!__
_ 2,698,285
Lewis County __

22, 815

Croghan town_-Croghan village_.-

2, 557

801

For total figures for New York city, see table

(2) C o u n t y n a m e s versus borough n a m e s . - I n general, if a table calls
specifically for counties, list t h e component areas of New York city as counties,
using t h e county names. Show t h e New York city t o t a l figures in footnotes,
not in t h e county list.
Where city data are t h e main consideration, list t h e component areas of
New York city as boroughs under their borough names, arranged alphabetically under t h e listing for t h e city as a whole.
c. Independent c i t i e s . - T h r e e States (Maryland, Missouri, and Virginia)
contain "independent cities"; t h a t is, cities independent of county jurisdiction.
In 1940 there were 26 such independent cities in t h e United States; 1 each in
Maryland and Missouri and 24 in Virginia. Each must be specifically included,
a n d shown separately, in a n y "county listing" which is intended to distribute
d a t a for t h e entire State. In States with independent cities the summation of
county figures alone will not equal t h e State total.
(1) In Maryland a n d M i s s o u r i . - W h e n listing Maryland counties, include
"Baltimore city" as well as "Baltimore County." T h e city is not a part of the
county.
Similarly, St. Louis city, Missouri, is independent of St. Louis County.
Therefore, when listing t h e counties of Missouri, both areas must be shown;
t h e one as " S t . Louis city," the other as " S t . Louis County."
Examples:

A. Maryland

B. Missouri

Allegany_Anne Arundel_-Baltimore city _
Baltimore County_-Calvert--

St. Clair_St. Francois_St. Louis city _
St. Louis County _
Ste. Genevieve_-

(2) In Virginia.-Because of t h e large number of independent cities (24) in
this State, and because several Virginia counties include t h e word "City" as
part of the county name, separate grouping is recommended for t h e independent
cities.
Separate groups.-Recommended. Arrange t h e independent cities in a separate group following t h e alphabetical listing of counties.
Interspersed.-Acceptable. Where both counties and independent cities
are merged into a single alphabetical list, include area designation after each.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

165

PAR.

1108

THE

STUB : STANDARD LISTINGS

sEc.

11-A

This reduces confusion in t h e case of counties which include "City" as part of
t h e county name, such as "Elizabeth City County" which is not a city. As
a minimum, "County" may be omitted from names of all counties except those
which include "City" as part of t h e county name; however, be sure to specify
"County" after such names. Thus, list "Elizabeth City County," never merely
"Elizabeth City," even though t h e s t u b boxhead specifies "County."
T h e term "city" is given an initial cap where a part of a county or city name;
it is completely lower cased (small letters) where it is appended solely as t h e
area designation.
Example C-1. Recommended:
County or independent city
Virginia_.

Example C-2. Acceptable:
County or independent city
Virginia_.

COUNTIES

Accomack_Albemarle_Alleghany_..
Amelia_Dickerson _
Dinwiddie_-Elizabeth City County__
Essex_-

Accomack County _
Albemarle County _
Alexandria city_.Alleghany County_Botetourt County _
Bristol city _
Brunswick County__Buena Vista city_.
Campbell County_-Caroline County_.

INDEPENDENT CITIES

Alexandria.
Bristol.
Buena Vista_--

Dinwiddie County.Elizabeth City County _
Essex County _

d. Yellowstone National P a r k . - T h i s National Park, geographically located
within t h e limits of Idaho, Montana, and Wyoming, is independent of county
organization; hence, must be included separately in listing counties in these three
States; otherwise the State total will not be represented fully. In each instance,
t h e listing should read "Yellowstone National Park 1 (part)" followed by t h e
statistics for t h e given part. The footnote should read "1 Geographically
located within limits of Idaho, Montana, and Wyoming." A total for the park
as a whole is usually added to t h e footnote; such footnotes rarely provide detailed
distributions for the park as a whole.

1109. Cities, towns, villages, places, etc.: General.-Cities, towns,
villages, incorporated places, urban places, etc., are normally listed
alphabetically unless rank order is required. Treatment of State
designation is as follows:
a. Abbreviation of State n a m e s . - S t a t e names should always be abbreviated
when included as part of individual listings for cities, towns, villages, etc. This
places t h e emphasis upon t h e place names and focuses the reader's attention
upon them.
Right:
Albany, N. Y - Chicago, Ill_-Denver, Colo __
Erie, P a Lowell, Mass_--

Undesirable:
Albany, New York_Chicago, Illinois_Denver, Colorado_-Erie, Pennsylvania _
Lowell, Massachusetts_-

b. When to include State n a m e s . - W h e r e t h e materials are not arranged under
S t a t e headings, and the listing includes places in various States, t h e State name

166

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

SEC.

l1-A

AREAS

PAR.

1109

(abbreviated) should be added to each individual listing if t h e list includes one
or more cities with a population of less t h a n 500,000.
Note: All listings (not under State headings for), and all references to, t h e following cities should invariably carry t h e appropriate State names to avoid confusion.
This list is comprised solely of cities of 50,000 inhabitants or more with duplicating
names.
Charleston, S. C.
Kansas City, Kans.
Springfield, Ill.
Kansas City, Mo.
Charleston, W. Va.
Springfield, Mass.
Springfield, Mo.
Portland, Maine
Columbus, Ga.
Springfield, Ohio
Portland, Oreg.
Washington, D. C,1
Columbus, Ohio
1

To avoid confusion with Washington State.

c. Arrangement under State h e a d s . - L i s t s of cities which include one or more
with a population of less t h a n 50,000 are best arranged under State headings.
Those restricted to cities of 50,000 inhabitants or more may be listed either in a
single alphabetical group or arranged alphabetically by State.

1110. Cities, towns, etc.: Size groups.-Frequently, cities, towns,
villages, etc., are listed in groups classified by "size"; that is, in groups
according to number of inhabitants. Some of the more common
groupings are given below with pertinent notes as to listing.
a. Cities of 1,000,000 inhabitants or m o r e . - U s u a l l y presented in a single alphabetical listing, irrespective of State of location. State names are omitted as
unnecessary, except that " D . C." is always specified for the city of Washington.
b. Cities of 500,000 inhabitants or m o r e . - S a m e procedure as for cities of
1,000,000 inhabitants or more.
c. Cities of 100,000 inhabitants or m o r e . - O n e of t h e best-known city stubs
of the Bureau of t h e Census. T h e cities are alphabetically arranged in a single
list. State names (abbreviated) appear after each place name. This stub is
planned for a 1-page presentation (census-size), assuming one line per area.
d. Urban places of 50,000 inhabitants or m o r e . - U s u a l l y presented as " u r b a n
places" rather than "cities" since not all urban areas in this size group are cities.
In 1940, one township, one town, one village, and one county (Arlington County,
Va.) were included. May be arranged(1) In a single a l p h a b e t i c a l l i s t , with State name (abbreviated) after each
place name. On a census-size page, this s t u b can be standardized to run two
tables to each five pages, with each city located at t h e same point on t h e page in
alternate tables. This standard arrangement is useful when a series of these
tables is desired.
(2) A l p h a b e t i c a l l y u n d e r S t a t e h e a d s . - S t a t e s are usually arranged
alphabetically.
e. Urban places of 25,000 inhabitants or m o r e . - L i s t in alphabetically arranged
groups under State headings. States also are arranged alphabetically.
f. Other city-size groups.-Groupings involving cities or places with fewer t h a n
25,000 inhabitants are normally arranged in State groups. Common groups of
this type arc:


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

167

PAR.

1110

THE

S T U B : STANDARD LISTINGS

SEC.

11-A

Urban places of 10,000 inhabitants or more
Urban places of 10,000 to 25,000 inhabitants
Urban places of 5,000 to 10,000 inhabitants
Urban places of 2,500 to 5,000 inhabitants
Rural incorporated places (that is, all incorporated places with fewer
t h a n 2,500 inhabitants)
Incorporated places of 1,000 to 2,500 inhabitants
Incorporated places of less t h a n 1,000 inhabitants

1111. Cities, towns, etc.: Specifying area designation.-The determination of the need for inclusion of area designation such as city,
town, village, etc., is a problem frequently encountered. No universally applicable rule can be laid down, but the following may prove
helpful.
a. Single specification in stub b o x . - W h e r e a single area designation is specified
in t h e s t u b box, places of t h a t t y p e do not require individual designation, but
places of other types do. T h a t is, if the term " C I T Y " appears alone in t h e stub
box, t h e term "city" need not be placed after each city name, but a n y village,
town, etc., listed should be specified as such.
Example A:

CITY
Allen_Braden __
Canton village__Dort_-Fallon town_Grayson__

b. Dual specification in stub b o x . - W h e r e two area types are indicated in t h e
s t u b box, places of neither t y p e require individual area designation, but it may be
inserted for t h e minority group if desired. All other types should be specified.
Example B-1. Right:
CITY OR TOWN

Allen_Braden_Canton village_Dort-X Fallon_-Grayson__

Example B-2. Right:
CITY OR TOWN

Allen _
Braden-Canton village_Dort_Fallon t o w n - Grayson __

c. Generalized area reference in stub b o x . - T h i s obviates specific need for
individual area designation. However, if one objective is to provide specific
description in this respect, t h e generalized reference does not prevent detailed
description.
Example C-1. Right:
URBAN PLACE

Allen
Braden_Canton_Dort_-Fallon_-Grayson__

Example C-2. Right:
URBAN PLACE

Allen city__Braden city _
Canton village_
Dort city_Fallon town_-Grayson city__-

d. Exceptional c a s e s . - W h e r e v e r t h e area designation for an individual place
is beyond normal expectation, or reflects a n y unusual situation, always specify
against t h e place name.
(1) Beyond normal expectation.-Example: In a listing of urban places
in Virginia, Arlington County should always be listed in full, not merely as
"Arlington," since t h e presence of a county in a city or urban-place list is
beyond normal expectation. (See par. 1110d, above.)

168

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

SEC.

11-A

PAR. 1111

AREAS

(2) Confusion of area n a m e s . - E x a m p l e : Baltimore city a n d St. Louis
city, when they appear in a county listing, should have "city" specified to
avoid confusion with t h e counties of t h e same names of which these cities are
not a part. (See par. 1108c.)

1112. Minor civil divisions.-The Bureau of the Census applies the
general term "minor civil division" to the primary political divisions
into which counties are divided (townships, districts, precincts, etc.).
The designation of these divisions varies from State to State; a few
States have more than one type. Ordinarily, incorporated places
form secondary divisions of the minor civil divisions in which located.
In some States, however, all or some of the incorporated places
themselves constitute primary divisions of the counties; and the larger
cities often contain all or part of several minor civil divisions. For
the situation in a particular State, and for a map of each State showing
minor civil divisions, see Population, volume I, 1940 (or of the most
recent decennial census).
a. Order of listing.-The order of listing (within counties) depends, in part,
upon whether the minor civil divisions in t h e given State are named or numbered.
(1) N a m e d . - A r r a n g e alphabetically.
(2) N u m b e r e d . - A r r a n g e numerically, beginning with t h e smallest number.
(3) N a m e d a n d n u m b e r e d . - I n some States both names and numbers are
assigned. Sometimes both appear for individual minor civil divisions. Where
some are numbered and others are named, list first those which have names,
followed by those t h a t are numbered. If individual areas have both names
and numbers (Township 4, Menango; Ranton, District 5), arrange according
to whichever comes first in t h e area designation.
b. Interspersing incorporated p l a c e s . - A t times, it is necessary to list both
minor civil divisions a n d incorporated places in a single stub, showing full interrelationship.

T h e m e t h o d i n v o l v e d is t o o c o m p l e x f o r d e t a i l e d c o v e r a g e h e r e ; s e e

table 4 for t h e given State in Population, volume I, 1940, for procedure.5 However,
several general rules may be mentioned:
(1) Incorporated places e n t i r e l y w i t h i n o n e m. c. d . - L i s t as subentry
under m. c. d. name; in letterpress work, use italics unless "Remainder of
m. c. d." is shown separately.
Example A-1:

Fort Osage township_Buckner cityLevasy town_-

2,575
571
126

Kaw townshipKansas City_-

_ 399,178
399, 178

(2) S p l i t b e t w e e n m. c. d . ' s . - U n d e r the first m. c. d. (in order of listing)
in which partly located, list t h e incorporated-place total, followed by specified
parts in t h a t m. c. d. and in other m. c. d.'s. Under each of t h e other m. c. d.'s
of which a part, list only t h a t part within t h a t m.c.d. (For example, see below.)
(3) Split b e t w e e n c o u n t i e s . - W h e r e split between counties, list t h e incorporated-place total under first affected m. c. d. in each county. For handling
m. c. d. make-up of portion within each county, follow rule 2, above.
1 Description of method used in 1940 and many earlier censuses. Subject to change in 1950 census
reports.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

169

PAR.

1112

THE

STUB : STANDARD LISTINGS

sEc.

11-A

Example A-2: In t h e following example, illustrating rules 1 to 3, note treatment for Neck City, Joplin city, a n d Cartersville city, designated in the margin
as (a), (b), a n d (c), respectively. This is an extract from Missouri table 4,
pp. 590, 592, Population, volume I, 1940.
Jasper County_.
(a)

Mineral township_-

Duval township_.

Neck City, total _
In Duval twp_-In M i n e r a l t w p __

A l b a city_-Cartersville city (part)Neck City (part) _
Oronogo city_Purcell city_-

(c)
(a)

Galena township_(b)

Preston township__-

Joplin city, total _
In Galena twp_In J o p l i n t w p _
In Newton County __

Newton County __

Benton township_Berwick township_Buffalo townshipDayton township__-

Jackson township_Joplin township
(c)
(b)

_

Cartersville city, total_-In Joplin t w p _
In M i n e r a l twp_Joplin city (part) __
Webb City__-

Shoal Creek township
(b)

Lincoln township__.

Joplin city, total_In Newton County __
In Jasper County_-

Van Buren township_--

1113. Urban-rural areas. This is a classification reflecting, broadly,
density of population. For Census of Population purposes in 1940,
t h e urban area is made up, for the most part, of cities and other incorporated places having 2,500 inhabitants or more. T h e remaining
territory is, in general, classified as rural. T h e population of the rural
area may be further divided into two subclasses: "Rural-farm" and
"Rural-nonfarm," on t h e basis of farm residence. T h e total farm population is not exactly t h e same as t h e rural-farm population, however,
because of t h e small number of persons living on farms located within
u r b a n a r e a s ; t h a t is, t h e " u r b a n - f a r m " population.

Because of t h e

small numbers involved, figures for the urban-farm population are
rarely shown separately.
Detailed description of t h e urban-rural classification and its application in the several States may be found in Population, volume I, 1940 (or of t h e most recent decennial census).
a. Order of listing and placement of hyphen.-Except for special purposes, the
order of listing is as indicated below. T h e "rural-nonfarm" and "rural-farm"
categories may, of course, be shown under a rural total. However, where both
subcategories appear, t h e rural total usually is omitted.
Note placement of hyphen in t h e expressions "rural-nonfarm" and "ruralf a r m "; erroneous placement or omission in copy constitutes a serious nuisance
at times, particularly since t h e error is frequently overlooked until t h e material
is received in proof.
Example A. Listing order:

Right
Urban
Rural-nonfarm
Rural-farm

Wrong
Urban
Rural-farm
Rural-nonfarm

Example B. Hyphen placement:

Right
Rural-nonfarm
Rural-farm

Wrong

Rural non-farm
Rural nonfarm
Rural farm

b. Urban under special r u l e . - I n all tables which present statistics for individual urban areas, each place classified as urban under special rule should be plainly
indicated. This should be done either by a footnote; or in a headnote if a standard
reference symbol, such as an asterisk (*), is used. See Population, vol. I,
1940, for places classed as " u r b a n under special rule" in t h e several States.

170

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEc. 11-A

AREAS

PAR.

1113

1114. Metropolitan districts.-A metropolitan district is not a
political unit, but an area including all of the thickly settled territory
in and around a city or group of cities. It tends to be a more or less
integrated area with common economic, social, and, often, administrative interests.
For the 1940 census, a metropolitan district was set up in connection with each city of 50,000 inhabitants or more, two or more such
cities (and their environs) sometimes being in one district. The
general plan was to include in the district, in addition to the central
city or cities, all adjacent and contiguous minor civil divisions or
incorporated places having a population of 150 or more per square mile.
a. System of n a m i n g . - M e t r o p o l i t a n districts are named according to the name
of the central city, as "Salt Lake City Metropolitan District." If there are two
or more central cities, t h e names are combined, with t h e largest city named first,
as "Norfolk-Portsmouth-Newport News Metropolitan District." However, only
those cities are named in t h e title which are considered "central" cities; this does
not necessarily include all cities of 50,000 or more within t h e given district. Thus,
Gary, Ind., is not mentioned in t h e title of t h e district in which it is located, the
Chicago Metropolitan District.
b. Use of State n a m e s . - W h e r e t h e district is specified in full, t h e State name
is rarely included, as "Buffalo-Niagara Metropolitan District." However, when
confusion is likely, specify the State (abbreviated). Two types of cases appear
where the State name is essential:
(1 ) N a m e includes c i t i e s in m o r e t h a n o n e S t a t e . - L i s t the State name
(abbreviated) after each city name, as "Huntington, W. V a . - A s h l a n d , Ky."
If the area designation is added, place the State names in parentheses, as " H u n t ington (W. V a . ) - A s h l a n d (Ky.) Metropolitan District."
(2) N a m e is o n e t h a t is d u p l i c a t e d in list of m a j o r c i t i e s . - R e f e r e n c e s
to metropolitan districts of the following cities should always specify State name.
Charleston, S. C.
Charleston, W. Va.

Kansas City, Kans.
Kansas City, Mo.

Columbus, Ga.
Columbus, Ohio

Portland, Maine
Portland, Oreg.

Springfield,
Springfield,
Springfield,
Springfield,

Ill.
Mass.
Mo.
Ohio

(3) W a s h i n g t o n , D. C., M e t r o p o l i t a n D i s t r i c t . - A l w a y s specify " D . C."
in referring to the Washington, D. C., Metropolitan District.
6 Two additional related areas may be noted: The metropolitan area and the metropolitan county. Whereas
the metropolitan district is drawn on minor civil division lines, thereby including parts of counties, the
metropolitan area is drawn on county lines-that is, either the entire county is included in the area or it is
omitted completely. The basic unit of the metropolitan district is the minor civil division; that of the
metropolitan area is the "metropolitan county."
For a discussion of the relationship between "metropolitan districts," "metropolitan areas," and "metropolitan counties," see Bureau of the Census, County Data Book, A Supplement to the Statistical Abstract of
the United States, Washington, 1947.
At the time of going to press, work was being completed by an Interdepartmental Committee which is
establishing the "Standard Metropolitan Area" as a statistical area recommended for use by all governmental agencies to take the place of the various areas described above, and also of industrial areas and labormarket areas. For principles adopted as a guide, see mimeographed statement, "Standard Metropolitan
Area Definitions," issued by Bureau of the Budget, Executive Office of the President (undated). Specific
details of presentation will be determined after work of this committee has been completed.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

171

PAR.

1114

THE

S T U B : STANDARD LISTINGS

sEC.

11-A

c. Use of h y p h e n . " - A s a m a t t e r of expedience, t h e hyphen (-) is used instead of
the dash ( - ) to separate the names of the central cities when more than one such
name appears in the district title. Exception: Use the dash ( - ) wherever(1) The title includes a city name which is itself hyphenated.
(2) T h e title includes State names.

Example A - 1 . Right:
S C R A N T O N - W I L K E S - B A R R E METROPOLITAN DISTRICT.
DAVENPORT
(IOWA)-ROCK
ISLAND-MOLINE
(ILL.)
METROPOLITAN
DISTRICT
KANSAS CITY ( M O . ) - K A N S A S CITY (KANS.) METROPOLITAN DISTRICT
KANSAS CITY, M O . - K A N S A S CITY, KANS., METROPOLITAN DISTRICT

Example A - 2 . Wrong:
SCRANTON-WILKES-BARRE METROPOLITAN DISTRICT
DAVENPORT ( I O W A ) - R O C K I S L A N D - M O L I N E (ILL.) METROPOLITAN D I S TRICT
KANSAS CITY ( M 0 . ) - K A N S A S CITY (KANS.) METROPOLITAN DISTRICT
KANSAS CITY, M O . - K A N S A S CITY, KANS., METROPOLITAN DISTRICT

d. Order of listing. Since many metropolitan districts are located in two or
more States, t h e usual order of listing is alphabetical by district name in a single
listing for t h e United States.
e. Inclusion in State reports or under State headings.-Districts located in two
or more States are subject to special rulings depending upon t h e purpose of t h e
presentation. A general rule, subject to modification as needed, is to place data
for a given district under t h e heading of t h e State in which t h e largest central
city is located.
Exception: T h e Davenport ( I o w a ) - R o c k I s l a n d - M o l i n e (Ill.) Metropolitan
District, in t h e 1940 Population Census Reports, was placed under both Iowa
a n d Illinois headings since t h e largest central city (Davenport) is in Iowa, but
t h e combined population of t h e other two central cities (Rock Island and Moline),
which a r e in Illinois, was larger t h a n t h a t of Davenport.

1115. Census t r a c t s . - C e n s u s tracts are small areas into which
certain large cities, and sometimes their adjacent areas, are subdivided
for statistical and local administrative purposes through cooperation
with a local committee in each case.
a. Order of l i s t i n g . - C e n s u s tracts are usually numbered; some have prefix
or suffix letters.
(1) N u m b e r e d . - L i s t in numerical order.
(2) Suffix letters.-Arrange first in numerical order with those of each
number grouped together. Within each number group, arrange t h e tracts
alphabetically by suffix letter.
(3) P r e i x l e t t e r s . - W h e r e prefix letters are not directional in character,
arrange first alphabetically by prefix letter. Within each prefix-letter group,
arrange t h e tracts in numerical order.
Where prefixes are directional in character, arrange first according to t h e
compass points; t h a t is, start with prefix letter " N " (for North), and moving
clockwise (N, E, S, W), group all tracts with t h e same prefix together. Within
each prefix-letter group, arrange t h e tracts in numerical order.
Technically, the en-dash, not the hyphen, is used here in letterpress (type-set) material.

172

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEc. 11-A

AREAS

PAR.

1116

1116. Area names as total-line captions.-Area names are commonly used as total and subtotal line captions, in place of, or in addition to, the term "Total."
a. Table for single a r e a . - W h e r e all d a t a in t h e table are for a single area,
and t h e name of t h e area is specified in t h e area segment of t h e title (or in t h e
running head of the page) 1 the grand total line may be generalized, asThe State_-The city_--

The metropolitan district_-The county _

b. Table with areas listed in s t u b . - W h e r e areas are listed in t h e stub, each
with its own block of subentries, t h e name of the area may be inserted in place
of "Total," or of " A l l classes," etc., wherever this will not abuse its function as a
t o t a l line. This saves space since otherwise it is necessary to insert each area name
as a centered head.
Alabama_White_Negro__Other races_

Akron, Ohio_White_-Negro__Other races_--

Arizona_-White_Negro__Other races __

Albany, N. Y - White__Negro__Other races _

c. Combining area name with " T o t a l . " - H e r e the question arises whether t h e
word "total" should be placed first (Total, Alabama) or last (Alabama, total)
in t h e given caption. T h e same principles apply as discussed in paragraph 957,
"Placement of 'Total' in combination with other wording."

1117. Spelling of area names.-Meticulous care is essential in spelling area names. Normal rules of spelling do not apply. Names pronounced the same way may be spelled differently in different States
and within t h e same county or in different counties in t h e same State.
Most Bureau of t h e Census tables are prepared under standard operating procedures which tend to avoid incorrect spellings. However,
constant watchfulness, and verification when t h e slightest doubt
arises, will sometimes catch errors in materials that presumably have
already been verified carefully. Special care is required for tables
prepared and typed in offices of analysts where elaborate verification
procedures are not employed.
T h e Geographer of the Bureau is t h e final authority for spelling of
area names in statistical tables of the Bureau of the Census.
For general purposes, reference is suggested to Population, vol. I,
of the most recent decennial census, since the spelling shown there is
t h e spelling provided by t h e Geographer as correct at the time the
volume went to press. T h e Geographer should always be consulted for
t h e most recent information.
Finally, t h e United States Postal Guide, commercial atlases, and
dictionaries should not be relied upon for area-name spellings. T h e
Postal Guide is t h e authority for the spelling of post-office names.
Post-office names are not always spelled the same way as t h e legal


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

173

,

PAR.

1117

THE

STUB : STANDARD LISTINGS

sEc.

11-A

n a m e of t h e community or area in which the post office is located or
which it serves. Thus, t h e post office i n S a n Buenaventura, Calif.,
is called Ventura; the post-office in Boise City, Idaho, is called Boise.
T h e following observations and examples illustrate t h e problems of
area-name spelling and the pitfalls commonly encountered. Lists of
names shown are illustrative only; they are not comprehensive since
they cover only a few of many similar cases.
a. Same county, different spellings of areas.-Following are examples:
In Alabama (Cherokee County), Centre town is located in Precinct 6, Center.
In Kansas, Salina city is in Saline County.
In North Dakota (Bottineau County), it is Newburg village but Newborg
township.
b. Same State, different counties.-Following are examples:
In Ohio, it is Greene township in Clark and Trumbull counties, but Green
township in other Ohio counties.
In South Dakota, it is Carlisle township in Brown County, but Carlyle
township in Beadle County; it is Linn township in H a n d County, Lynn in other
counties, and Lien township in Roberts County; it is Clare township in Moody
County, but Claire City town in Roberts County.
c. County spellings.-The following list of similar county names is by no means
complete:
Carver in Minnesota
Aiken in South Carolina
Carter in all others
Aitkin in Minnesota
Allegany in Maryland and New York
Cheboygan in Michigan; Sheboygan in
Wisconsin
Alleghany in North Carolina and Virgin1a
Clarke in Alabama, Georgia, Iowa,
Allegheny in Pennsylvania
Mississippi, and Virginia
Andrew in Missouri
Clark in all others
Andrews in Texas
Coal in Oklahoma
Aransas in Texas
Cole in Missouri
Arkansas in Arkansas
Coles in Illinois
Baca in Colorado; De Baca in New Coffee in Alabama, Georgia, and TenMexico
nessee
Baker in Florida, Georgia, and Oregon Coffey in Kansas
Banner in Nebraska
Cook in Georgia, Illinois, a n d Minnesota
Bonner in Idaho
Cooke in Texas
Barber in Kansas
Davidson in North Carolina and TenBarbour in Alabama a n d West Virginia
nessee
Davison in South Dakota
Barren in Kentucky
Barron in Wisconsin
Davie in North Carolina
Daviess in Indiana, Kentucky, a n d
Brevard in Florida; alsoMissouri
Broward in Florida
Davis in Iowa and Utah
Brooke in West Virginia
Dickenson in Virginia
Brooks in Georgia and Texas
Dickinson in Iowa, Kansas, and MichiBrown in all States
gan
Bulloch in Georgia
Dickson in Tennessee
Bullock in Alabama
Dixon in Nebraska
Burnet in Texas
Douglas in all States
Burnett in Wisconsin
Forrest in Mississippi
Callaway in Missouri
Forest in all others
Calloway in Kentucky
Glascock in Georgia
Cannon in Tennessee
Glasscock in Texas
Canyon in Idaho

174

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

SEC.

11-A

AREAS

Glenn in California
Glynn in Georgia
Green in Kentucky a n d Wisconsin
Greene in all others
Greensville in Virginia
Green ville in South Carolina
Hamblen in Tennessee
Hamlin in South Dakota
Harding in New Mexico and South
Dakota
Hardin in all others
Harford in Maryland
Hartford in Connecticut
Hayes in Nebraska
H a y s in Texas
Hendry in Florida
Henry in all others
Highland in Ohio and Virginia
Highlands in Florida
Huntingdon in Pennsylvania
Huntington in Indiana
Johnston in North Carolina and Oklahoma
Johnson in all others
Kanabec in Minnesota
Kennebec in Maine
Kearney in Nebraska
Kearny in Kansas
Kimball in Nebraska
Kimble in Texas
King in Texas and Washington
Kings in California a n d New York
Laurens in Georgia and South Carolina
Lawrence in all others
Lea in New Mexico
Lee in all others
Leflore in Mississippi
Le Flore in Oklahoma
Linn in Iowa, Kansas, Missouri, and
Oregon
Lynn in Texas
Loudon in Tennessee
Loudoun in Virginia
Manatee in Florida
Manistee in Michigan
Merced in California
Mercer in all others
Morton in Kansas and N o r t h Dakota;
Norton in Kansas
Muscogee in Georgia
Muskogee in Oklahoma

PAR.

1117

Ouachita in Arkansas a n d Louisiana;
Washita in Oklahoma; Wichita in
Kansas a n d Texas
Park in Colorado, Montana, Wyoming
Parke in Indiana
Pottawatomie in Kansas and Oklahoma
Pottawattamie in Iowa
Prince George in Virginia
Prince Georges in Maryland
S t . Francis in Arkansas
St. Francois in Missouri
St. Mary in Louisiana
St. Marys in Maryland
Sanders in Montana
Saunders in Nebraska
Sheboygan in Wisconsin; Cheboygan in
Michigan
Smyth in Virginia
Smith in all others
Stafford in Kansas and Virginia
Strafford in New Hampshire
Stanley in South Dakota
Stanly in North Carolina
Stark in Illinois, North Dakota, and
Ohio
Starke in Indiana
Stephens in Georgia, Oklahoma, and
Texas
Stevens in Kansas, Minnesota, and
Washington
Storey in Nevada
Story in Iowa
Terrell in Georgia and Texas
Tyrrell in North Carolina
Tooele in Utah
Toole in Montana
Uinta in Wyoming
Uintah in Utah
Vermilion in Illinois and Louisiana
Vermillion in Indiana
Wabash in Illinois a n d Indiana
Wabasha in Minnesota
Warrick in Indiana
Warwick in Virginia
Washita in Oklahoma; Wichita m
Kansas and Texas; Ouachita in
Arkansas a n d Louisiana
Woods in Oklahoma
Wood in all others
Wyandot in Ohio
Wyandotte in Kansas

1118. Transcribing and proofreading of area names.-Area names
should never be listed'without' reference to an authoritative Bureau
of the Census source. They should always be verified after each
typing or transcription by a careful spelling out letter by letter.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

175

PAR.

1121

THE STUB : STANDARD LISTINGS

SEC.11-B

Sec. 11-B. Listing Time Periods (1121-1124)
1121. General.-Greater precision in statement of time periods is
necessary in the table stub than in table titles. (See par. 364.)
E v e n here, however, exact specification frequently gives way to conventionalized practices, as follows:
a. Regular census d a t e s . - W h e n reference is to regular census dates, such as
those for the Censuses of Population, Agriculture, etc., specification of the year is
generally sufficient except(1) Where the difference between census dates may affect comparisons
because of seasonal factors or for other reasons.
(2) In introductory summary tables, where it is desirable to include the
exact effective date of the census as a matter of record.
Usual practice

Only where essential
1940 (April 1 ) . 1930 (April 1).-1920 (January 1 ) - 1910 (April 1 5 ) - -

1940--1930--1920--1910--

b. Regular census periods.-For statistics which cover a regular census period
(rather t h a n a given date) as in t h e instance of t h e Census of Manufactures or the
Census of Business, the time reference is normally stated merely as "1939,"
"1937," etc., rather t h a n in the longer and more precise form, such as "During
1937," "During fiscal year 1937," etc.
c. Other dates and time periods.-Specify exact d a t e or period only where(1) Misinterpretation is likely if stated in brief form.
(2) It is not enough to explain the exact effective date or period in the accompanying text.
(3) No accompanying text is presented and t h e exact statement must appear
in t h e table if it is to be mentioned at all.

1122. Order of listing : General.-In a chronological arrangement,
t h e primary question is whether to use (a) t h e direct method; t h a t is, to
place first t h e most recent statistics followed in order by those for
earlier time periods; or (b) t h e indirect method; that is, to show first the
statistics for t h e earliest time period followed by those for more recent
time periods. (See pars. 365-366.) T h e method selected depends
upon t h e nature of t h e statistics, t h e purpose of presentation, and t h e
probable convenience to t h e user. T h e following statements are
suggestive only:
a. Arrangements involving future t i r n e . - H e r e , since t h e usual purpose of
presentation is to project a series into the future by some method of estimate, t h e
normal practice is to run the series forward in time; that is, to place the earlier
time periods at the top of t h e list, and t h e later dates at the bottom. This
facilitates following t h e trend from the "known" into t h e "unknown."
Example A-1. Preferred:
1920-1930 _
1940--1950 I __
1960 1
1970 11

Estimated.

176

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Example A-2. Rarely desirable:
1970 1__
1960 1__
1950 11940
1930--1920
1

Estimated.

T I M E PERIODS

SEC.11-B

PAR.

1122

b. Arrangements involving past t i m e . - A r r a n g e m e n t here also depends largely
on presentation purpose. The following general statements are supplemented by
specific suggestions in paragraphs 1123 and 1124, below.
(1) Backward in t i m e . - I n a report devoted primarily to presentation of
statistics concerned with the present, the most recent time period should appear
first.8 In such cases, statistics for earlier time periods appear, primarily, to aid
in interpreting the current information. It is assumed, therefore, t h a t from the
standpoint of the present the user is looking into the past, and t h e materials are
arranged accordingly. Example: Any decennial or quinquennial census report.
Example B-1. Recommended:
1940--1930--1920--19101900

Example B-2. Inappropriate:
1900
1910-

1920--1930--1940---

(2) Forward in t i m e . - T h i s arrangement is appropriate in a report devoted

to presentation of a historical progression or development, as such.° Here, the
current statistics are shown merely to indicate t h e latest development in, or endproduct of, t h e phenomena reflected in the series. Arrangement of statistics
should logically follow t h e development [in discussion; t h a t is, the earliest statistics
should be listed first. Example: A history of t h e Census of Population, or a
report on developments in cotton production since 1850.
Example C-1. Recommended:
1790--1800--1810--1820--1830---

Example C-2. Inappropriate:
1890--1880--1870--1860-1850--

1123. Order of listing: Decennial, quinquennial, biennial, and
annual data, and special surveys.-The following statements cover
normal treatment for presentation of statistics based on the indicated
types of censuses and surveys.
a. Decennial and quinquennial census d a t a . - A r r a n g e backward in time since
the reason for the report's existence is to present the statistics gained in t h e
particular census.
b. Annual surveys, not involving data by m o n t h s . - A r r a n g e backward in time;
t h a t is, place the most recent year first. For listing order for statistics involving
weeks and months, see par. 1124, below.
8 The Statistical Abstract constitutes the outstanding exception to this practice among publications of the
Bureau of the Census. Since its first issue (that for 1878), the Statistical Abstract has placed the earlier time
periods at the top of the list, although the primary purpose of the annual volume is to provide current statistics. This style bas been retained for two reasons: (1) Only a small proportion of tho type of the Statistical
Abstract is set new each year. For most tables, the old type is retained and figures for an additional year
are added, with figures for an earlier year deleted to make room. This creates difficulties in making over-all
changes in existent Statistical Abstract style. (2) When, after many years, the type became badly worn and
it was found necessary to reset completely for the 1947 edition, it was intended that the order of listing be
reversed so as to place the latest figures at the top of the stub (and in the first column of statistics). However, each of 1,059 tables would have had to be recopied by band, or on the typewriter, to provide acceptable
printer's copy. Since the considerable expense involved did not seem justified, the traditional arrangement
was retained.
9 A Statistical Abstract supplement (Historical Statistics of the United States, 1789-1945) constitutes an
outstanding exception to this rule among publications of the Bureau of the Census. Briefly, the supplement
in question presents about 3,000 statistical time-series, with source citations and some annotation. Because
of the nature of the mass-presentation problem, the data could be presented in the least number of pages if
all series were arranged with the most recent year first. Here again, mass-production requirements, and the
cost of printing, were the determining factors, rather than analytical considerations, as such.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

177

PAR.

1123

THE

sEc. 11-B

S T U B : STANDARD LISTINGS

Recommended:

Inappropriate:
1941
1942 _
1943--1944--1945--

1945--1944-194
19421941

c. Special s u r v e y s . - W h e r e statistics from earlier surveys or censuses also are
shown, place the current figures first, since their presentation usually represents
t h e purpose of the report or table.

1124. Order of listing: Days, weeks, and months.-Listings involving days of the week, weeks of the month, and months of the
year are more readily grasped by the user if they are arranged to run
forward in time. Whereas most readers can readily grasp a backward listing of years and decades, a backward listing of months of the
year or days of the week tends to be confusing.
In the work of the Bureau of the Census this problem arises most
frequently where a monthly series is presented covering several
years. Three possible stub arrangements are listed below with arrows
to indicate the eye movements required to read forward in time.
In the second case, note the "rotary" reading forced upon the user;
note, also, the difficulty in comparing December 1944 with January
1945.
Recommended:

V

1944 __

--

Confusing:

V

--

1945 __

Unnatural:
--

1945 __

January_February_.March_-

January_-February __
March_-

December_November_October.-

October_November_December ___

October_November_December_--

MarchFebruary __
January __

V

1945 __

--

January ___
February_March_
October_November_December___

V

3

1

1944

1944-

January__February__
March_-

December ___
November_October_-

October_November_December_--

March_February_January_-

Sec. 11-C. Quantitative and Qualitative Classifications (1131-1133)

1131. General.-Census listings involving quantitative and qualitative classifications are so numerous and varied in type they cannot
be covered in this manual. In all fields, however, more or less standard
stub listings have been developed and are well illustrated in the published materials of the Bureau. The standard arrangements should be
followed meticulously since omitted lines, lines out of proper order,
changed word-order, or differences in wording, can cause considerable

178

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEo.

11-C

QUANTITATIVE CLASSIFICATIONS

PAR.

1131

a

difficulty in proof. Also, such variations can be
real nuisance to
the user if they should pass unnoticed in the publication process.
1132. Quantitative classifications.-The general rule is to arrange
the classes in order of magnitude; that is, magnitude as shown in the
class caption, not in terms of number of cases covered by the individual
class. Thus, in the standard 5-year age classification, the class "Under
5 years" is listed first, not the class which includes the greatest number
of persons.
Note: Order of rank, or magnitude, of number of cases reported is a special
arrangement used only in special-purpose tables, with the basis of the listing
order clearly indicated in the title or headnote.

1133. Qualitative classifications.-The problems of qualitative
classification are too complex and varied to be stated in a general rule.
F or most classifications used by the Bureau of the Census, standard
listings have been developed, ranging from those two or three lines
long to those many pages in length. These should be followed carefully except where there are strong analytical reasons for doing
otherwise. (See also par. 835.)
Sec. 11-D. Clarifying Relationships Between Stubs of Varying Detail

(1141-1144)
1141. General.-Frequently, when statistics are being presented
for a given field, they are shown in detail of classification varying
from one table or report to another. When planning standard stubs of
different degrees of condensation or expansion, and when improvising
such stubs to suit a given purpose, the first consideration is to insure
convertibility from the standpoint of content. Once this has been
achieved, attention should be turned to devices which may aid the
reader in noting interstub relationships. The following discussion
relates to such visual aids. (See also par. 1030 concerning the tie-in
of stubs by standard placement of space breaks.)
1142. Prime lines in summary list should appear as subtotal or
prime lines in the more detailed listings.-The relationship between
line captions in stubs of varying detail can be made more readily
apparent to the user if each line shown in a summary or condensed list
also appears, usually as a subtotal or as a colon line, in the listing of
next greater detail.
a. It makes stub relationship obvious.
b. It automatically provides an index of composition of the given lines.

776554
49
13

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

179

PAR.

1142

THE

Intermediate list

Detailed list
This is accurate

Artists and art teachers_.Authors, editors, and reporters
Chemists, assayists,

sEc. 11-D

S T U B : STANDARD LISTINGS

Artists and art teachers_.Authors and editors-Reporters_Chemists, assayists,

(but)

This is easier to relate to the
intermediate list
Artists and art teachers _
Authors, editors, and
reporters:
Authors and editors-Reporters_Chemists, assayists,

1143. Residual items.-The residual is literally what its name
implies; that is, it represents the residue, or left-over portion, of a
category after all components have been shown separately which have
analytical usefulness in terms of the given presentation purpose.
Special problems arise in phrasing such lines, largely because of a
failure to differentiate residual line captions from specific line captions
in the following respects:
a. Nature of the " r e s i d u a l . " - A "residual" is not an entity in itself. It has
significance only in terms of t h e other members of t h e classification appearing with
it in t h e given table. It cannot stand alone, or out of its original order, unless it
is shown in quotes and carefully defined. In contrast, t h e specific line caption
represents more nearly a homogeneous class or subclass a n d is better able to s t a n d
by itself.
b. Appears after last specific entry in same c l a s s . - S i n c e it constitutes " t h e
remainder," t h e residual always appears at t h e end of t h e listing to which it
applies; t h a t is, it follows t h e last specifically listed component of t h e whole of
which t h e residual is a part. T h e coverage of a residual entry is always determined by t h e number and coverage of t h e items listed immediately above it. In
contrast, as long as t h e specific line caption is under t h e appropriate headings its
coverage is not affected by its relative position although improper or inconsistent
placement may be confusing.
c. Not included in comparative ranking of categories.-A residual is not ordinarily included in a discussion or listing of t h e members of a classification in terms
of relative occurrence or rank. When items are ranked in order of magnitude, t h e
residual ordinarily remains at t h e end, irrespective of size. T h e observations to
which it refers are not homogeneous except t h a t they are all "not in a n y class listed
above."
d. Not subject to consistency in v a l u e . - T h e residual is not subject to t h e
requirement of consistency in numerical value from one table to another in the
same manner as are all specifically described members of t h e classification. Thus,
t h e residual category "Other races" m a y and does vary in size from one table to
another, or even within t h e same table, depending upon t h e degree of race detail
shown separately before t h e residual is reached.
A residual merely makes up t h e difference between t h e total and t h e sum of t h e
entries for t h e classification members shown separately. Hence, in a classification
of greater detail, t h e residual entry, though described identically, may well be
smaller in numbers t h a n in a s t u b of lesser detail.
Residual treatment is theoretically correct in all three of the following examples.
However, for preferred treatment of example C-1, see example C - 2 , par. 1144,
below.

180

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEc. 11-D
Example A-1:
All classes_White_-Negro_X Other races_-

INTERSTUB CLARIFICATION
Example B-1:

350
150
125
75

All classes_White_-Negro_Indian __
Chinese __
X Other races--

PAR.

1143

Example C-1:
350

150
125
10
40
25

All classes_White_-Negro_Indian_
Chinese __
Japanese_
X Other races_--

350
150
125
10
40
15
10

1144. Maintaining identity of residual.-Although not essential
from the theoretical standpoint, there may be a decided advantage in
maintaining a residual identity where feasible.
a. Aids in clarifying stub interrelationships.-If t h e position (and coverage)
of t h e residual is shifted between tables, it may increase t h e reader's troubles when
comparing t h e stubs. Note, below, t h a t t h e relationship between stubs A - 2 and
C - 2 is more readily apparent t h a n between stubs A - 2 and B - 2 .
Erample A-2. S u m m a r y
list:
All classes_.
350
150
White_-Negro_125
75
x Other races_-

Example B-2. R e s i d u a I

shifted in detail list:
All classes __
White_-Negro_Indian_.Chinese_Japanese_
X Other races--

350

150
125
10

40
15
10

Erample C-2. R e s i d u a l

maintained in detail list:
All classes __

White_Negro __
X Other races:
Indian __
Chinese_Japanese_
All other_-

350
150
125
10

40
15
10

b. "All other" versus "Other."-Where residuals are maintained as in example C-2, above, a terminology question arises; t h a t is, " I s it permissible to have
t w o line captions of identical wordings, but different coverage, in t h e same stub,
as long as they are on different indent levels?" Theoretically, t h e answer is
"Yes." In practice, such cases should be avoided.
Where subresiduals appear, it frequently is helpful to t h e user, as well as to those
preparing the tables, if t h e wording of t h e subresidual is modified.
(1) Use "Other," or its equivalent, for t h e first level of residual.
(2) Use " A l l other," or a similar modification, for a subresidual.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

181

Chapter 12
THE BOXHEAD (1201-1272)
Sec. 12-A. General (1201-1211)

1201. Definition. The table boxhead is that portion of the table
in which is located the individual column heads or captions describing
the data in each vertical row or column, together with needed classifying and qualifying spanner heads.
Placed at the top of the table, the boxhead is separated from the
title and headnote by a parallel top rule. The boxhead is set off from
the stub and field, which appear below it, by a single horizontal rule
running the full width of the table.
1202. Component parts defined.-In terms of construction, the
component parts of the boxhead are few in number. For illustration
of the various parts of the boxhead, see fig. 18, pp. 184-185. Some
of the more important components may be defined as follows:
a. Stubhead or stub b o x . - T h e column head or caption of t h e s t u b which
describes the s t u b listing as a whole. Although it is physically a part of t h e boxhead, it is classified with the stub for discussion purposes. (For treatment, see
sec. 9-A.)
b. Individual column head or c a p t i o n . - T h e basic unit of t h e boxhead. A classifying, descriptive, or qualifying title which is located directly above the individual column to which it refers and is framed on either side by the column rules
of t h a t column. It may or may not be qualified, supplemented, or described by
one or more spanner heads above it. (See secs. 12-B and 12-C.)
c. Spanner h e a d . - A classifying, descriptive, or qualifying caption spreading
across (and above) two or more individual column heads or across two or more
lower spanners. In its normal range, it applies, in varying degree, to all columns
or subordinate spanners thus covered. This is t h e boxhead counterpart of t h e
s t u b center head or subhead and of the colon or dash (read-in) line-caption.
(See sec. 12-E.)
d. Undercut spanner.-Undesirable. A spanner appearing on t h e lowest level
of box directly over, and spanning, two or more columns. Since it appears
below t h e individual column heads, it breaks t h e column rules and separates t h e
column captions from the d a t a which they describe. Most commonly it is used
to specify t h e presentation unit for all columns covered. T h e undercut spanner
should be used sparingly. (See sec. 12-G.)

182

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

smc. 12-A

GENHRAL

PAR.

1202

e. Banner head.-Undesirable. A special type of spanner extending across and
covering the entire boxhead of the table (except the s t u b box). It should be used
only under carefully restricted circumstances. (See sec. 12-F.)
A quasi-banner is the same as a banner except t h a t t h e total column is not
covered. This form is useful but, at times, may be subject to the same criticism
as t h e banner head. (See par. 1264.)
f. Column n u m b e r s . - R a r e l y used by t h e Bureau of t h e Census. A device for
numbering columns to facilitate reference. Normally, Arabic numerals are used
a n d the s t u b is not included in t h e numbering.
Where present, column numbers may be located within an auxiliary single-level
box running directly under t h e boxhead proper, and set off above and below by a
single horizontal rule. In such cases, parentheses are not used.
On occasion, column numbers are shown parenthetically within t h e individual
column boxes.
g. P a n e l . - A distinctive segment of t h e boxhead consisting of a group of
related column heads with their attendant spanners, frequently a self-contained
unit. It represents t h e boxhead counterpart of t h e s t u b block. T h e term "panel"
is largely a term of convenience used when referring to distinctive groups of
columns.

1203. Heads in field, not in b o x . - T h e following heads are not a
part of the table boxhead, as such, but are commonly confused with it.
Although they perform a function similar to that of the boxhead, they
appear in the field of the table. Their use is discussed in detail in
Chapter 14, The Field.
a. Field spanner.-Undesirable, but necessary in special cases. A type of
spanner appearing recurrently within, and constituting a component part of, t h e
field. I t s confusion with t h e boxhead may be minimized if t h e field spanner is
thought of as a stub center head transposed into t h e field in cases where(1) No room is available to repeat t h e stub, and
(2) The omitted subentries for the spanner would, if shown, be identical with
those in t h e s t u b as it appears at t h e left.
A common error is t h e inclusion of t h e uppermost of a series of field spanners
within t h e boxbead of a table, usually as t h e uppermost spanner of t h e boxhead.
This practice leads to confusion.
Compare this form (example A-1, par. 1431) with t h e repeated boxhead found
in divide tables (example A-2, par. 1431) and with t h e undercut spanner (example
A-1, par. 1271).
Note: At times, field spanners are used where stub center heads are too long
to be shown in a narrow stub. Here, each head is r u n in the field, set off by rules.
This should be avoided unless absolutely necessary.
b. Unit-indicator.-Normally, a component of t h e field of t h e table, not of t h e
boxhead, even though t h e lower box may, in special cases, be used for this purpose.
(See sec. 1 2 - G . ) T h e unit-indicator, as such, is a device designed to make clear
t h e exact unit of measurement to which t h e given statistics refer. When it
appears at t h e top of a column, below t h e boxhead, it normally is set in italic in
letterpress; it is usually placed in parentheses in typewriter-offset.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

183

FIG.

18

FIGURE

THE B O X H E A D

18

SEC.

12-A

- B O x H E A D S AND T H E I R COMPONENT P A R T S (See p a r s .
1201-1203)

Example A. Boxhead with two levels:

Division and S t a t e ®

Operated by nonwhite
owners

Operated by white
owners®

Total
farms

Part
Full
Part
owners owners Total @ owners owners

Total

F

.

@ Parallel top rule.
@ Stubhead.
© Single bottom rule.

® Individual column head.
E Spanner head.
® Panel.

Example B - 1 . Boxhead with three levels and with column numbers separated
by horizontal rule:
Living in the specified
region @

Born in the specified
region @

Region and sex

Total

1©

@ Upper spanner.

Born in
Living in other and living in the
regions
specified
region
Num
Perber
cent

2©

3©

4©

® Lower spanner.

Per
cent

Net gain
( + ) or
loss(-)
through
interregional
movement
(col. 5
minus
col.l)

7©

8©

Born in other
regions®
Total
Num
ber

5©

© Column number.

Example B - 2 . Boxhead with four levels and column numbers in parentheses:
Consumption @

Stocks in consuming establishments at end of year@

Cotton, excluding linters
Domestic@

Yearly average
or year
Total
Total
(2)@

@ Upper spanner.
® Lower spanner.

184

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Upland

Sea Island

American

For
eign

(6) @

Linters

Cotton, excluding
linters @

Total
(7) @

ForDomestic eign

(8) ®

. © Third-level spanner.
@ Column number.

Lintters

(10) @ (11) @

sEc. 12-A

GENERAL

F I G U R E 1 8 . - B O x H E A D S AND

FIG.

T H E I R COMPONENT PARTS

18

(See pars.

1201-1203)-Con.
Example C. Boxhead with undercut spanner:

Group or industry

Census
year

Value
Cost of
Wage Wages materials Value
added by
of
Number earners
manupurproducts
of estab- (average
facture
chased
lishfor the
ments
year)
@ In thousands of dollars

@ Undercut spanner comprising a unit indicator.
Example D. Divide table (compare with example E, table with field spanners) :
The State

Urban

Country of birth
Total

Male

Female

Total

Male

Female

England_-Scotland ___
Azores_All o t h e r Rural-non farm

Rural-farm

Country of birth @
Total

Male

Female

Total

Male

Female

England ___
Scotland ___

® Repeated box.

@ Repeated stub.

Example E. Table with field spanners (compare with example D, divide table);
field spanners are not a part of t h e boxhead:
Country of birth

Total

Male
The S t a t e @

Female

Total

Male

Female

Urban

England_.
Scotland _
Azores.-All other_Rural-nonfarm

Rural-farm

England_Scotland_-

@ Field spanner.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

185

Pax.

1203

T H E BOXHEAD

s c . 12-A

For a more detailed statement concerning t h e nature and use of this device in
t h e stub, see sec. 9 - F . For treatment in t h e field, see sec. 14-C. Although
details of placement differ between s t u b and boxhead, t h e general principles
involved are t h e same wherever t h e device is used.

1204. Limitations on column width, caption lineage, and number of
box levels.-Arbitrary rules on these points are not possible since the
situation necessarily varies from table to table, and even within the
boxhead of the given table. However, a few general principles may
be laid down, as follows:
a. Column w i d t h . - A major objective in Bureau of the Census tables is to
present t h e maximum amount of d a t a within the given space. Therefore, t h e
width of t h e column box normally should be based on t h e horizontal space required
by t h e largest figure to be shown in any cell in t h e given column, usually t h e grand
t o t a l figure. In other words, where possible, t h e column width should be determined by space demands of t h e cell entries and t h e caption adjusted accordingly.
A desirable column width would be at least two spaces more t h a n t h a t required
by t h e maximum cell entry. This allows for a "bear-off" for the maximum figures
of one space on either side of each column rule.
b. Lines per c a p t i o n . - C o l u m n captions more t h a n three lines long should be
avoided where possible, particularly where t h e column box is extremely narrow.
In particular, captions involving many 2- and 3-letter hyphenations are hard to
read, as are words which are hyphenated twice because they cannot be fitted into
two lines.
c. Number of box l e v e l s . - W h e r e possible, hold the number of box levels to
not more t h a n three; two is preferable as a maximum. If four levels are essential,
it is desirable that one level indicate merely t h e unit of presentation or contain
some other t y p e of caption readily grasped. Beyond three levels, t h e boxhead
becomes too complex for many readers to follow.

1205. Space problem caused by wordy column captions.- Column
captions and other boxhead captions must necessarily make clear the
nature of the data. However, excessive wordiness not only wastes
space but may require sacrifice of columns of data.
a. Effect on columns per p a g e . - T h e number of categories which can be shown
in t h e box on a n y given page is strictly limited. Excess wordage in column boxes,
even where only two more spaces are added in each column, may squeeze two or
more columns from t h e page.
Example: A census-size table m a y be thought of as being 172 typewriter spaces
wide. If the s t u b occupies 24 typewriter spaces, a n d t h e maximum cell entry in
each column takes 6 spaces (99,999), a maximum of 21 columns is possible, since
1 space must be provided for each column rule. This represents a squeezed table
a n d t h e above figures assume, also, t h a t no column caption will exceed t h e 6-space
width. Expansion of all column captions to an 8-space width will mean deletion
of about 5 columns of data from t h e page.
b. Effect of continuing box to next p a g e . - W h e r e wordy column captions
squeeze columns over to t h e next page, considerably more publication space m a y
be required t h a n t h e amount necessary to show those particular columns, if only
because of repetition of t h e s t u b on t h e new page.

186

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEc. 12-A

GENERAL

PAR.

1205

(1) S t u b r e p e t i t i o n s . - S t u b repetition is expensive in terms of space. In
the above example, t h e 24-space stub, narrow though it is, takes only a little less
t h a n four 6-space, or three 8-space, data columns. Many stubs may run double
or triple this width.
c. "Last page" treatment for continuing box as compared with continuing
stub. If the stub continues, t h e left-over on the last page is run as a fraction of
a book-page as long as a reasonable number of lines of data are to be shown. T h a t
is, the last page of a table need not occupy all the vertical space of the book-page.
Another table may be started on t h a t page, or the remainder of the [page may,
in some cases, be left blank.
Where the box continues, however, the last table-page must cover the full width
of t h e book-page. Hence, if the number of columns (with normal spacing) is too
few to extend across t h e full width of t h e last table-page, one of four steps must
be taken:
(1) T h e columns on the last page must be widened;
(2) More columns must be brought over from previous pages;
(3) The columns on the last page must be pushed back into previous pages
and the "last page" eliminated; or
(4) The last table-page must be run fractional-measure with the stub doubled
or tripled up. (See par. 215 for definition, and fig. 4 for illustration, of fractionalmeasure tables.)

1206. Colon heads not used.-Whereas, in the stub, a caption ending in a colon serves as a subhead (pars. 813 and 921), the construetion of the boxhead prevents use of a similar device. Instead,
spanner heads must be employed for the purpose. That is, a colon
line, as it appears in the stub, becomes a spanner head if shifted to the
box.
Example A. In stub:

Example B. In box:

Color of head:
White_--

Nonwhite.-

Color of head
White

Nonwhite

a. Case of single subentry.-In the s t u b where a "colon line" has only one
subentry, the subentry is run in with the colon line and t h e colon preserved
(par. 924). Where an equivalent situation arises in t h e box, the same basic
principle is observed. Spanner heads must cover more t h a n one column.
Example C-1. In stub (right):
Increase: 1930 to 1940-Example C-2. In stub (undesirable):
Increase:
1930 to 1940.---

Example D-1. In box
(right):

Example D-2. In box
(wrong):

Increase:
1930 to 1940

Increase
1930 to 1940

1207. Captions should run across, not up.-Boxhead captions should
be planned to run across the page, not up; that is, they should be
run in the same direction as the stub and the table title, not at right


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

187

PAR.

1207

sEc. 12-A

T H E BOXHEAD

angles to them. This practice should be maintained even though it
means heavy use of abbreviations or word-divisions. Run captions
"up" only as a last resort.
Example A-1. Preferred:
Municipally
owned electric
utilities

Example A-2. Acceptable:

Example A - 3 . Undesirable:

Municipally
owned
electric
utilities

EE

1208. Rules for running captions up.- When necessary to run captions "up" instead of across, several basic rules should be followed:
a. Must read up the page, not d o w n . - C a p t i o n s running " up," as t h e term
implies, should always start at t h e bottom. of t h e box and read up. They should
not start at t h e top and read down. T h i s practice is universal and t h e reader
is accustomed to it. Exception: See paragraph 1208c, below, concerning broad
tables.
b. "Run up" only those entries which cannot be inserted horizontally.-If it
is essential to run one caption " u p , " there is no need to run all of them up on
t h a t page or deck. R u n up only those where horizontal insertion is impracticable.
Avoid run-up on spanner heads even though t h e subordinate column heads run
up. Never run up a read-in caption (caption ending in a dash). Common
sense should prevail, however. Thus, if only one or two column heads can be
run horizontally and t h e rest must be run up, it would be sensible to run all of
them up. In case of doubt, run everything horizontally t h a t can be inserted
t h a t way.
Example A - 1 . Preferred:
Example A - 2 . Undesirable:
Note: Examples assume that space demands that the "white" captions must be run up.

Age

Total

Negro Other

0

Age

aB
E

0

c. Undesirable on broadside t a b l e s . - C o l u m n captions on broadside tables
should always read crosswise; t h a t is, they should always run in t h e same direction as t h e title and stub. Never run captions up in a broad table.
( I ) Reason w h y undesirable.-Broad tables are used where t h e table is
too wide to be run upright. T h e y are objectionable in themselves since t h e
user must t u r n t h e report sidewise to read them.
Insertion of run-up heads in a broad table forces t h e reader to t u r n t h e book
upside down to read them. Comparison of d a t a in such tables with d a t a in
upright tables in t h e same report becomes constructively impossible, particularly if comparisons of box wording are also necessary.

188

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEo. 12-A

PAR.

GENERAL

1208

Example B-1. Upright table:
Total

Age

White

Negro

600

300

Other
0130N

All ages-

1,000

100

0JI4A

uroq

Note: Comparison of column captions in example B - 1 with those in B - 2 is constructively
impossible. T h e difficulty is increased where
(a) t h e captions are long and involved, (b) an
important distinction lies in t h e difference of a
single word, a n d (c) t h e tables are several pages
apart in a bound volume.

I810,L

0

E

(2) A l t e r n a t i v e . - T h e only possible alternative, other t h a n redesigning the
table, is to make such captions read down, instead of up, on broad tables. If
this were done, the user, to read them, would tend to restore the report to a
normal position; this would simplify comparisons with upright tables. However, the only real solution is to redesign the table.

1209. Style and punctuation.-The same general principles apply
as for other parts of the table.
a. Style.-Telegraphic style is used as elsewhere in t h e table. However,
because of space limitations, special attention must be paid in boxhead writing to
reduction of wordiness and to emphasis upon presence and location of keywords
or phrases.

b. Punctuation.-Normal punctuation is used. Special care is required since
a misplaced punctuation mark coupled with the extreme condensation of phrasing
may obscure the meaning of the column head or even change it completely.
No punctuation mark is used to terminate captions as such. T h e dash ( - ) is
used for read-in beads, but note that its appearance means that the caption is incomplete. (For definition and explanation of dash or read-in lines, see par. 926.)
Terminal periods are not used, but abbreviation periods are.
Example A - 1 . Right (no terminal periods):

Example A-2. Wrong (terminal periods are incorrect):

Number of employees
Number of employees.
Total

School

Nonschool
Total.

School.

Nonschool.

Example B. Right (periods are for abbreviations, not to indicate ends of captions) :
Total

Jan.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Feb.

Mar.

Apr.

May

June

July

Aug.

189

FIG. 19

THE

SEC. 1 2 - A

BOXHEAD

F I G U R E 1 9 . - U s E OF A B B R E V I A T I O N S TO D E C R E A S E D E P T H OF B O X -

HEAD

Example A.

Spelling-out increases box depth:
• Persons not In labor force

Area, color, and age
Total

Example B.

Percent of total

EmEnploygaged
Engaged in In institu- Employment in
In
ment
own
status not
own home
status
home insti
tions
reported
housework
1
house- tutions not
rework
ported

Consistent abbreviation decreases depth, but diminishes clarity:
Persons not in labor force

Area, color, and age
Total

In own
home
housework

staIn institu- Empl.
tus not
tions
reptd.

Percent of total
In own
home
housework

Empl.
In
insti- status
not
tutions reptd.

Example C. Spelling-out, where possible, gives reader clue as to meaning in
the panel where the same words are abbreviated to save space:
Persons not In labor force
Area, color, and age
Total

190


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Percent of total

own
Engaged in In institu- Employment Inhomo
own home
status not
housetions
reported
housework
work

Empl.
In
insti- status
not
tutions reptd.

sE0.

12-A

GENERAL

FAR.

1209

c. Abbreviations.-Abbreviations may be used heavily if t h e depth of head is
materially reduced thereby. Use of an abbreviation in one column or spanner
box does not demand its use in all column boxes on t h a t page. Where nons t a n d a r d abbreviations are "invented" for use in a tight portior of a boxhead, it
will help to make the meaning clear if the term is spelled in full in those boxes
where sufficient space is available. In the example (see fig. 19, p. 190), note t h a t the
obvious identity of corresponding column heads in the two panels of the table even
makes possible shortening of the "housework" caption without danger of confusion.

1210. Capitalization.-The current style of the Bureau of the
Census calls for setting or typing all boxheads in lower case (This is
in lower case). Full caps (THIS IS IN FULL CAPS), small caps
( T H I S Is IN SMALL C A P S ) , and caps and lower case (This Is in Caps and
Lower Case) are no longer used in tabular boxheads of the Bureau of
the Census.
The use of caps (or of small caps) in the stub box, and in the uppermost spanners of the boxheads, was traditional in the tabular style
of the Bureau of the Census. Analytical objections to this practice
led to its abolition, effective February 4, 1948.
1211. Vertical and horizontal spacing of captions.- T h e following
rules reflect standard practice.
a. Vertical spacing:
(1) Box captions are centered vertically within their box.
up at the top or bottom.
Ezample A-1. Right:
Total

White

Example A-2. Wrong:

Negro Other

races

Total

White

They do not line

Example A-3. Wrong:

Negro Other

races

Total

White

Negro

Other

races

(2) They are run solid; t h a t is, no space is allowed between the lines, irrespective of t h e space available in the given box, openness of t h e remainder ot
the table, or the amount of space which may be left blank on the page because
the table runs short.
Example B-1. Right:
Native
white

Example B-2. Wrong:
Native
white

(3) Depth of box is controlled by the deepest box (the controlling box) on
the same level, unless t h e total depth of boxhead would thereby be increased.
In examples C-1 to C - 4 , column captions shown as A, B, C, etc., represent
1-line captions.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

191

PAR.

1211

Example C-2. Wrong (no saving in depth):

Example C-1. Right:

Total

This is the
deepest box
on top level
C

D

Total

Other

E

F

A

G

Example C-3. Right (saving in depth justifies
this):
This deep box
usually controls depth
of top level

Total
Three
line
head

sEo. 12-A

THE BOXHEAD

B

This is the
deepest box
on top level
C

Other

F
D

E

Example C-4. Wrong (if lower rule of spanner
does not line up with other spanners on its level,
it must line up with spanner on next lowest
level. Never leave it hanging between):

Other

B
C

D

E

(4) In t h e controlling box, a 1-em space (letterpress) is left above and below
t h e horizontal rules defining the upper and lower limits of the box. In the
remaining boxes on t h a t level, the captions are centered vertically. In typewriter-offset, either one line or one-half line may be left blank above a n d below
t h e rule. Examples D - 1 to D - 3 relate to t h e controlling box only.
Example D-2. Wrong (too open):

Example D-1. Right:
Sound spacing
for the
controlling
box

Too much
space above
and below

Example D-8. Wrong
(too tight, avoid):
Too little
space above
and below

b. Horizontal spacing:
(1) Each line of each boxhead is centered in the given box; the lines are not
blocked. R u n each line as wide as possible; leave only one space blank on
either side of column rules. Exception: Where a box is 10 ems or more wide as
printed (20 characters or more in typewriter), a caption making 3 lines or more
is set or typed with a hanging indent.
Example E. Illustrations of horizontal spacing:

T h i s is
right

This
is wrong

192

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Run first line full
width

Do not run
lines short
in center of
wide column

sEc. 12-A

PAR.

GENERAL

1211

Example F-1. Right (typewriter-offset):

This caption is in a wide spanner or column that is more than 20
characters wide, hence this hanging indent is appropriate
for a 3-line caption

Example F-1. Right (letterpress):
This caption is in a wide spanner or column that Is more than
10 ems wide as printed, hence this hanging Indent is
appropriate for a 3-line caption

(2) Comparative line l e n g t h . - W h e r e v e r practicable use an inverted
pyramid effect; t h a t is, make t h e first line the longest. (See example E, above.)
This makes for easier reading. However, 2-letter word divisions should be
avoided. Also, if a key phrase or word combination would otherwise be split,
it might be helpful to t h e user if the first line were shortened to preserve t h e
conceptual unity. Finally, it is better to run t h e first line short if running it
full width will result in two successive lines ending in word divisions.
c. Spacing for run-up h e a d s . - W h e r e heads run up, they are not centered in t h e
width 1 of t h e box. This rule applies both to letterpress (type-set) composition
and to typing for offset reproduction. Instead, t h e first line of t h e given box is
given a bear-off 2 (from t h e horizontal rule) of one en in letterpress; one typewriter
space for offset. For t h e 2-line run-up heads, letterpress and typewriter-offset
procedure differs, as indicated below.
(1) 1-line r u n - u p h e a d s . - I f it is a short line, start one en (letterpress) or
one typewriter space (typewriter-offset) from the bottom rule and let the line run
short.
Erample G. Right:

Example H. Right:

a
rn
0)

0

bl)•-

rn

0

(2) 2-line r u n - u p h e a d s . - I n letterpress, t h e first line is run full width of
the box (except for bear-off at either end). The second line is centered under
the first. In typewriter-offset, t h e first line will rarely come out even to full
length of box, and attempts to center second lines on either box width or on first
lines will create a ragged appearance. Therefore, a hanging indent is used for a
run-up head of two lines or more; t h a t is, the second and following lines are indented one space under t h e first line. All lines run as close to t h e right-hand
edge as possible (except for a bear-off of one space). No attempt is made
to space them on the typewriter to line them at the right.
1

Where heads run up, "width" ls the dimension otherwise referred to as "depth."
is the space left between the'rules and the beginningor ending of a line.

2 "Bear-off"


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

193

PAR.1211

THE

sEc. 12-A

BOXHEAD

(3) 3-line h e a d s . - A hanging indent is used, both in letterpress and typewriter-offset. In letterpress, the overrun is indented one en; in typewriter-offset,
it is indented one space.
Example J-1. Right (In letterpress):

0

O

0

-

0

s:I O 0.

E
Example J-2. Right (In typewriter-offset):

0

.....o Ill
..... 0
b0·.-1

o > o

Q ) Q )

•rl

Sec. 12-B. The Column Head or Caption (1221-1227)

1221. Definition.-The basic unit of the boxhead. T h e column
head or column caption is a classifying, descriptive, or qualifying title
of an individual column. It is located directly above the individual
column to which it refers and is framed on either side by the column
rules of that column. It may or may not be qualified, supplemented,
or described, in turn, by one or more spanner heads above it.
1222. General class of columns: Prime and nonprime. Prime
columns are those columns of data constituting the hard core of the
table. From them, all other, or nonprime, columns of data can be
derived by mathematical processes. This classification is significant
in table planning since, in general, inclusion of prime columns
increases the amount of basic information provided the user; inclusion of nonprime columns contributes to his convenience. Ordinarily,

194

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

src. 12-B

THE

COLUMN HEAD OR CAPTION

PAR.

1222

if columns must be omitted because of lack of space, it is best, in
Bureau of the Census presentation, to reduce first the number of
nonprime columns other than basic totals.
A detailed statement of this classification is presented in pars. 942-944 where
t h e discussion is in terms of t h e stub. T h e same general principles apply to the
boxhead.

1223. Additive and nonadditive columns.-A classification of columns useful in placing vertical rules. (See sec. 13-C, below, for
placement and use of vertical rules. See also par. 945 for discussion
of additive and nonadditive lines.)
a. Additive c o l u m n . - A n y one of two or more columns of entries which, taken
together, will add to a total or subtotal column shown in t h e table. T h e individual
columns in such a "group" may appear consecutively, or they m a y be dispersed.
Each may, in itself, be prime or nonprime; thus, one of t h e group may be a
prime d a t a column, and one (or more) of the others may be a subtotal (nonprime) column.
b. Nonadditive c o l u m n . - A column not a member of a n y group of columns
which together add to a total or subtotal shown in the table; or any column not
subject to, or not intended to be subjected to, the additive process, such as a line
of means, medians, etc.

1224. Word order.-Carefully phrased column heads can clarify
the whole basis of column arrangement. In general, it is well to
stress differences between adjacent columns rather than similarities.
That is, place at the beginning of each column head the key word or
phrase which indicates the point of difference or of contrast. However, at times the opposite procedure is desirable, particularly when
it is important to stress the unity of a group of columns.
a. Stressing difference.-Within a related group of columns, it is ordinarily
desirable not to have two or more adjacent column heads start or close with t h e
same phrase. Frequently, this is a signal t h a t a spanner head is needed.
Example A-1. Preferred:
Native white
Native
parentage

Foreign or
mixed
parentage

Example B-1. Preferred:
Electric utilities
Privately
owned

776554
49

Municipally
owned

14


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Example A - 2 , Questionable except to save a 4-level box:
Native white,
native
parentage

Native white,
foreign or
mixed
parentage

Example B-2. Questionable:
Privately
owned
electric
utilities

Municipally
owned
electric
utilities

195

PAR.1224

THE BOXHEAD

SEC.

12-B

b. Stressing similarity.-At times t h e exactly opposite holds true in t h a t similarity between columns should be stressed. In general, this is the case where
spanners are impracticable and t h e individual column heads must be used to
indicate column grouping.
Example C-1. Stressing
difference where spanner indicates grouping:

Example C-8. Indicating column pairing by similarity stress-no spanner:
Native
father

Mixed parentage
Father
foreign

Native
mother

Foreign
father

Foreign
mother

Mother
foreign

1225. Column heads should be single level.-Column heads should
not be divided into two levels. That is, a I-column "spanner" is
undesirable. This question most frequently arises as follows:
a. Treatment of unit-indicator.-If a unit of measure is to be included t h e box
should not be split horizontally to create a single-column spanner. Either place
t h e unit-indicator in parentheses immediately after t h e caption, or show it as a
formal unit-indicator above the first cell-entry in t h e field.
Example A-2. Acceptable:

Example A-1. Preferred:
Number
reporting
on value
65
25

Number
reporting
on value

Total
value
Dollars
3, 200
1, 000

65
25

Example A-3. Wrong:

Total
value
(dollars)

Number
reporting
on value

Total
value
Dollars

3, 200
1,000

b. Two classification-levels in one b o x . - I n a table where spanner heads a r e
used for indicating a superior level of classification, one of t h e classes designated by
spanners may have only one component part. This results in a single column
extending to t h e top of t h e spanner level. Here, both t h e "spanner" classification
and t h e individual column classification should be included in t h e single column
head.

Solution: Do not divide t h e box. Place t h e spanner classification first to maintain parallelism; then run in t h e column classification by means of a comma,
preposition, dash, etc., as t h e circumstances warrant. (In t h e following example,
it is assumed t h a t t h e sex classification is shown for t h e urban group, but only the
total for t h e rural group.)
Example B-2. Wrong:

Example B-1. Right:

Total

Male

Rural

Urban

Urban
Female

196

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Rural,
total

Total

Male

Female

Total

sEc. 12-B

THE COLUMN HEAD OR CAPTION

PAR.

1226

1226. Capitalization.-Column heads are invariably set or typed in
lower case; never in caps, or in caps and lower case. (See par. 1210.)
Comparative classification-level, as such, has nothing to do with
capitalization in the box.
Right:

Wrong:

Total
white

TOTAL
WHITE

Wrong:
Total
White

Wrong:
TOTAL
WHITE

1227. Column rules.-Normally, the individual column is set off
by single lightface (hairline) column rules. However, moderately
bold rules (about %-point face) are normally used to set off panels,
sometimes appear after (to the right of) nonadditive total columns,
and occasionally are used as a substitute for the parallel total rule.
For a discussion of use of vertical rules, see section 13-C.
Sec. 12-C. The Column Head: Total and Subtotal Column
(1231-1233)

1231. General.-The total (and subtotal) head is a special case of
the column head, comprising the descriptive title (or caption) of any
column composed of grand totals, group totals, or subtotals.
Most of the discussion of total and subtotal data-line captions (pars.
952-959) applies equally to total and subtotal column captions and
will not be repeated here.
1232. Various meanings of term "total."-In the boxhead as in
the stub, the term "Total" is subject to a number of different interpretations. Some of the more important meanings are listed below.
a. The result of a summation process applied to t h e subentry columns; t h e
sum of t h e parts actually shown separately. This is t h e most commonly understood meaning of t h e term. It is t h e normal meaning in such phrases as " T o t a l
column," " T o t a l entry," etc.
b. The aggregate, or total number, as an independent item. This usage is
typified by such individual columns (without subordinate columns) as " T o t a l
value (thousands of dollars)."
c. The total, or the whole, as a class in itself, or as representing a class of classes.
This is typified by t h e " T o t a l " class in t h e area classification : "Total," "Urban,"
"Rural."

In most instances, as in the stub, these differences tend to be academic in that a given total column may reflect two, or even all three,
of these meanings at the same time.
At times, however, it is important that the reader understand which
meaning is intended or that one of the meanings does not apply.
Here, the table-designer should keep the above distinctions in mind


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

197

PAR. 1232

sEc. 12-C

THE BOXHEAD

and should govern his treatment of spanner heads and of column
heads for total and subtotal columns, accordingly.
See paragraphs 953-958 for application of above discussion. In general, the
statements made there (in respect to t h e stub) apply also to the boxhead. In
most instances the effect of transposing t h e s t u b examples to t h e box situation
will be obvious.

j

1233. The "grand-total" c e l l . - T h e grand-total cell probably
tends to be less well described than any other cell in the statistical
table. Intent upon making clear the description of lines and columns
of data, the table-designer finds it easy to assign "Total" as the caption
of the first line, and "Total" as the caption for the first column.
Occasionally, he may cap this situation by assigning "Total" as the
spanner caption for the first panel. The following result is frequently
passed unnoticed:
Total
Item
Total
Total.--

A

B

X

At t h e left, the grand-total cell
(X) is described as "Total, total,
total." This sends the reader to the
table title for a description which
may not be too well stated there nor
too easy to decipher.

a. Alternative terms for " T o t a l . " - A simple solution of the above problem is
to describe at least one, preferably two, of t h e above total classifications in definite
terms. Thus, if one means t h e total for both sexes combined, substitute " B o t h
sexes" for "Total." If one means t h e total for all race groups, substitute "All
races" or "All classes," etc. T h e third t e r m may be left standing if t h e other two
are described properly, since t h e combined description may reasonably read
" T o t a l , both sexes, all races."
b. General r u l e . - E a c h case must be handled on its merits. In general,
instances are rare where space limitations prevent a solution to t h e problem. T h e
general rule should be:
(1) R u l e . - I n describing grand-total cells t h e single word "Total" should
never appear more t h a n once. If three or more captions apply to t h e given
grand-total cell, no more t h a n one should be labelled simply " T o t a l " ; no
more t h a n one other should have a similarly generalized term, such as "All
classes"; t h e remaining one or more captions should always be explicit in
description.
(2) A p p l i c a t i o n . - T h e above rule is more reasonable t h a n it m a y seem at
first glance. Where more t h a n two captions apply to t h e grand-total cell, t h e
additional captions must necessarily appear either as centered heads in t h e s t u b
or as spanner heads in t h e box. In such cases t h e availability of full s t u b width
or of spanner width simplifies adequate description.
Example A-1. Wrong (spanner and column space available for more detailed description):

Urban

Total

Rural

Age

Total

Total_-

198

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Male Female

Total

Male Female Total

Male

Female

SEC•

12-C

THE COLUMN HEAD: T O T A L

PAx.

1233

Example A - 2 . Right (wise use of spanner space):
The State

Urban

Rural

Age
Total

Male Female

Total

Male Female Total

Male

Female

All ages ..
Example A - 3 . Right (where spanner total is hard to describe,! adjust column heads):
Total 1
Ag0

Both
sexes

Urban

Malo Female

Both
sexes

Rural

Male Female Both
gexes

Male

Female

All ages__1 T h a t is, assume that the above spanner entry ("Total"), if spelled out in full, would read "Balance of
metropolitan district outside central cities." In such a case, it is simpler to adjust the caption of the grandtotal column.

Sec. 12-D. The Dash (or Read-in) Head (1241-1248)
1241. General.-For definitions and general principles see pars.
813, 921, 926-929, and 931 where the "read in" is discussed in terms
of stub presentation. Read-in heads should be used sparingly in the
boxhead. They are best reserved for cases where a major gain in
clarity is involved.
Briefly, the dash or "read-in" caption is a fragment; it is not complete in itself. It is the beginning portion of each caption in the
boxes under it, or of the cell entries if it appears in a column head
and not in a spanner head. Thus, in example A-1, the full caption
of the first column is "Born in New York"; of the second column is
"Born in New Jersey," etc. For further illustration, see par. 1252c,
examples H - 1 and H-2.
1242. Clarification of paired distributions.-Certain paired or cross
distributions tend to be confusing, particularly those involving
identical stub and column captions. (See example, below.) Here,
the dash head may go far to assist the user in reading the table.
Example A-1. Helpful:

Example A - 2 . Helpful:

Born in
Residence in 1940

New
York

Middle Atlantic:
Now York ..
New Jersey ..
Pennsylvania..


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

New
Jersey

Births by
place ofPennsylvania

City
Occurrence
Akron.
Albany __
Atlanta_Boston..

199

Residence

PAR.

1243

sEo. 12-D

THE BOXHEAD

1243. Avoidance of misunderstanding.-At times a misconception
may be avoided by proper use of the "read-in" principle. The following example illustrates this and emphasizes the need for care in
wording boxheads. In example A-1, the read-in head makes clear
that the entries refer to the price for the 1939 crop and the 1940 crop;
that is, prices obtained for products grown in that year. In example
A-2, "Crop price" might mean prevailing price in the given year for
type of crop specified, irrespective of production year.
Example A-1. Clear:

Example A - 2 . Confusing:
Crop price

Price for crop of1939

1940

1939

1940

1244. "Read in" rarely made to cell entries.R e a d - i n heads
should normally appear as spanners reading in to lower spanners or to
column captions. "Read ins" are rarely used in the lowest level of box
where the "read in" would be to the individual cell entries. The latter
usage should be confined to reader columns (columns containing words
or phrases in the cells); even then, it is rarely needed.
1245. Should be avoided over total columns. Read-in heads
should not be used over total columns except where the total caption
is phrased for proper inclusion. In example A-2, note "Families
having total families."
Example A - 1 . Right:

Example A - 2 . Wrong:
Families having-

Families havingTotal
families

No
child

2 or
more

1

child

Total
families

No
child

1

child

2 or
more

1246. Exemption of residuals from read-in rule.-Where necessary,
residual entries, such as "Other," or "Remainder," etc., may be
exempted from the rule that all subentries of a read-in caption must
"read in" smoothly. Where practicable, however, the wording of the
read-in spanner or of the residual caption should be adjusted to conform. (See par. 928.)
Example A - 2 . Acceptable:

Example A-1. Preferred:

Total,
owned
homes

200

Occupied b y White

Nonwhite


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Other
races

Total,
owned
homes

Owned homes withWhite
occupants

Nonwhite
occupants

Other
races

sEc.

12-D

THE

DASH

(OR R E A D - I N ) H E A D

PAR.

1247

1247. "Reading in" to other terminal groups.-Terminal groups
other than residuals are not exempt from the read-in rule. However,
note that here, as elsewhere, a listing terminating an obvious series
may reasonably omit the term indicating what is being classified.
Thus, in example A - 1 , paragraph 1245, above, it is sufficient to say
"2 or more" instead of "2 children or more" since the previous column
captions make it clear that "children" is meant.
1248. Incorrect use of the read-in d a s h . - I n the boxhead, as in
the stub, dashes frequently are used unnecessarily. Improper use of
the dash may be confusing.
a. Unnecessary.-In example A, below, t h e dash is useless. True, the spanner
"reads i n , " but this would be t r u e of many spanners. If t h e dash is not essential,
it should be omitted, thereby focusing t h e reader's attention on instances where
real assistance is offered by t h e read-in dash.
b. Confusing.-In example B, t h e dash after the d a t e ( " J u n e 3 0 - " ) is not
merely useless; it falsely implies that the effective year varies from cell to cell. No
such variation appeared in t h e published table from which this example was
extracted. In fact, careful examination of t h e table showed no meaning for t h e
dash. I t s function, in such cases, is to invite t h e reader to search for something
t h a t does not appear.
Example A. Unnecessary:
Power available90 percent
of the
time

50 percent
of the

time

E:rample B. Confusing:
Number of items,
June 30
Type A

Type B

Sec. 12-E. The Spanner Head (1251-1255)

1251. Definition.-The spanner head is a classifying, descriptive,
or qualifying caption spreading across (and above) two or more
individual column heads, or across two or more lower spanners. In
its normal range, it applies, in varying degree, to all columns or
subordinate spanners thus covered. The spanner head represents
the boxhead counterpart of the stub center head or subhead and of
the colon and dash (read-in) line caption. General principles outlined
for such stub devices apply also to the spanner head. (See secs.
9-B and 9-C.)
1252. Purpose and function.-The purpose of the spanner head is
similar to that of stub center heads, etc., but an increased burden is
laid upon it by the demand for extreme brevity of individual column
heads. Some of the more common uses of spanner heads are as
follows:


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

201

PAR.

1252

THHE B O X H E A D

sE0.

12-E

a. To indicate class or kind of information shown in the cells, particularly when
this is not stated in the individual column captions.
Example B:

Example A:
Males

Value of land
1946

1945

Full

Part

owners

owners

b. To describe the classification shown in the next lowest level of box.-Simple
cases are illustrated below. At times, the nature of the caption may be such
t h a t it may be thought of as referring to t h e next lowest level or to the cells,
either or both, without likelihood of confusion. At other times, it may specifically refer to the next lowest level of box. (For cases where confusion might
arise, see par. 1253.)
(1) Reference b o t h to lower box-level and to cells. H e r e t h e spanner
caption is such as to apply equally, at least by implication, to both t h e next
lower level and to the cells. In examples C and D the user will tend to assume
t h a t t h e cell entries represent t h e number of dams, or number of employed
workers, unless some other unit of measure is specified clearly elsewhere.
Example D:

Example C:
Dams
Diversion

Employed workers

Storage

White

Negro

Other

(2) Reference to lower box-level o n l y , not to c e l l s . - H e r e t h e upper
spanner has no direct relationship to the cell entries. It specifically describes
the captions appearing in t h e next lowest level of box. In example E the unit
of measure for t h e cells is not clear and must be specified or made clear elsewhere in the table. T h e cell entries might refer to number of dams, number
of men employed in construction, storage capacity, water flow, cost, etc. In
example F the cell entries might refer to t h e number of whites, Negroes, etc.,
but not necessarily. These combinations of heads are satisfactory, but must
not be relied upon exclusively to supply the unit of measure for the cells.
Example F:

Ezample E:

T y p e of dam

Diversion

Storage

Race

White

Negro

Other

c. To carry descriptive words common to all subordinated boxes on next lowest
level.-This is one of t h e most common uses of t h e spanner. In general, where
two or more adjoining box captions begin or end with the same words, or involve
t h e same concept, the common phrase or concept may be transposed to a spanner.
(See par. 1224a.) Frequently, this releases space pressure in individual column

202

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEc.

12-E

THE SPANNER HEAD

PAR.

1252

captions. It may, or may not, decrease the over-all depth of the boxhead. Where
necessary, such a head may be made to "read in" by terminating it with a dash
( - ) , but this should be avoided where possible. (See par. 1248a.)
Example F-1. Without spanner:

Municipally
owned
electric
utilities

Example F-2. With spanner:

Electric utilities

Privately
owned
electric
utilities

Municipally
owned

Privately
owned

Example G-1. Without spanner:

Example G-2. With spanner:

Wage rates
Wage rates
per month,
per month,
with board without board

Wage rates per month

Example H-1. Without spanner:

Price for
1939 crop

Price for
1940 crop

With board

Without board

Example H-2. With read-in spanner:

Price for
1941 crop

Price for crop of1939

1940

1941

Note: Spanners should cover only distinctly related columns. T h e mere fact
t h a t two adjoining columns start or end with t h e same word or phrase does not,
in itself, justify use of a spanner since they may be totally unrelated in terms of
the presentation. T h e columns must be both adjoining and related.
d. To describe specific measure employed.-This is particularly useful when
different panels employ different measures. Two general situations arise:
(1) Specification for all p a n e l s . - H e r e , t h e nature of the d a t a are such
that it is desirable to specify the measurement unit in spanners throughout the
entire boxhead.
Example J:

Example K:

Number
1945

1946

Percent
1945

1946

Thousands of
pounds
1945

1946

Dollars
1945

1946

(2) Specification for selected p a n e l s . - F r e q u e n t l y , t h e statistics are such
t h a t t h e title of the table, or other presentation feature, makes the unit of
measurement obvious. In such cases it is not necessary to specify the unit of
measure further, except for such panels as may depart from this normal expectation.
T h e most frequent case is represented by "Number." Where the measurement unit is unitary, the word "Number" usually may be omitted from the
boxhead without loss of clarity. At times, advantage may be taken of this.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

203

PAR.

1252

sEc. 12-E

THE BOXHEAD

T h a t is, where a table is divided into two panels, "Number" and "Percent,"
the "Number" section may be used for more general description instead of
merely to specify "Number."
Example L-1. Right:
Number

Total
Employed

Seeking

work

Percent of total
Other

Employed

Seeking
work

Other

Example L-2. Preferred:
Percent of total

In labor force
Total
Employed

Seeking
work

Other

Employed

Seeking
work

Other

1253. Clarifying point of reference.-Spanners may refer to subspanners, to column captions, or to the cell entries themselves.
Spanners must be written in terms of the framework in which they
are to appear, not as something in themselves. Otherwise, the point of
reference, though it may be clear to the designer, may not be made
clear to the user. The designer, confronted by a specific problem of
description, must write an appropriate spanner. The user must reverse this process; that is, upon reading the spanner, he must note
its applicability. If the user cannot reconstruct readily what the
designer had in mind, the spanner is defective.
a. Identification of major class or statement of universe.-Where t h e spanner
states the universe involved, or states the major class being distributed, no difficulty is likely to arise. Usually, the applicability of such spanners is obvious.
Example A:

Example B:

Urban population
White

Male

Fe
male

Contractors

Nonwhite

Male

Fe-

male

Resident
Number of
establishments

Value of
work
performed

Nonresident
Number of
establishments

Value of
work
performed

b. Description of subclassifications.-These spanners should be thought of as
applying solely and strictly to the next lower level of box unless they specify
otherwise. It follows t h a t they should always appear immediately above the
captions (whether column captions or subspanners) to which they refer.

204

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEc. 12-E

PAR. 1253

THE SPANNER HEAD
Example C-2. Wrong:

Example C-1. Right:

Color

Sex

Total

Female

Male
White

Nonwhite

Nonwhite

White

Male

Total

White

Example D-1. Right:

Female

Nonwhite

Nonwhite

Example D-2. Wrong:
Sex

Color
Total

White

White

White

Nonwhite

Nonwhite

Total
Male

Fe
male

Female

Male

Male

Fe
male

Male

Female

(1) Description of numerical intervals.-Spanners describing classifications involving numerical intervals require special care in wording and use.
Otherwise, since both the subcaptions and the cell entries will be composed of
numbers, confusion may arise concerning the point of reference for the spanner.
Example E-1. Confusing:
(Is reference to cell entries or to intervals?)

Example E-2. Clear:
(Number reference Is obviously to intervals.
Cell unit also is clear)

Number of children
Total
None

1

Women having specified number
of children

3 or
more

2

Total
women
None

1

2

3 or
more

Example F-1. Confusing:
(Do the cell entries represent the number of persons
in 2-person families, etc., or the number of families
with 2 persons, etc.?)
Number of persons In family
Total
1

2

3

5 or
more

Example F-2. Clear:
(Number reference Is obviously to Intervals. Cell
unit also Is clear)
Families comprising specified number
of persons
families


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

1

2

3

5 or
more

205

PAR.

1253

sm0. 12-E

THE BOXHHAD

c. Unit of measure.-Here too, confusion is possible unless t h e spanner is
handled in terms of its surroundings. T h e same general problem arises as in
par. 1253a, above.
In example G-2, below, does the "Percent" refer to t h e intervals shown in the
subordinate caption, or does it mean t h a t the cell entries are 'percentages? In
this specific case, the reader can determine the answer by examining the cell
entries. However, in more complex materials, the answer may not be apparent.
Where possible, the spanner and column head treatment should be such as to
avert uncertainty.
Example G-2. Confusing:

Example G-1. Clear:
Percent
Under 45
years

Percent
Under 45

45 and
over

45 and over

Example H-1. Clear:

Example H-2. Confusing:

Percent by age

Percent by age (years)

Under 45
years

45 and
over

Under 45

45 and over

d. Use of " b y " and "of" in s p a n n e r s . - I n many cases, t h e use of " b y , "
o f " or a similar term, in the spanner is likely to solve the general problem and
add considerably to clarity.
T h e typical case is that given below where the use of either portion of the
spanner standing alone would be confusing. T h e obvious need is to specify both
t h e significance of the classification shown in the column heads and the unit of
measure employed in the cells.
Example J-1. Confusing:

.

Example J-2. Confusing:

(What Is the column-bead
classification?)

(What Is unit in cells?)
Value of all farm products

Number of farms

$1,000
$1,000

to

$1,499

$1,500

to

$1,999

$2,000

and
over

Example J-3. Clear:
Number of farms by value of
all farm products

206

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

$1,000

$1,500

to

to

$1,499

$1,999

$2,000

and
over

$1,500

$2,000

to

to

$1,499

$1,999

and
over

sEc. 12-E

THE SPANNER HEAD

PAR.

1253

1254. Spanning total columns: General.-The inclusion of a total
column under a spanner depends, in part, upon the spanner wording
and, in part, upon the relationship of the given panel to other panels
in the table. The possibilities involved are too numerous for complete
coverage here. A few common problems are discussed briefly.
a. Statement of universe or presentation u n i t . - I n general, spanners comprising a statement of the presentation unit (or of the universe involved) should cover
the total column. Otherwise the reader may find t h a t the components have been
adequately described but t h a t the total has not been. In situations exemplified
by example B-2, a minimum requirement would be to change the total caption to
read " T o t a l exports."
Example A-1. Right:

Example A-2. Confusing:
Number

Number
Total
Total

Female

Male

Example B-1. Right:

Male

Example B-2. Confusing:
Exports

Exports
Total

Female

U. S. mer- Reexports
chandise

Total

U. S. merchandise

Reexports

b. Description of subclassification.-At times, a spanner is used to describe the
distribution which appears in the columns to the right of a total column, such as
age or sex. Normally, this type of spanner should not cover the total column.
(1) M i n o r a p p l i c a t i o n . - I n some instances this rule may seem academic.
Thus, no reader is likely to be confused if the spanner "Sex" covers the total
column, even though such treatment is theoretically incorrect. (The example
is over-simplified. In practice, the spanner is unnecessary for the subject
matter used.)
Example C-2. Poor style:

Example C-1. Right:
Sex

Sex

Total
Male

Female

Total

Male

Female

(2) M a j o r application.-At times, serious misapprehension may result
from violation of this rule. The type of error shown in example D - 2 , below,
may well pass unnoticed by the table-designer with whom the correct style has
not become an established habit.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

207

PAR.

1254

src. 12-E

THE BOXHEAD

Here, t h e inclusion of t h e total column under t h e spanner results in a description of t h e column as showing " T o t a l value," a term normally used to signify
t h e aggregate value in dollars. T h e case is thereby misstated since t h e total
column in example D - 2 is intended to show the total number of units which are
being distributed according to value.
E v e n though this total did represent " T o t a l value," this form of presentation
would still be defective since it would leave the way open for misunderstanding
in the opposite direction. Aggregates of value, amounts, etc., should always be
signalled clearly. Conversely, caption treatment of totals of units being distributed should avoid a n y possibility of misapprehension. Examples D - 1 and
D - 2 would be clarified somewhat if the total captions were expanded to read
" T o t a l units," as in examples E - 1 and E - 2 .
Example D-2. Confusing:

Example D-1. Right:
Value
Total

Under
$1,000

Value

$1,000

Total

and over

By value

Under
$1,000

$1,000

$1,000

and over

Example E-2. Better than
D - 2 but still undesirable:

Example E-1. Preferred:

Total
units

Under

Total units by value
Total
units

$1,000

and over

Under
$1,000

$1,000

and over

c. "Double-duty" t o t a l s . - A t times two or more distributions may appear in
t h e boxhead, and each may add independently to an identical total. Usually,
repetition of the total column is undesirable in such cases.
Normal solution: Cover each distribution'with its own spanner. Then separate
t h e panels by a medium-weight (letterpress, No. 3) rule. (See sec. 1 3 - C for
discussion of vertical rules.) Place the total column at the left. Do not include
it under the first spanner; do not repeat it for the second panel. (In the case of
sex and color, it is obvious t h a t the same total will serve for both. In t h e more
complex classifications, the relationship may not be so clear.)
Example F-1. Right:
("Double-duty" total; serves both panels)
By sex

By color

Total
Male

Female

208

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Ezample F-2. Confusing:
(Where is "Total" for color distribution?)

White

Nonwhite

Sex
Total

Male

By color
Female

White

Nonwhite

sEc. 12-E

THE

SPANNER HEAD

PAR.

1255

1255. Effect of spanner omissions.-At times, space or other considerations may make it necessary or desirable to omit an upper
spanner ordinarily deemed necessary. They may even make it
necessary to delete all spanners. In such cases, special precautions
are desirable. In particular, t h e captions of the total columns and
t h e types of vertical rules employed should be reviewed and necessary
adjustments made. Two typical cases are described below.
a. Omission of upper spanner only, leaving lower spanners.-Where t h e box
contains several distributions, each with its own total, both major and minor
spanners may appear; such as universe spanners establishing major panels, and
class-description subspanners. Removal of the upper spanner in such cases
requires adjustments in the material remaining.
Example A-1. Complete form:
Workers reporting marital status

All workers

Marital status

Sex
Total

Total
Male

Single

Female

Married

Other

(1) Spanner removal necessitates changes.-With the upper spanner removed
(example A-2, below), note t h a t (a) the total columns have no indication of the
class referred to; and (b) the presence of the two parallel total rules rising to
t h e same level is objectionable. Since no heavy rule or medium rule intervenes,
t h e range of influence of t h e left-hand parallel rule includes the second total
column. T h i s picture is confusing.
Example A-2. Wrong (Result of removing spanners without adjusting wording or vertical rules*
Class descriptions for total columns are missing, and the two parallel vertical rules rising to the same
level are confusing):
Marital status

Sex
Total

Total
Male

Single

Female

Married

Other

(2) Suggested t r e a t m e n t . - T h r e e suggestions are illustrated below. In al]
three, the classes referred to by the totals are made clear, and the range of
influence of the left-hand parallel rule is stopped after the "Female" column,
as it should be.
Example A-3 is preferable since it reestablishes the basic panel arrangement,
even though neither spanner properly belongs over the total column. (See
example C-2, par. 1254b.)
Example A-3. First choice:
Sex
All
workers

Male

Marital status
Female


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Workers
reporting

Single

Married

Other

209

PAR.

1255

SEC.12-E

THE BOXHEAD

In example A-4, the right-hand total caption is cumbersome.
Example A-4. Second choice:

All
workers

Sex
Male

Female

Workers
reporting
marital
status

Marital status
Single

Married

Other

In example A-5 (not recommended), t h e parallel rule used to separate t h e
table sections is fairly effective but this usage is undesirable in itself. (See
par. 1330b.) Also, it forces t h e designer to use medium weight rules for total
rules. (See par. 1328d.)
Example A-6. Not recommended:

All
workers

Sex

Male

Female

Workers
reporting
marital
status

Marital status
Single

Married

Other

b. Omission of all spanners.-In a badly squeezed table, all spanners, even
those normally essential, may have to be omitted. This is permissible only as a.
last resort.
Example B-1. Intermediate stage (same as example A-3; for full form, see Example A-1):
Bex

workers

Male

Marital status
Female

Workers
reporting

Single

Married

Other

(1) Spanner removal necessitates' c h a n g e s . - I n example B - 2 , below,
note t h a t the class description for the right-hand total column is missing and
t h a t the box is no longer visibly divided into two independent sections.
Example B-2. Confusing:
All
workers

Male

Female

Workers
reporting

Single

Married

Other

(2) Suggested t r e a t m e n t . - T h e following suggestion is similar to t h a t
proposed for cases in which only t h e upper spanner is removed.
In example B-3, t h e right-hand total caption has been clarified and t h e box
sectionalized with a medium weight rule.
Example B-3. First choice:

All
workers

Male

210

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Female

Workers
reporting
marital
status

Single

Married

Other

sEc. 12-E

THE SPANNER HEAD

PAR.

1255

In example B - 4 (not recommended), the rule usage has been reversed.
A parallel is used as t h e dividing rule; medium weight rules are used for total
rules. (See pars. 1328d and 1330b.)
Example B-4. Not recommended (see example A-5, above):

workers

Male

Female

Workers
reporting
marital

Single

Married

Other

status

Sec. 12-F. The Banner Head (1261-1266)

1261. Definition. The banner head is a special type of spanner
extending across and covering the entire boxhead of the table (except
the stub box). Normally, it is undesirable. Since it applies to the
table as a whole, it represents an item better covered in the title or
headnote, or better presented as a main centered head in the stub.
Sound usage tends to restrict it sharply to the circumstances indicated
below. A good rule is: When in doubt, avoid the banner head.
1262. Acceptable u s e . - I n general, banner heads are best reserved
for two principal uses:
a. To emphasize an essential qualification.-At times, it may be essential to
warn t h e reader of a highly important qualification which must be kept constantly
in mind if the data are not to be misinterpreted. Ordinarily, this can be taken
care of in t h e title, headnote, or in t h e stub. On rare occasions, however, a
banner head is justified.
b. To emphasize a major distinction between standard t a b l e s . - A common
objective in Bureau of the Census tables is to provide strictly comparable coverage for different universes or areas at different points in time. T h e customary
method is to employ a series of standard tables. Each is identical in form but
t h e statistics shown relate, in each case, to a different universe or time period.
(1) Universe d i s t i n c t i o n . - W h e r e each table provides a distribution for a
different major class or universe, t h e distinctive feature should be emphasized
by placing t h a t feature a t , or near, t h e beginning of t h e table title. Here,
banner heads are undesirable.
(2) Area a n d t i m e d i s t i n c t i o n s . - W h e r e area and time differences are the
distinctive features, title-distinction might rest on starting t h e title with a key
phrase (THE NoRTH, 1 9 4 0 - ) , but this use of t h e key phrase technique is not
desirable. Otherwise, t h e reader must note the difference in t h e area and time
segments at t h e title's end.
Where t h e titles are brief, end-differences will stand out fairly well. Where
t h e titles are long, end-differences tend to be overlooked. In t h e latter case,
t h e use of t h e banner head may be justified.
T h e following example assumes a series of six tables, identical in stub and
box, with long descriptive titles. T h e titles are assumed to be identical with
t h e exception of t h e area and time segments at t h e end.

776554-49-1


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

211

PAR.

1262

sEc. 12-F'

THE BOXHEAD

Example A. Banners are useful if the preceding portion of title is long and involved:
TABLE

1.-

F O R THE N O R T H :

1940

The North, 1940
Color and marital status

Under
5
years

All

ages

TABLE

2.

•

5 to
14
years

15 to
24
years

25 to
34
years

FOR T H E N O R T H :

35 to
44
years

45 to
54
years

55 and
over

45 to

55 and
over

1930

The North, 1930
Color and marital status

Under

All

5

ages

years

5 to
14
years

15 to
24

years

25 to
34
years

35 to
44
years

years

(Tables 3 to 6 continue in the same vein and provide statistics for the South,
1940; t h e South, 1930; the West, 1940; and the West, 1930.)

1263. Improper use of banner heads. Banner heads should not
be used casually. They should be reserved for emergencies. In
particular, their use should be avoided to specify something already
stressed in the table title or which can as well be placed in a headnote
or in a stub center head. Examples of these uses follow:
a. To specify a universe clearly emphasized in t h e table title.
Example A. Banner is useless:
T A B L E 7 . - E M P L O Y E D W O R K E R S BY A G E A N D R A C E :

1940

Employed workers
Age
All

classes

White

Negro

Other
races

b. To specify a general qualification of t h e table better placed in a headnote.
Example B-1. Banner unnecessary; place in headnote:
T A B L E 9 . - N A T I O N A L INCOME BY INDUSTRIAL D I V I S I O N S :

1937

TO

1941

Millions of dollars
Industrial division
1937

212

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

1938

1939

1940

1941

SEC.12-F

THE B A N N E R HEAD

Example B-2.

1263

PAR.

Improved:

T A B L E 9 . - N A T I O N A L INCOME BY INDUSTRIAL D I V I S I O N S :

1937

TO

1941

[Millions of dollars]
Industrial division

1937

1938

1939

1940

1941

c. To specify an area, time period, or general class where the statement can as
effectively be placed as a center head at the top of the stub (not in stub box).
Example C-1. Is banner essential?
All experienced workers
Area and age

All
classes

Native
white

Foreignborn
white

Negro

Other

The State__
Under 45 years __
45 and over_
Urban

Under 45 years __
45 and over_

Example C-g.

Will stub center head serve the purpose?

Area and age

All

classes

Native
white

Foreignborn
white

Negro

Other

ALL EXPERIENCED

WORKERS
The State._
Under 45 years __
45 and over..

1264. Quasi-banner heads: General.-This is a spanner deceptively similar to the banner head in physical appearance since the only
difference is that the quasi-banner does not cover the total column.
A quasi-banner

A banner bead

Stub box

Stub box

Total

Total

This similarity in physical appearance promotes overuse and abuse
of the quasi-banner, just as in the case of the banner head. That is,
the quasi-banner also tends to be used for general descriptive or
qualifying statements better expressed elsewhere.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

213

PAR. 1264

THE

sEc.

BOXHEAD

12-F

However, the quasi-banner, used properly, performs one or both of
two extremely important functions. Specifically, it may clarify an
otherwise unintelligible series of column heads, or it may facilitate
reference to the column heads by eliminating repetitive words or
phrases.
1265. Quasi-banner heads: Classification description.-For purposes of this discussion, classifications may be divided into two main
groups: Those normally self-identifying, and those normally requiring
explanation or description. Each group is discussed briefly below in
terms of use of the quasi-banner.
a. Classification normally self-identifying.-Many classifications are so commonly used that identification is unnecessary, although it may be essential to indicate the applicable group.
Example: To reduce the case to its simplest form, a table with t h e three
columns "Total," "Male," and "Female," does not need "Sex" inserted as a
quasi-banner to identify t h e sex classification as such. Further, in a series of
tables devoted entirely to t h e characteristics of t h e total population, it would
scarcely be necessary to specify "Sex of person." In such cases, the quasibanner is rarely justified.
However, if t h e table deals with characteristics of individuals in terms of
their family connections, or with persons in family groups as such, it might be
imperative to make clear t h a t "Sex of head" is meant, not "Sex of child" or
"Sex of principal wage earner," etc. Here, t h e quasi-banner performs a valuable function.
b. Classification normally requiring identification or description.-Most classifications involving numerical intervals a r e of this type. Thus, t h e expression
" U n d e r 5" means nothing in itself. It might mean years of age, years of schooling,
thousands of acres, thousands of proof gallons, millions of dollars, etc. In general,
where numerical intervals appear in the boxhead, identifying and explanatory
spanners are essential. Where the box is devoted to a single classification adding
to a total column at t h e left, t h e use of t h e quasi-banner may be desirable. However, such spanners should be clearly descriptive of the column-head classification,
as such; if descriptive of cell entries, the total column should be covered also. (See
par. 1253.)

1266. Quasi-banner heads: Reduction of wordy column heads.Frequently, wordy column heads may be avoided by use of descriptive
spanners. The saving is likely to be particularly great where identification or descriptive problems demand that a given word or phrase
be repeated in each individual column head. (See par. 1252c.) Here,
the quasi-banner is useful. The following example assumes that the
stub also is in years, such as years of school completed. This makes
specific reference to age important in the boxhead.
Example A-1. Column heads meaningless:

Total

Under
15

15 to

214

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

25 to

35 to

45 to

55 to

65 to
74

75 to
84

85 and
over

sEc. 12-F

THE BANNER HEAD

PAR.

1266

Example A-2. Column heads wordy:

Total

Under
15
years
old

15 to
24

years
old

25 to
34
years
old

35 to
44
years
old

45 to

55 to

years
old

years
old

65 to
74
years
old

75 to
84
years
old

85 and
over

65 to
74

75 to

85 and
over

Example A-3. Quasi-banner clarifies the picture:
Age (years)
Total

Under
15

15 to

25 to

34

24

35 to
44

45 to

55 to

84

Sec. 12-G. The Undercut Spanner (1271-1272)
1271. Definition.-The undercut spanner is a spanner appearing on
the lowest level of box directly over, and spanning, two or more
columns. Since it appears below the individual column heads, it
breaks the column rules and separates the column captions from the
data which they describe. For this reason, it is generally undesirable.
& a m p l e A-1. With undercut spanner:
Receipts
Trailers,
etc.

Motorcycles,
etc.

Total
receipts

Registration

Thousands of dollars

Example A-2. Undercut avoided:

Trailers,
etc.

Motorcycles,
etc.

Receipts (thousands
of dollars)
Total
receipts

Registration

1272. Conditions of u s e . - T h e following brief discussion indicates
instances where use of the undercut spanner is justified and where it
is not justified. In general, its use should be confined to indication of
presentation unit or unit of reference.
a. Justified u s e . - T h e undercut spanner is justified primarily where all of t h e
following conditions are present:
(1) The presentation unit for the given group of adjoining columns differs
from t h a t used in the remainder of the table.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

215

PAR. 1272

sEc. 12-G

THE BOXHEAD

(2) It cannot be included parenthetically within each of the individual
column heads concerned.
(3) There is no way to place it in an upper spanner in parentheses after the
spanner statement, or as a spanner head in itself.
( 4) It is too long to appear in italic (letterpress), or in parentheses (if
typewriter), in each individual column just below the boxhead a n d above t h e
first column entry.
(5) Specification in a headnote, as "Amounts are in thousands of dollars,"
is impracticable or undesirable.

b. Not j u s t i f i e d . - I t s use is not justified(1) Where the unit-indicator is brief and can readily be repeated outside t h e
box a n d placed within the field at the head of each column where it belongs.
(2) Where other of t h e above-mentioned means of presentation can be
made to serve the purpose.
Example A-1. A void if possible:

Census year

Number of
estab
lishments

Wage
earners
(average
for the
year)

Wages

Value of
products

Value
added by
manufacture

In thousands of dollars

Example A-2. Preferable:
Number
Census year

216

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Establishments

Wage
earners
(average
for the
year)

Thousands of dollars

Wages

Value of
products

Value
added by
manufacture

Chapter 13
HORIZONTAL AND VERTICAL RULES (1301-1334)
Sec. 13-A. General (1301-1302)
1301. Definitions. The terms "horizontal rule" and "vertical
rule" as used here cover all types of continuous horizontal and vertical
lines appearing as part of the table.
1302. Terminology : "Dashes" versus "rules."-In typographical
language a rule is sometimes referred to as "dashes." Thus, in the
G. P. 0. Style Manual (revised edition, January 1945, p. 127) the
following statement appears: "152. Rules may be used in place of
dashes in a crowded table."
Throughout this manual, the term "rule," as in "horizontal rule"
or "vertical rule," covers both "dashes" and "rules," as used above.
The difference is a mechanical matter that does not directly affect the
table-designer. The following statement is for information purposes
only and does not affect the ensuing discussion.
a. " D a s h e s . " - A close scrutiny of tabular rules in letterpressed (type-set)
publications of t h e Bureau of t h e Census will reveal t h a t the horizontal rules are
not continuous, but are made up of tiny, closely-joined segments. These small
individual segments, set by machine, are the "dashes" referred to in the G. P. 0.
Style Manual. " R u l e s " are frequently made up in this way since the dashes do not
require hand-insertion, and the deletion or repair of a n y part of such a "rule" can
be made without handling the entire line.
b. " R u l e s . " - S t r i c t l y speaking, a "rule" is a long strip of metal with a polished
printing surface which must be cut to t h e proper length and inserted in t h e t y p e
page by hand. For technical reasons, this t r u e type of "rule" can be cast to occupy
less space t h a n t h e "dashes," hence it is used where t h e table is crowded.
c. " D a s h e s * versus "leaders.
Dashes" should not be confused with
"leaders." T h e leader line is one in which t h e tiny dashes do not run together, but
are separated by white space. Therefore, t h e leader line, when printed, is a broken
line. In contrast, printed "dashes" form a continuous line looking like a rule.
T h e comparative effect is t h e same as if, on t h e typewriter, one were to use a row
of hyphens (lined up at the bottom of t h e line) for leaders (not recommended), a n d
the conventional underscore for "dashes" or rules. The "dash segments," shown
below have had spaces inserted between each to illustrate their structure.
Leaders_--

Dash segments opened up

Rule strip

Dashes closed to make rule


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

217

PAR.

1311

sEc. 13-B

HORIZONTAL AND VERTICAL RULES

Sec. 13-B. The Horizontal Rule (1311-1316)

1311. General.-The term "horizontal rule" as used here covers
all types of continuous horizontal lines appearing as part of the table.
1312. Types of horizontal r u l e s . - T w o types of horizontal rules are
used in Census tables: The single rule and the parallel (double) rule.
As the name implies, the single rule is composed of a single line.
The parallel (double) rule is composed of two parallel lines run closely
together, each of the same weight or heaviness as the single rule.
With the exceptions noted below, the table-designer need have no
concern with mechanical differences between rule practices in letterpress (type-set) composition and in typewriter-offset.
a. Total rule versus boldface.-Letterpress

and

typewriter-offset

differ

radically at this point.
(1) L e t t e r p r e s s . - G r o u p totals and grand totals are set in boldface type.
T h e single rule is used sparingly; t h e parallel rule is reserved for grand totals,
where used at all.
Example A - 1 :
All classes__

White_Negro_.Other races_

75

10

20

25

20

50

5

20

10

17
8

1

1

15
3
2

9
1

20
15
3
2

25
20
1

33
20

45
39

Example A-2:
Adams C o u n t y . _

White_
Negro
Other races_
White_

Brown County_

75
50

5

1

100

11

15

8
2

10

10

1

3

75

Negro_.Other r a c e s -

10

17
8

20
10
9

1

1

11
8

2
1

(2) Typewriter o f f s e t . - S i n c e boldface is not available on t h e standard
typewriter, t h e single rule is. used to set off group totals. Avoid hand-drafted
horizontal parallel rules in offset work; use t h e single rule instead and rely on
indent and spacing to carry the story. In all but rare instances t h e use of the
horizontal parallel rule is not essential.
Example B:
All c l a s s e s .

White . . . . . . . . . . . . . . . . . . . . . . .

Negro . . . . . . . . . . . . . . . . . . . . . . .
Other r a c e s . . . . . . . . . . . . . . . . .

75

10

20

25

20

50
17

5
4

15

20

10

2

4

1

8

1

1

9

b. Offset space requirements.-The insertion of horizontal rules under totals
and subtotals necessarily increases the depth of the page. Also, it may make it
necessary to increase the space above t h e s t u b entry of the line t h a t is underscored. Otherwise a subtotal may appear to be a subentry of the preceding group,
instead of a total for t h e following group. T h e following details are provided
for informational purposes. T h e reference to half-space refers to typewriters
equipped for half-rachet turns, an extremely useful device in statistical typing.

218

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEc. 13-B

THE

HORIZONTAL RULE

PAR.

1312

(1) Single r u l e . - N o more t h a n one-half typewriter line of space is needed
for t h e rule. However, one full space (line) should be left open above t h e subtotal or total line. If one line is left open below for t h e single total-rule, then
one a n d one-half lines, preferably two lines, should be left free above the total
line. Furthermore, t h e underscore rule should be placed up under the entries
underscored; it should not be placed midway in the space left for it. (See
par. 1313e.)
Example C-1. Desirable spacing:
Other races . . . . . .

2

All c l a s s e s . . . .
White . . . . . . . . . . . .
Negro . . . . . . . . . . . .
Other races . . . . . .

75
50
20

Example C-2. Acceptable spacOther races . . . . . .

2

classes . . . .

75

Other races . . . . . .
All c l a s s e s . . . .

2
75

White . . . . . . . . . . . .
Negro . . . . . . . . . . . .
Other races . . . . . .

50
20
5

White . . . . . . . . . . . .
Negro . . . . . . . . . . . .
Other races . . . . . .

50
20
5

All

5

Example C-3. Bad spacing:

ing:

(2) Parallel r u l e . - A s indicated above, avoid horizontal parallel rules in
off set work. Where used, one full typewriter line of space should be allotted.
This means at least one and one-half, preferably two, lines should be left blank
above the total line.
Example D-1. Desirable spacing:
Other races . . . . . .
All c l a s s e s . . . .
White . . . . . . . . . . . .
Negro . . . . . . . . . . . .
Other races . . . . . .

Example D-2. Acceptable spacing:

Other r a c e s . . . . . .

2

All

classes . . . .

75

White . . . . . . . . . . . .
Negro . . . . . . . . . . . .
Other races . . . . . .

50

75
50
20
5

2

20
5

Example D-3. Incorrect spacing:

Other races . . . . . .
All

2

classes . . . .

75

White . . . . . . . . . . . .
Negro . . . . . . . . . . . .
Other races . . . . . .

50
20
5

1313. The single rule: Function and purpose. The single horizontal rule is used for four distinctive purposes in letterpress. In
offset, it serves the same four purposes plus one additional purpose.
These purposes are listed and illustrated below with a statement, in
each case, of the horizontal extension limits of the rules. Since
various rules are shown in the illustrations, an arrow () points
to the rule under discussion in each example.
a. Bottom rule of boxhead.-Runs full width of table.
remainder of table.

Sets off boxhead from

Example A:
All classes

White

Cities
Male

Female

Male

Female

Anderson
Ashton__

(1) W i t h tracer n u m b e r s . - E x t e n s i o n of rule to cover tracer columns
depends on presence or absence of entry in column box. If the tracer-column
box is left blank, the bottom rule of boxhead does not cover the column. If


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

219

PAR.

1313

sEc. 13-B

HORIZONTAL A N D VERTICAL RULES

an e ntry ("Line No.") appears, extend the bottom rule of the boxhead clear
across. Examples below show a left-hand tracer column only; identical treatment is applicable for the right-hand tracer column.
Example B-1. No entry in tracer box:

Line
No.

Oitles
1
2

Example B-2. Entry in tracer box:

Anderson_Ashton__-

1
2

Cities
Anderson_Ashton . . .

b. End or bottom rule of table. Runs full width of table. It designates the
end of the table, reading downward; that is, the end of the stub, not the end of
t h e box. It is not used to indicate the bottom of the page, as such. (For detailed discussion of the proper use of this rule, see par. 1316, below.) Indention
of t h e final s t u b caption, and the presence or absence of tracer columns, has no
effect on extension of the bottom rule to full table width.
Example C-2:

Example C-1:
All classes_--

80
50
25
5

White_-Negro_-Other races_--

56
57
58
59

65
40
23
2

Pacific_-Washington_Oregon.
California_-

80
50
25
5

65
40
23
2

c. Terminal rule for additive s e c t i o n s . - R u n s width of field only; does not run
through stub. Its use is optional. (See par. 1314b, below, for examples a n d for
discussion of use of parallel rule to set off additive grand total blocks.
d. Upper and lower rules of field spanners.-Runs width of field only; does not
r u n through the stub since s t u b box is not repeated. Field spanners are a part of
the field, but their presence necessarily affects s t u b spacing.
Example D:
Other races. . .

9

2

3

New York City Bronx Borough
All classes__

White-Negro__Other races__-

80
50
25

5

65
40
23
2

32
20
10
2

21
15
5
1

e. Offset only: Total and subtotal r u l e . - R u n s width of field only; does not run
through the stub. It is used to distinguish additive totals in offset as a substitute
for boldface in letterpress (type set). (In letterpress, when boldface is used for
additive total captions and their field entries, no total rule is ordinarily used except
for grand totals or aggregates which represent summation of major group totals.
These sometimes require use of both boldface and a total rule to set them off
sufficiently.)

220

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEc. 13-B

PAR.

THE HORIZONTAL RULE

1313

Example E-1. Subentries add to total:

* *

.

7

4

All c l a s s e s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . •

100

36

19

75

25

10

Other races

White
Negro
Other races

··

.
.

20

9

5

2

7
2

Example E-2. Right-subentries add except for missing component explained in footnote:
All ages

1. • • • • • • • • • • • • • • • • • • • • • • • • • • • . • • • • • • •

100

28

25

Under 45 y e a r s
.
45 y e a r s and o v e r . . . . . . . . . • . . . . . . . . . . . . . . . . . . • . . . . . .

75
20

20
5

15

8

1 I n c l u d e s age n o t r e p o r t e d , n o t shown s e p a r a t e l y .
Example E-S. Wrong (subentries do not add to total):
F i l l m o r e County, all a g e s

100

28

25

2l y e a r s and o v e r . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . • 55

14

7

Note: T h e offset total rule should not be used excessively.

Where the stub is
composed of many 2-, 3-, and 4-line classifications, the insertion of many total
rules may detract from t h e clarity of t h e presentation, rather t h a n add to it.
Here again, common sense must be enlisted. A table that is a maze of horizontal
rules rarely reflects sound presentation. The same rule applies to excessive use of
boldface lines in letterpress, particularly to the combined use of boldface and t h e
total rule.

1314. The parallel rule: Function and purpose.-The parallel
(double) rule may be used for four purposes, both in letterpress and
typewriter-offset. However, its use in offset should be avoided where
possible because of the tendency to allow excessive space between
component lines of the rule.
Where used in offset, the rules should be drafted with fine lines and
the space between held to about twice the width of the line. A full
typewriter, or half typewriter, space between the component rules is
too much, even allowing for the tendency to fill when reduced. The
parallel horizontal rule is not intended to frame the space left for i t ;
neither does it represent two single rules in succession. It is a double
rule, intended as a unit, and should be drafted accordingly. Where
used to set off important total entries (and space permits), it should be
placed up under those entries; that is, closer to the entries to which it
relates than to the line below to which it does not relate.
a. Top rule of boxhead.-Runs full width of table, including tracer columns.

It is used at the top of the boxhead, whether the boxhead is at the top of the tablepage, or is the boxhead of a second or following deck. In the examples, an
arrow () points to the rule being discussed.
(1) Divide t a b l e s . - I n a "divide" table, t h e entire boxhead is repeated at
intervals down the page, once for e a c h deck. Here the parallel rule is used
above each such boxhead. (The " t o t a l rules" have been omitted as not pertinent
to the example.)


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

221

Pax. 1314

sE0. 13-B

HORIZONTAL AND VERTICAL RULES

Example A. Right (divide table, entire boxhead repeated) :
T A B L E 8 . - F O R E I G N - B O R N W H I T E , BY C O U N T R Y
OF B I R T H _ - -

T h e State

Urban

Country of birth
Total

Male

Female

Total

Male

Female

All countries_-

England-Scotland __
Azores_All other..
Rural-nonfarm

Rural-farm

Country of birth
Total

Male

Female Total

Male

Female

All countries_

England . . .
Scotland.Azores.
All others_-

(2) Tables with field s p a n n e r s . - S i n c e t h e entire boxhead is not repeated,
neither is t h e stub box. T h e parallel top rule is not repeated and t h e single
rules employed at top and bottom of field spanners are not carried across t h e
stub. (See also par. 1313d, above.) T h e " t o t a l rules" have been omitted as
not pertinent to t h e example.
Example B. Right as shown here; use of parallel above lower field spanner would be wrong (table with

field spanners):

Country of birth

Total

Male

Female

Total

Male

T h e State

Urban

Rural-nonfarm

Rural-farm

All countries._.

England_Scotland ____
All o t h e r -

Al countries.-

England..
Scotland __
All other __

222

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Female

SEC.13-B

PAR.

THE HORIZONTAL RULE

1314

b. Terminal rule for additive total block.-Sometimes used, but not recommended. Extends width of field only. This rule indicates t h e termination of a
total section which represents the summation of corresponding lines in the subordinate sections. (The subordinate sections are each terminated with a single
rule if t h e additive total section carries a parallel terminal rule.)
Example D. Not recommended:

Example C:
United States_-

THE STATE

New England ..
Middle Atlantic.-East North Central_West North C e n t r a L .
South Atlantic.
East South Central__West South Central_Mountain_Pacific_--

All ages..

325

80

40

195
130

25

All ages.

200

55

25

Under 45 years_45 and over_-

120
80

40
15

15

All ages_

125

35

15

Under 45 years_45 and over_-

75

20
15

10

Under 45 years_45 and over_-

25

15

URBAN

New England:
MaineNew Hampshire_-Vermont.
Massachusetts. . .

10

RURAL

50

Note: T h e above usage is not recommended because (1) it rarely signals a fact
not readily ascertainable by a glance at t h e s t u b centered heads or total captions,
a n d (2) it clouds t h e more essential use of horizontal parallel rules to set off one
table section from another, as described in par. 1314c, below.
c. To s e t off one table section from a n o t h e r . - R u n s full width of field only.
This use of the parallel rule is easily abused. In general, it is well not to use the
parallel rule in this way unless one of the listed situations exists. In particular,
omit it if the conditions of use will require its appearance after every block in t h e
stub.
(1) To set off comparatively unrelated s e c t i o n s . - I t is sometimes desirable to include in a s t u b a section which, in terms of classification, is comparatively unrelated to the rest of the stub. Usually such a section is placed at the
end of the table and may be set off by parallel rules.
Example E. Assume in the following that the stub is comprised of three sections of housing data, followed by one section (at the end) of basic population figures. T h e example shows only the last
two sections of the stub.
VALUE OF HOME
All owner-occupied units

Number reporting value ..
Under $300. . .
$300 to $499-$500 to $699.$700 to $999-$1,000 and over_-POPULATION
Ali occupants..

White.
Nonwhite. . .
Under 15 years old.---15 to 44 years old.-45 years old and o v e r Single_.
Married.

widowed and divorced._---


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

223

PAR.

1314

HORIZONTAL AND VERTICAL RULES

sEc. 13-B

(2) To group closely related b l o c k s . - I f , in a given table, one group of
s t u b blocks is more closely related t h a n are the remaining blocks, t h e more
closely related blocks m a y be set off in a special section by means of parallel
rules. This is the obverse of the case cited above. This usage should be restricted closely.
Example F. Stub blocks have been condensed for purposes of the example:
TYPE OF STRUCTURE
All dwelling units..
1-family detached_-I-family attached_Other. . .
YEAR BUILT
All dwelling units..
1935 to 1940-1930 to 1934--

Other..

RACE OF OCCUPANTS
Occupied units ..
White_
Nonwhite_PERSONS PER ROOM
Occupied units_1 person.
2 persons__-

3 or more persons ..

GROSS RENT
Tenant-occupied units_Under $14..$15 to $39.$40 and over_FURNITURE IN RENT
Tenant-occupied units_-Furniture included.
Furniture not included_-

(3) To m a k e clear t h a t successive blocks of d a t a each add to a t o t a l
line shown o n l y o n c e . - W h e r e a given universe is repeatedly classified in
different ways (see each section in above example), it sometimes is undesirable
to repeat t h e identical total line at t h e top of each block. Two solutions are
common:
Where no subtotal rules appear.-If no other total or subtotal rules appear in
t h e stub, t h e insertion of a total rule under t h e entries for t h e several universe
total-lines will sectionalize t h e page sufficiently. (See example G-1.) Colon
lines or centered heads usually appear above each minor grouping. However,
if colon lines or centered heads are used, t h e single presentation of t h e universe
total should appear above t h e colon line or head of t h e first block, not as an
integral part of it. (See example G-2.)

224

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEc. 13-B

THE HORIZONTAL RULE

Example G-1:
All dwelling units_-

By t y p e of structure:
I-family detached. __
I-family attached _
Other __
By year built:
1935 to 1940--1930 to 1934--Other_
AUoccupied units__
By race of occupants:
White_Nonwhite_By persons per room:
1 person __
2 persons __
3 or more persons __
Tenant-occupied units_.

By gross rent:
Under $14--

PAR.

1314

Example G-2:
All dwelling units __
TYPE OF STRUCTURE

I-family detached_1-family attached.Other_
YEAR BUILT

1935 to 1940-1930 to 1934-Other.
All occupied units
RACE OF OCCUPANTS

White_NonwhitePERSONS PER ROOM
1 person-

2 persons_3 or more persons_-

Where other subtotal rules appear.-Here, t h e effect of inserting more total
rules (for t h e universe totals) will be negligible since t h e eye will not select
them readily from t h e other total rules. T h e simplest solution is to introduce
each subblock either with a colon line ("By . . . . · :") or with a centered
head, and sectionalize by use of t h e parallel rule between t h e two sections.
Ezample H-1. With parallel rules:
All dwelling units__
OCCUPANCY STATUS

Owner occupied_-Tenant occupied __
Vacant, for sale or r e n t Vacant, not for sale or rent_STATE OF REPAIR

Needing major repairs _
Not needing major repairs__
Not reporting_
NUMBER OF PERSONS
All occupied units __

1 person2 persons __
3 or more persons ..
Tenant-occupied units __

1 person2 persons_3 or more persons_Vacant units, Cor sale or
rent
ORDINARY OR SEASONAL

Ordinary_.Seasonal__TYPE OF STRUCTURE

1-family detached _
1-family attached_-Other-


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Example H-2. Without parallel rules-undesirable:
All dwelling units_OCCUPANCY STATUS

Owner occupied_
Tenant occupied __
Vacant, for sale or rent_Vacant, not for sale or rent..
STATE OF REPAIR

Needing major repairs_
Not needing major repairs_Not reporting_
NUMBER OF PERSONS
All occupied units __
1 person __
2 persons __
3 or more persons __
Tenant-occupied units_-

1 person2 persons __
3 or more persons_-Vacant units. Cor sale or rent.
ORDINARY OR SEASONAL

Ordinary_Seasonal _
TYPE OF STRUCTURE

1-family detached_I-family attached __
Other __

225

PAR.

1314

HORIZONTAL A N D VERTICAL R U L E S

sEc.

13-B

d. Grand-total or aggregate r u l e . - R u n s width of field only; does not extend
into stub. Usually, this rule distinguishes a grand-total or aggregate line which
represents a summation of additive group-total lines which, in turn, are distinguished by single total-rules.
(1) Use r e s t r i c t e d . - I n general, this rule should be avoided in offset.
It is recommended only where t h e aggregate is of unusual significance, or is not
commonly shown; or where the use of the single total-rule may reasonably result
in misunderstanding of total and subtotal interrelationship.
Example J:
Aggregate__

Continental United States __
Territory in 1790-Louisiana Purchase_-Florida_Territories and possessions _
Alaska_Hawaii.-

1315. End or bottom rule of table: General.-Problems concerning placement of the end or bottom rule of the table are largely confined to situations in which the purpose or significance of the end rule
is misunderstood.
Definition.-The end or bottom rule of the table is a horizontal
single (hairline) rule placed at the bottom end of the table and. extending across the full width of the table-page.
a. T h e end rule is more properly called t h e " b o t t o m rule." It denotes t h e
end of t h e table reading downward, the bottom end; not t h e end of t h e table reading across, t h e right-hand end or edge. T h a t is, it indicates the end of the s t u b
presentation for the categories shown in the boxhead on the given page.
b. Repetition of s t u b caused exclusively by additional boxhead classifications
requires insertion of the end rule upon completion of each such repetition.
c. "Repetition" of stub under different center heads which appear at the top of,
or within, the stub does not require insertion of the end rule at the end of each such
'"repetition," since the repetition is not complete.
d. In divide tables, where t h e stub is so short t h a t the boxhead is continued
in decks on t h e same page, t h e same principle applies. However, the parallel
top rule of the repeated box for the deck below makes the end rule unnecessary
except for the last deck on the page. Of course, if the last deck is incomplete
(carried over to the next page) the end rule is not shown until the end of that deck.

1316. End or bottom rule of table: Specific.-The following
examples will clarify the above general discussion. (They are illustrated in fig. 20.) Note that the distinction between "bottom end"
and "right-hand end or edge" is academic as long as the classifications
shown in the boxhead can be presented in one page or less.
a. Single-page or partial-page t a b l e s . - E n d rule will appear at t h e bottom of the
table page. (See fig. 20, examples A-1 to A-3, inclusive.)

226

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

SEC.

13-B

THE

FIG.

HORIZONTAL RULE

F I G U R E 2 0 . - P L A C E M E N T OF B o T T o M R u L E

20

(See par. 1316)

Group A. Single-page or partial-page tables
Example A - 1 . Narrow table:
Age

Total

Male

Female

All ages__

Under 5ti to 14-15 to 2 4 - 25 to 3 4 - 35 to 44--45 to 54-55 to 6 4 65 to 74--75 and over_-

Example A - 2 , 1 Half-measure table:
Age

Total

Male

All ages_-

Age

Female

Total

Male

Female

35 to 4L
45 to 54-55 to 64___
65 to 7 4 75 and over.

Under 5_____
5 to 14___
15 to 24___
25 to 3 4 -

1 In any fractional-measure table, if the end rule belongs at the bottom of the last partial-measure shown
on the page, it is extended across the entire table.

Example A - 8 .
e

Narrow-divide table:
Total

White

Indian

Chinese

Negro

All ages_.

Under 55 to 14--15 to 2 4 - 25 to 34-35 to 4 4 - 45 to 5 4 - 55 to 64__
65 to 74-75 and over.
Age

Other

All ages__

Under 5-5 to 1 4 15 to 24___
25 to 3 4 - 35 to 44-45 to 54.
55 to 64-65 to 74-75 and over __

776554-49

16


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

227

FIG.

20

sEc.

HORIZONTAL A N D VERTICAL RULES

F I G U R E 2 0 . - P L A C E M E N T OF B o r T o M R u L E

(See

par.

13-B

1316)-Con.

Group B. I-page box, multipage stub 1
Example B - 1 . Continuing stub:
(P. 1 of 2 pp.)
Age

Total

( P . 2 of 2 pp.)
Female

Male

Age

Total

Male

Female

25 to 3 4 35 to 44-45 to 54_55 to 64-65 and o v e r -

All ages_.

Under 55 t o l 4__
15 to 2 4 -

Example B - 2 . Continuing stub with changing center heads:
(P. 1 of 2 pp.)
Age

Total

( P . 2 of 2 pp.)

Male

Female

WHITE

Age

Total

Male

Female

NONWHITE

All agee __

All a g e s _

Under 5-5 to 14-15 to 24-25 and o v e r -

Under 5.5 to 1 4 15 to 2 4 25 and over.-

1 The expression "I-page box" means that after the first page the entire box is repeated without change,
spanner heads and all. The term "multipage stub" means that the stub entries differ from page to page
even though the sole difference lies in variation of a center head at the top of the stub on each page.

Group C. 1-page stub, multipage box 1
Example C-1. Continuing box:
(P. 1 of 2 pp . )
Age

Total

White

( P . 2 of 2 pp.)
Negro

All ages__

Age

Indian Chinese Other

All ages_.

Under 5 __
5 to 1 4 15 to 24-25 and over_.-

Under 5__
5 to 14-15 to 24-25 and over_-

Example C-2. Continuing box with changing banner head:
(P. 1 of 2 pp.)

(P. 2 of 2 pp.)

White

Negro

Age

Age
Total

All a g e s _

Under 55 to 1 4 15 to 2 4 25 and over_-

Male

Female

Total

Male

Female

All ages_.

Under 55 to 14-15 to 24.25 and over_

1 By "I-page stub" is meant that after the first page, the entire stub is repeated without change, center
heads and all. By "multipage box" is meant that the box entries differ from page to page, even though
the sole difference lies in variation of a spanner extending across the entire boxhead on each successive page.

228

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

13-B

SEC.

THE HORIZONTAL RULE

F I G U R E 2 0 . - P L A C E M E N T OF B o T r o M R u L E

FIG.

20

(See par. 1316)- C o n .

Group D. Multipage stub, multipage box 1
Example D-1. Stub completed first for each page of box:
(P. 1 of 4 pp.)
Age

Total

White

(P. 2 of 4 pp.)
Negro

All ages__

Age

White

Negro

35 to 4 4 45 to 54_55 to 6 4 65 to 7 4 75 and over_-

Under 55 to 14.
15 to 2 4 25to34__

(P. 4 of 4 pp.)

(P. 3 of 4 pp.)
Age

Total

Indian

Chinese

Other

All ages __

Age

Indian

Chinese

Other

35 to 44-45 to 54__
55 to 6 4 65 to 7 4 75 and over_-

Under 55 to 1 4 15 to 2 4 25 to 34--

Example D-2. Box completed first for each page of s t u b :
(P. 1 of 4 pp.)
Age

Total

White

(P. 2 of 4 pp.)
Negro

All ages_...

Other

Chinese

Other

Under 55 to 1 4 15 to 24-25 to 34--

(P. 3 of 4 pp.)
Total

White

(P. 4 of 4 pp.)
Negro

35 to 44__
45 to 54. _
55 to 6 4 65 to 74-75 and over_

1 For

Chinese

All ages __

Under5 __
5 to 14
15 to 2 4 25 to 34--

Age

In
dian

Age

Age

In
dian

35 to 4 4 45 to 54-55 to 64___
65 to 74-75 and over.-

definitions, see footnotes to groups B and C.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

229

PAR. 1316

HORIZONTAL AND VERTICAL RULES

sEc.

13-B

In a n y fractional-measure table, if the end rule belongs at the bottom of the
last partial-measure shown on the page, it is extended across the entire table.
b. One-page box, multipage s t u b . - E n d rule will appear only on t h e last page
of the table since that ends the stub. (See fig. 20, examples B - 1 and B-2.)
T h e expression "1-page box" means t h a t after the first page the entire box is
repeated without change, spanner heads and all. T h e term "multipage stub"
means t h a t t h e stub differs from page to page, even though t h e sole difference lies
in variation of a center head at t h e top of t h e s t u b on each page.
c. One-page stub, multipage b o x . - E n d rule will appear at the bottom of
every page since the stub presentation is completed at the bottom of each page.
(See fig. 20, examples C - 1 and C-2.)
By "1-page stub" is meant t h a t after the first page, the entire stub is repeated
without change, center heads and all. By "multipage box" is meant t h a t t h e
box entries differ from page to page, even though the sole difference lies in variation
of a spanner extending across t h e entire boxhead on each successive page.
Note: Comparison of examples B - 2 a n d C - 2 brings out t h e arbitrary nature
of t h e basic rule, since t h e presence of t h e end rule at bottom of page 1 is made
dependent on t h e location of t h e caption specifying color.
d. Multipage stub, multipage b o x . - T h e end rule will appear at t h e bottom
of as many pages as there are distinctly different pages of boxhead. T h e location
of the pages which require the end rule depends upon whether the stub is completed
in successive pages before continuation of the box, or contrariwise.
( I ) S t u b c o m p l e t e d for e a c h page of b o x . - E n d rule appears at periodic
intervals. Thus, with a 4-page table with two pages of stub and two pages of
box, every second page will show the end rule if the stub is completed for the
first page of box before a n y portion is shown for the second page of box. The
end rule will appear on two pages in all. (See fig. 20, example D-1.)
(2) B o x c o m p l e t e d for e a c h page of s t u b . - E n d rule appears on each of
the last group of pages. Assuming again a table with two pages of s t u b and
two pages of box, the end rule will appear at t h e bottom of each of t h e last
two pages of the 4-page table if the entire box is completed on successive pages
for t h e first page of stub before a n y portion of the box is shown for t h e second
page of stub. (See fig. 20, example D-2.)

Sec. 13-C. The Vertical Rule (1321-1334)

1321. G e n e r a l . - T h e term "vertical rule" as used here covers all
types of continuous vertical lines appearing as part of the table.
In both letterpress and typewriter-offset work the same general
types of vertical rules are used. In letterpress, they are part of the
typographical composition. In offset, they are normally drafted by
hand to imitate the appearance of their letterpress counterparts.
1322. Types and weights of vertical rules.-Differences between
vertical rules in letterpress and offset work are a matter of mechanical
limitations. Letterpress usage is basic. Rules drafted on offset
tables, and those preprinted on large table forms on which the data
are to be filled by the typewriter, should be so planned that when
reduced for printing they will duplicate the appearance of the rules
used in ordinary letterpress work.

230

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

src. 13-C

THE

Pax.

VERTICAL RULE

1322

a. Single r u l e . - U s e d in three weights,1 as follows:
(1) H a i r l i n e . - T h e standard column rule.

This is t h e lightest weight of

rule available.
(2) M e d i u m w e i g h t . - T y p i f i e d in letterpress by a rule % point (%a") wide
when printed. Primarily, it is used to set off independent (nonadditive)
panels of columns. In the process, it improves the vertical legibility of the
table.
(3) B o l d , or heavy w e i g h t . - T y p i f i e d by t h e 1-point ( 2 " ) or 1% point
( 4 9 " ) rule in letterpress work. It is used infrequently by the Bureau of the
Census because of its excessive boldness.

b. Parallel r u l e . - A double rule composed of two rules running in parallel as
a single rule-unit. As used in this manual, "parallel" always refers to the double
hairline rule. The parallel rule is never used by the Bureau of the Census in
medium or heavy weight except when preprinting standard-table forms which
will be reduced for publication. (See par. 1323b, below.) Normally, the parallel
rule is used to set off additive totals, to precede s t u b continuations in fractionalmeasure tables, and, rarely, to set off independent sections of the table.
Example A. Comparison of types and weights of rules employed:
[All rules single hairline unless otherwise specified. " P " denotes parallel rule; single rules other than
hairline are denoted by: " M , " medium; and "B," boldface]
Panel 1
Stub head

Addi-

tive
total

Panel,2

Components
X

y

z

Non
additive
total

Aver
age

Panel 3
Percent
of total

Additive
total

Components
X

y

z

p
p
B
M
B
N o r E . - I n the above example, the 1)4 point boldface rule has been used as a panel separator in order to
illustrate its comparative weight. In practice, the medium rule would be used to separate the panels as
well as to set off the nonadditive total.

1323. Technical specification for rules.-Following are quasitechnical specifications for vertical rules. These are shown here as
a matter of information. In practice, the table-designer is not expected to indicate such technical detail. The use of the initials
"P," "H," "M," and "B" (parallel, hairline, medium, a n d bold)
will cover all cases where rule specification is essential as a part of
table design. Rules are always set as single hairline unless otherwise
specified.
Note: In Bureau of Census work, do not use t h e above initials, or t h e general
terms "medium" or "bold" in marking copy for the Government Printing Office.
T h e printer must have a kind of technical description best supplied by t h e office
of the Printing Section.
1 All references to letterpress (type set) rules as 1 point, 2 point, etc., signify width of the printing surface
of the rule; that is, the line width or thickness as printed, disregarding varying absorption capacities of
different printing papers. They do not refer to the thickness of the "body" or base of the rule. Thus,
both hairline and I-point rule are cast on 1-point, 2-point, 4-point, and 6-point bodies. To refer to a hairline
rule cast on a 2-point body as a "2-point rule" is incorrect and misleading. T h e term "2-point rule" always
means a rule with a 2-point printing face, regardless of the body on which it is cast.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

231

PAR. 1323

HORIZONTAL AND VERTICAL RULES

sEo. 13-C

a. Letterpress:
(1) Hairline: Single hair or hairline.
(2) M e d i u m : % point.
(3) Bold: 1 point or 1% point.
(4) Parallel: Parallel hairline.
b. Offset: Preprinted f o r m s . - T h i s refers to standard page or table forms prepared double-size in which the cell entries are added by the typewriter. Specifications here assume a 50-percent reduction before the report is reproduced.
(1) Hairline: Substitute % point.
(2) Medium: Substitute 1% point.
(3) Bold: Substitute 2 point (avoid this; it is very black when reproduced in
spite of t h e reduction).
(4) Parallel: Substitute parallel % point.

c. Offset: Drafted rules.-This refers to all rules drafted by hand. Specifications shown assume a 50-percent reduction before reproduction. In each case,
imitate t h e printed rules on standard offset tables as closely as possible. As a
practical matter, it is well to err by making the rule lighter, rather t h a n heavier,
t h a n its letterpress counterpart.
(1) Hairline: Duplicate % point as closely as possible (lighter, rather t h a n
heavier).
(2) M e d i u m : Duplicate 1% point as closely as possible.
(3) Bold: Duplicate 2 point as closely as possible (avoid).
(4) Parallel: Duplicate parallel % point as closely as possible but, draft the
rules lighter, rather t h a n heavier, t h a n % point. In particular, do not
draft t h e component lines too far apart. T h e vertical parallel rule is not
intended to frame t h e space left for it.

1324. Space requirements: General.-Because of typewriter limitations, the space required for vertical rules tends to be greater in
offset than in letterpress. However, the table-designer is not affected
unless he is planning a tight table in letterpress. Even then he would be
well advised to assume full normal spacing unless he is technically
qualified to determine exactly how much space each part of the table
will occupy typographically, not merely the vertical rules. The following specifications are rule-of-thumb.
a. Letterpress.-Allow 1 digit space (3 points in tables set in 6 point) for each
vertical rule. Normally, allow also 1 digit space on either side of the column rule
for normal "bear-off."2 Where necessary, in a tight table, t h e bear-off m a y be
eliminated. However, elimination of t h e bear-off (which means running figures
against t h e rules with no space between) may result in the last (or first) digit
being printed imperfectly or not at all. In a n y case, the table-designer should
always allow 3 points (1 digit space) for each column rule whether it be light,
medium, or bold single rule, or parallel rule.#
"Bear-off" refers to the white space left between the column rule and the maximum cell entry.
Technically each hairline rule will occupy 2 points of space; each medium or bold single rule, and each
parallel rule, will occupy 4 points of space. The rule-of-thumb given above will work successfully, therefore, in all tables where at least 50 percent of the rules are hairline. Few Bureau of Census tables are outside
this class. Calculation of space allotment to exact technical limits should be left to the Printing Expert.
He will take all typographical features of the table into account.
2

232


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEc. 13-C

THE

VERTICAL RULE

PAR.

1324

b. Typewriter o f f s e t . - H e r e , the space limitation tends to be absolute since the
horizontal spacing of the typewriter is involved. Every vertical rule, without
exception, must be allotted at least one full typewriter space. This is true irrespective of amount of reduction planned.
An additional typewriter space should be allowed on either side for bear-off,
particularly in preprinted standard table forms. Otherwise, it may be extremely
difficult for the fill-in typist to keep the columns of figures from touching the
printed vertical rules as she progresses down the page.

1325. Allocation of white space to stub and columns.-A common
and undesirable practice is to allow normal bear-off (white space
between maximum entry and vertical rule) in each column and, if
space is left over, to throw it into the stub in the form of leaders.
a. Stub leaders waste s p a c e . - T h i s practice is objectionable for several reasons.

First, it results, particularly when columns are narrow, in an overwhelmingly
vertical effect caused by the multiplicity of vertical rules extending from top to
bottom, unrelieved by white space. This makes it difficult for t h e eye to follow
a line horizontally across the page. Second, t h e over-abundance of leaders in
t h e s t u b makes it hard for the eye to follow across to the crowded field.
b. Preferred practice.-Allocate to t h e s t u b enough space to make possible two
leader dots after the longest line, where the longest line is not more than one-half
as much longer than the average line. (If it is longer, overrun it.) Allocate the
remaining "extra space" to the columns. First make sure each total and subtotal
column is at least as wide as its widest component column. Then even out the
width of the component columns until each has at least two spaces clear on either
side of each column rule. Only then return to the stub and assign extra space to
it in the form of additional leaders. Where the stub leaders still r u n to more t h a n
two or three times the longest s t u b line, assign more space to the columns.
c. Example of unnecessary crowding.-The examples shown in figure 21
illustrate a "before" and "after" effect of this general principle.

Example A - 1 shows the effect of insufficient white space in the columns
coupled with the corresponding type of error in the s t u b - l a c k of reader breaks.
Example A - 2 shows both errors as corrected in final proof. The minute
additional space in the columns makes a decided difference in legibility. Also,
it makes it possible to run the column heads across, not up.

1326. The hairline or lightface rule.-This discussion covers the
use of the hairline rule in letterpress and its corresponding rule
(¾ point) in offset planned for 50-percent reduction.
The hairline rule is the standard tabular rule, used where no other
type or weight of rule is specified.
a. As a column r u l e . - H a i r l i n e is normally used at t h e right of t h e stub, and at
the right of all columns which are not total columns or panel-ending columns.
However, no vertical rule appears after the last column on a n y table-page, unless
it is followed by a tracer or code column. Exception: T h e last column on the
left-hand page of a parallel table invariably carries a hairline column rule. Note,
"Bear-off' refers to the white space left between the column rule and the cell entries.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

233

FIG.

21

sro. 13-C

HORIZONTAL AND VERTICAL RULES

F I G U R E 2 1 . - A L L O C A T I O N OF W H I T E S P A C E TO T A B L E S

(See par. 1325)
&ample A-1. No white space in columns; no reader breaks in stub:
Station

0

0

Alabama_Arizona_
Arkansas.California_Colorado_Dist. of Col_Florida_Georgia_Idaho-Illinois ___
Indiana-Iowa_Kansas_Kentucky_Louisiana.Maine_
Massachusetts __
Michigan_.
Minnesota_
Mississippi_Missouri_Montana.Nebraska-Nevada_New Jersey __
New Mexico_New York.
North Carolina_
North Dakota_Ohio __
Oklahoma_
Oregon ___
Pennsylvania_South Carolina_South Dakota_-Tennessee ___
Texas_-

Utah_VermontVirginia__
Washington __
West Virginia__
Wisconsin_-Wyoming__
Alaska_Hawaii_Puerto Rico_-

_ Mobile __
Montgomery__Phoenix_-Little Rock__ Fresno __
Los Angeles__San Francisco_-_ Denver __
Washington_-_ Jacksonville ___
Miami_-- Atlanta_Boise_-Chicago__.
Indianapolis_Des Moines_-Wichita_-_ Louisville __
New Orleans_Eastport_Boston___ Detroit __
Sault Ste. Marie __
_ Minneapolis __
Vicksburg__Kansas City_St. Louis ___
Helena_-Miles City ___
North Platte_-Omaha_-Winnemucca_--

- Atlantic City_- Albuquerque__.
- Albany_
New York_Rochester_- Asheville_Raleigh_
-Bismarck_Cleveland_-Oklahoma City_-Portland_-Harrisburg__Pittsburgh___ Charleston ___
_ Huron _
_ Nashville __
- Amarillo_El P a s o - Fort Worth_-Houston_
Salt Lake City_-_ Burlington ___
Norfolk_
Richmond.Seattle_Spokane__ Parkersburg __
_ Madison __
Cheyenne_..
--- Juneau_-Honolulu_
San Juan_--

234

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

51. 5.54. 7 59. 7 66.3 74. 4 80.3 81.4 81. 0 78. l 69. 3 58.6 52. 2 76. 3
48. 2.51. 6 57.8,65. 3 73.4 79.6 81. 7 80.8 76. 3 66. 5 55.8 49. 4 65. 5
51. 2 55. 1 60. 7 67.0 75.0 84.5 89.8 88.5 82. 7 ,o. 6 59. 7 52.0 69. 7
41.4 44. 9,53. 0 62.1 70.3 77. 4 80.9 79.8 74.1 63. 6 52. 1 44. 2 62.0
45. 5 50. 3,54. 4 60.2 67. 2 75. 7 81.3 79. 5 72.0 62.3 52.4 45. 1 62. 2
54. 6 55. 5 57. 5 59. 4 62. 2 66.4 70.2 71.1 69.0 65. 3 60. 9 56. 6 62. 4
49. 9 52. 2 54. 2 55.0 56.8 58. 5 58. 5 59.1 60. 9 60. 5 56. 3 51.3 56. l
29.8 32. 7 39. 3 47. 1 56. 2 66.3 72.2 70. 7 62. 9 51. 2 39.8 32.3 50.0
33.4 35.3 42. 6 53.3 63. 7 72. 2 76.8 75.0 68.1 57.4 45. 2 36. 6 55.0
55. 4 58.0 62. 6 68. 7 75. 0 79. 9 82.1 81. 7 78.3 71.1 62.2 56.3 69. 3
66. 5 67.1 70. 2 72.8 76. 4 80.0 81.0 81.4 80.1 77.0 71.8 68.0 75.4
42. 6 45. 3 52.0 61.0 69.9 76.0 78.1 77.0 72.4 63. 0 52.1 44. 7 61.2
29.8 34.8 42. 7 50.4 57. 1 65.3 72.9 71.8 61.9 51.1 41.0 32. l 50. 9
23. 7 26.3 35. 3 46. 9 57. 5 67. 3 72. 5 71.6 65. 2 54.0 40.1 28.8 49.1
28.4 31. 1 40.0 52.1 62. 9 71. 6 75. 7 73. 7 66. 9 55. 7 42. 3 32. 2 52. 7
20.1 23. 7 35. 9 50.1 61.3 70. 6 75. 4 73.1 65. 6 53. 4 38.4 26.0 49. 5
31.3 34. 4 45.1 56. 4 65.1 74. 4 79. 4 78.3 70.6 58.6 44.8 34. 6 56.1
34.4 37. 2 45. 4 56. 4 66. 6 74. 7 78.6 77.0 70. 5 59. 3 46. 7 37. 6 57.0
54. 2 57. 3 62.8 68.8 75. 4 80. 6 82. 4 82.2 79. 2 71.0 61. 6 55. 6 69. 3
20.4 21. 5 28. 9 39.0 47. 7 55. 1 60.4 60. 7 55.8 47. 5 36. 7 26. 3 41. 7
27. 9 28.8 35. 6 46.4 57.1 66.5 71. 7 69.9 63. 2 53. 6 42.0 32. 5 49. 6
24.4 25. 3 33. 4 46. 2 58.0 67.4 72.1 70.3 63.5 52. 5 39.3 29. 3 48. 5
13. 3 12. 6 21. 6 37. 4 49. 0 58. 6 63.8 62.1 55.5 44. 6 32.0 20. 5 39.2
12. 7 15. 9 29. 6 46. 4 57. 7 67. 5 72.3 69.9 61.4 48.9 32. 4 19. 6 44. 5
48. 2 51.8 58. 5 65. 6 72.9 79.0 81.3 80.8 76. 3 66. 7 56.6 50. 0 65.6
28. 2 31. 2 42. 7 54.8 64.8 73. 6 78.1 76. 6 68.9 57. 7 43. 7 32. 5 54.4
31.1 34. 8 44.1 56.1 67.0 75.0 78.8 77. 5 70.5 58.8 45.4 34. 9 65.2
20. 2 23.0 32. 4 43. 5 51.6 59. 2 65. 7 65.0 56.6 44.9 32. 2 24. 2 43. 3
17.1 19. 4 31.8 47.0 56. 9 66. 9 74.3 71. 7 60.4 48.0 33.6 22.3 45. 9
22. 9 26. 6 36. 6 48. 6 58. 7 67.5 72.9 70.8 62.1 49. 7 36.6 26. 7 48.3
21. 9 25. 5 37.0 51. 2 62. 4 71.6 76. 7 74.4 66.8 54. 3 38. 5 26. 4 50. 6
28. 6 33. 5 40.0 46. 7 53. 9 62.8 70.6 69.3 59. 2 48. 3 38. 4 30.0 48.4
32. 5 33. 6 38. 6 47.8 58. 1 66. 6 72. 1 72. 5 66. 8 56.9 45.6 36. 4 52. 3
34.1 40. 5 45. 9 54. 0 63.3 72. 6 76. 7 73. 9 67. 9 56. 6 43. 3 34. 5 55.3
23.1 24.1 32. 7 46. 8 59.3 68.0 72.6 70.8 63. 1 52.1 39.3 28. 5 48. 4
30. 9 31.3 37. 7 49. 4 60.6 68.8 73.8 73.1 66.8 56.3 44. 2 35.0 52. 3
24.6 24. 6 31.8 44. 9 57.1 66.1 70. 7 69. 2 62. 4 51. 5 38. 7 29. 3 47. 6
35. 4 38. 5 44. 9 53. 9 62.6 68. 7 71. 70. 5 65.0 55.3 45. 1 37.8 54.1
41.4 43. 2 50. 2 59.4 68. 5 75. 7 78.8 77.0 71.1 62, 0 51. 0 43. 0 60.1
7.8 10.3 24. 2 42.1 54. 5 63. 7 69.8 67.3 58.1 44. 9 28. 5 14. 7 40. 5
26. 5 27. 4 34. 6 46.2 57.9 67.1 71. 4 70.0 63.9 53. 6 40. 9 31.2 49. 2
36.4 39. 6 50.0 59.8 67. 7 76.0 80.6 79. 7 72.8 61. 5 48.8 39. 3 59. 4
39. 4 42.1 46. 9 51.8 56. 9 62. 4 66. 7 66. 7 61. 7 54.2 46.8 41. 2 53.1
29.0 30.3 38. 9 50. 9 61.8 70.3 74.8 72.6 65.8 54.8 42. 8 32. 7 52.1
30. 7 32.3 39. 6 51.2 62.4 70. 7 74. 6 72. 9 66. 4 55. 7 43.2 34.2 52. 8
49. 9 52. 4 57.4 64. 5 72. 7 78. 9 81.4 81.0 76. 6 67.8 58.1 51. 7 66.0
11. 3 14. 3 28. 9 45.1 56.4 66. 2 71.8 69. 4 61.3 47. 7 31. 5 18. 7 43. 6
38. 6 41.6 49. 2 59.0 68.2 75.6 79.1 77. 8 71. 8 61.0 49. 41.0 59. 3
35.3 38.1 46. 9 55.8 64.1 72.8 76.8 75. 7 69.3 57. 7 45. 5 37.0 56.3
45.0 49.0 55. 8 63.4 71.5 79. 6 81.1 79.2 73.9 63. 5 52. 7
63.3
45.4 48.3 57. 7 65.0 72.3 79.9 83. 6 83.0 76.9 66. 7 55. 5 47. 5 65.2
52. 7 55. 9 63.3 69.3 75. 5 81. 4 83. 83. 2 79.0 70.3 61.0 54. 4 69.1
29.2 33.8 41, 7 49. 6 57. 4 67.4 75. 7 74. 5 64. 4 52. 5 41.1 31.9 51.6
18.8 19. 4 29. 1 43. 3 56. 5 65. 7 70.3 67.9 60.3 49. 2 36. 3 24. 4 45.1
40.6 42. 7 48. 2 56.8 66.2 74. 4 78. 7 77.4 71. 6 62. 5 51.4 43. 1 59. 5
37. 9 39.6 47. 2 56. 6 66. 5 74.1 78. 5 76. 5 70. 5 59. 6 48. 3 9. 8 57. 9
40.8 42. 8 46. 4 51.1 56. 6 61.4 65. 5 65.1 60.3 53. 7 46.8 42. 6 52.8
27. 5 31.3 39. 7 48.4 55. 5 62.8 69.0 68.1 59. 2 48. 3 38. 5 30. 5 48. 2
32. 5 34. 2 42.8 53.4 63.8 71.4 75. 4 73.9 67.3 56.1 43.8 35.2 54. 2
16. 7 19. 1 30. 6 45. 4 57. 6 67. 2 72.1 69.8 62.4 50.3 35. 2 22.8 45.8
25. 5 27.3 33. 1 40.9 50.3 60.4 66. 7 65. 6 57.0 44. 8 34. 8 28. 5 44. 6
27. 7 30.0 33. 7 40. 7 47. 7 54. 2 56. 7 55.3 50.4 43. 3 35. 7 30.9 42. 2
71.4 71. 5 71.8 73.0 74.8 76. 6 77.6 78.3 78. 2 77. 2 75. 0 73. 1 74. 9
75.0 74. 9 75. 4 76. 6 78.6 79. 7 80.1 80. 5 80.5 79.8 78. 4 76. 3 78.0

sEo. 13-C

THE

VERTICAL RULE

F I G U R E 2 1 . - A L L O C A T I O N OF

FIG.

21

W H I T E S P A C E TO T A B L E S

(See par. 1 3 2 5 ) - C o n .
Example A - 2 . Columns opened u p ; reader breaks added in s t u b :
Station
Mobile_-Montgomery__Phoenix_-Arizona-Arkansas_- Little R o c k California _ Fresno_Los Angeles_-San Francisco_Colorado_- _ Denver __
Dist.ofCoL Washington_-Jacksonville.-FloridaMiami_Alabama_-

Georgia.
Idaho_Illinois_Indiana_
IowaKansas_Kentucky__Louisiana_Maine_Mass

Atlanta_Boise_-Chicago__Indianapolis_Des Moines__
Wichita_-Louisville.-New Orleans_Eastport_Boston_--

Jan. F e b . Mar. Apr. M a y June J u l y Aug. Sept. Oct Nov. Dec. Annual
51. 5 54. 7 59. 7 66.3 74.4 80.3
48. 2 51. 6 57.8 65. 3 73.4 79.6
61. 2 55.1 60. 7 67.0 75.0 84.5
41.4 44. 9 53.0 62.1 70.3 77.4
45.5 50.3 54. 4 60. 2 67. 2 75. 7
54. 6 55. 5 57. 5 59.4 62. 2 66.4
49. 9 52. 2 54. 2 55.0 56. 8 58. 5
29. 8 32. 7 39.3 47. 1 56.2 66.3
33.4 35.3 42.6 53.3 63. 7 72.2
55. 4 58.0 62.6 68. 7 75.0 79. 9
66. 5 67.1 70.2 72.8 76.4 80. 0

81. 4 81.0 78.1
81. 7 80. 8 76.3
89. 8 88. 5 82. 7
80. 9 79.8 74.1
81. 3 79.5 72.0
70.2 71. 1 69.0
58. 5 59. 1 60. 9
72. 2 70. 7 62.9
76.8 75.0 68.1
82.1 81. 7 78.3
81.0 81.4 80.1

69.3
66.6
70. 6
63.6
62.3
65. 3
60. 5
51. 2
57. 4
71. 1
77.0

58.6 52.2
55.8 49.4
59. 7 52.0
52.1 44.2
52.4 45.1
60. 9 56. 6
56.3 51. 3
39.8 32.3
45. 2 36.6
62.2 56.3
71. 8 68.0

72. 4
61. 9
65. 2
66. 9
65.6
70.6
70. 5
79.2
55.8
63.2

63. 0
51.1
54.0
55. 7
53.4
58.6
59.3
71.0
47. 5
53.6

52.1
41. 0
40. 1
42.3
38. 4
44. 8
46. 7
61. 6
36. 7
42. 0

72.1 70.3 63. 5
63.8 62.1 55. 5
72.3 69. 9 61. 4
81.3 80. 8 76.3
78.1 76. 6 68. 9
78.8 77. 5 70. 5
65. 7 65.0 56.6
74.3 71. 7 60.4
72.9 70.8 62.1
76. 7 74.4 66. 8

52. 5
44. 6
48.9
66. 7
57. 7
58.8
44. 9
48.0
49. 7
54. 3

39.3 29.3
32.0 20. 5
32. 4 19.6
56.6 50.0
43. 7 32.5
45. 4 34.9
33. 2 24. 2
33.6 22. 3
36.6 26. 7
38. 5 26. 4

48. 5
39.2
44. 5
65.6
54. 4

59. 2
66.8
67. 9
63.1
66. 8
62. 4
65.0
71.1
58.1
63. 9
72.8

48.3
56. 9
56.6
52.1
56.3
51. 5

38.4
45. 6
43.3
39. 3
44. 2
38. 7

30.0
36.4
34. 5
28.5
35.0
29.3
37. 8
43.0
14. 7
31.2

48.4
52.3
55.3
48. 4
52. 3
47.6
54.1
60.1
40. 5
49.2
59.4

42.1 46. 9 51.8 56. 9 62. 4 66. 7 66. 7 61. 7 54.2 46. 8 41. 2
42.8 32. 7
43. 2 34. 2
58.1 51. 7
31. 5 18.
49.0 41. 0
45.5 37.0
49. 0 55.8 63.4 71.5 79. 6 81.1 79.2 73. 9 63. 5 52. 7 44. 9
48.3 57. 7 65.0 72.3 79. 9 83.6 83.0 76.9 66. 7 55.5 47. 5
55. 9 63.3 69.3 75. 5 81.4 83. 7 83.2 79. 0 70. 3 61.0 54. 4

53.1
52.1
52. 8
66.0
43. 6
59.3

67. 3

65. 5
69. 7
62.0
62. 2
62. 4
56. 1

50. 0
55.0
69.3
75. 4

42. 6
29. 8
23. 7
28. 4
20.1
31. 3
34. 4
54. 2
20. 4
27.9

45. 3
34. 8
26.3
31.1
23. 7
34. 4
27. 2
57. 3
21.5
28.8

52. 0
42. 7
35. 3
40.0
35. 9
45.1
45.4
62.8
28. 9
35.6

69. 9
57. 1
57.5
62. 9
61. 3
65.1
66.6
68.8 75.4
39.0 47. 7
46. 4 57.1

Detroit_
24. 4
Sault Ste. Marie __13.3
Minnesota__ Minneapolis_12. 7
MlssissippL Vicksburg.-48. 2
Missouri.
Kansas City_28.2
St. Louis _
31.1
Montana_- Helena_20.2
Miles City _
17. 1
Nebraska_- North Platte_22.9
Omaha_21.9

25.3
12. 6
15. 9
51. 8
31. 2
34.8
23.0
19. 4
26.6
25.5

33.4
21. 6
29.6
58. 5
42. 7
44. 1
32.4
31. 8
36.6
37. 0

46. 2
37.4
46. 4
65.6
54. 8
56.1
43. 5
47. 0
48.6
51. 2

58.0
49. 0
57. 7
72.9
64. 8
67.0
51. 6
56. 9
58. 7
62.4

Nevada_
Winnemucca__
New Jersey_ Atlantic City_New Mexico_ Albuquerque_._
New York __ Albany__
New York __
Rochester_-N. Carolina_ Asheville_-Raleigh
N. Dakota.- Bismarck_ Cleveland _
OhioOklahoma_- Oklahoma City_-

28. 6
32.5
34. 1
23.1
30.9
24. 6
35. 4
41.1
7. 8
26. 5
36.4

33. 5
33.6
40. 5
24. 1
31.3
24. 6

40.0
38.6
45. 9
32. 7
37. 7
31. 8
38. 5 44. 9
43. 2 50. 2
10.3 24. 2
27. 4 34.6
39. 6 50.0

46. 7
47. 8
54. 0
46.8
49. 4
44. 9

53. 9 62.8
58.1 66.6
63.3 72.6
59.3 68.0
60.6 68.8
57.1 66.1

_ Portland __
Harrisburg__Pittsburgh_-S. Carolina_- Charleston_S. Dakota_- Huron_
Tennessee__ Nashville __
Texas_
Amarillo.
El Paso __
Fort Worth_Houston_-

39. 4
29. 0
30. 7
49. 9
11. 3
38.6
35.3
45.0
45. 4
52. 7

Utah_Salt Lake City__Vermont_- Burlington__Virginia__- _ Norfolk _
Richmond __
Washington_ Seattle_-Spokane__
W. Virginia_ Parkersburg__.
Wisconsin _ Madison_Wyoming__ Cheyenne__-

29. 2
18. 8
40.6
37.9
40. 8
27. 5
32. 5
16. 7
25.5

31. 9
24. 4
43.1
39.8
42. 6
30. 5
35.2
22.8
28.5

59. 5
57. 9
52.8
48. 2
54. 2
45.8
44. 6

Juneau_Alaska_Honolulu_Hawaii_Puerto Rico_ San Juan_-

27. 7 30.0 33. 7 40. 7 47, 7 54. 2 56. 7 55.3 50.4 43. 3 35. 7 30. 9
71.4 71. 5 71. 8 73.0 74.8 76.6 77.6 78.3 78.2 77.2 75. 0 73.1
75. 0 74. 9 75.4 76.6 78.6 79. 7 80.1 80.5 80.5 79.8 78.4 76. 3

42. 2
74. 9
78. 0

Michigan_

Oregon_-

Pa


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

61.0
50. 4
46. 9
52. 1
50.1
56.4
56.4

78.1
72. 9
72.5
75. 7
75. 4
79. 4
78.6
80.6 82.4
55.1 60.4
66. 5 71. 7
76.0
65.3
67.3
71. 6
70.6
74. 4
74. 7

67.4
58.6
67. 5
79.0
72.6
75.0
59.2
66. 9
67.5
71. 6

53. 9 6 2 . 6 68. 7
59. 4 68. 5 75. 7

42.1 54. 5 63. 7
46. 2 57. 9 67. 1
59. 8 67. 7 76.0

70.6
72. 1
76. 7
72.6
73.8
70. 7
71. 7
78.8
69.8
71.4
80.6

77.0
71. 8
71.6
73. 7
73.1

78.3
77.0

82. 2
60. 7
69. 9

69.3
72. 5
73. 9
70. 8
73. 1

69. 2
70. 5
77.0
67.3
70.0
79. 7

44. 7
32.1
28.8
32. 2
26. 0
34. 6
37.6
55. 6
26.3
32. 5

55. 3

45. 1

62.0
44.9
53.6
61. 5

51. 0
28. 5
40. 9
48.8 39.3

30. 2 38. 9 50.9 61.8 70.3 74. 8 72.6 65.8 54.8
32.3 39.6 51. 2 62. 4 70. 7 74.6 72. 9 66.4 55. 7
52. 4 57. 4 64. 5 72. 7 78. 9 81.4 81. 0 76.6 67. 8
14. 3 28. 9 45.1 56. 4 66. 2 71. 8 69.4 61. 3 47. 7
41.6 49. 2 59. 0 68. 2 75.6 79.1 77.8 71. 8 61.0
38.1 46. 9 55. 8 64.1 72.8 76.8 75. 7 69.3 57. 7

33. 8
19. 4
42. 7
39.6
42. 8
31.3
34. 2

41. 7
29.1
48. 2
47.2
46. 4
39. 7
42.8
19.1 30. 6
27. 3 33.1

49. 6
43.3
56.8
56.6
51.1
48.4
53.4
45.4
40. 9

57.4
56. 5
66. 2
66.5
56. 6
55. 5
63.8
57. 6
50.3

67. 4
65. 7
74. 4
74.1
61.4
62.8
71.4
67. 2
60.4

75. 7
70.3
78. 7
78. 5
65. 5
69.0
75. 4
72.1
66. 7

74. 5
67.9
77.4
76.5
65.1
68.1

73. 9
69.8
65. 6

64. 4
60.3
71.6
70. 5
60. 3
59. 2
67.3
62.4
57.0

52.5
49. 2
62.5
59. 6
53. 7
48.3
56.1
50.3
44. 8

41.1
36.3
51.4
48.3
46. 8
38. 5
43. 8
35.2
34. 8

235

61. 2
50. 9
49. 1
52. 7
49. 5
56.1

57. 0
69.3
41. 7
49. 6

56.2

43. 3
45. 9
48.3
50. 6

56. 3

63. 3
65.2
69.1

51. 6
45.1

PAR.

1326

HORIZONTAL A N D VERTICAL RULES

s o . 13-C

however, that this is not the last column on the table-page since t h e table-page in a
parallel table consists of the two facing book-pages taken together. (See fig. 9,
p. 22.)
T h e hairline rule also is used at the left of the stub to set off tracer columns or,
on occasion, line or code number columns.
b. As a nonadditive total r u l e . - W h e r e a total column is nonadditive, t h a t
is, does not represent the summation of the columns shown immediately to the
right, it is normally followed by a hairline rule.
On occasion, however, a medium rule is used for this purpose if the nature of t h e
total warrants special emphasis. (See pars. 1327 and 1328a.)
c. As a panel r u l e . - W h e r e more t h a n a few major panels appear in t h e boxhead, it is sufficient t h a t they be closed by hairline rules extending to the top of the
panel spanner. A multiplicity of medium-weight rules terminating spanners may
defeat the purpose of their insertion. (See par. 1328.)

1327. Medium and bold rules: General.-Medium and bold rules
are best used to set off independent (nonadditive) panels or columns.
In the process, they also improve the vertical legibility of the table.
a. Basic principle.-Medium and bold rules should not be inserted to separate
the members of a series of columns or panels which, taken together, add to a total
column within the same panel or to a total panel within the same table. In other
words, these rules should stress the independence, not the interdependence (if any),
of the columns or panels which they separate. This function should take precedence over the improvement of vertical legibility of the presentation.
b. Signalling additive-nonadditive situations.-The principle stated above relates to a basic problem in tabular presentation; t h a t is, the need for devices
which will help the reader to identify, quickly and surely, the additive and t h e
nonadditive relationships of t h e various columns and panels. Two signalling
methods are available:
(1) A positive signal is provided by the presence of the parallel total rule at the
right of a total column or panel. This tells the user t h a t t h e columns (or panels)
on the right add to make t h e total column or panel.
(2) Within limits, a negative signal is provided by t h e presence of t h e medium
or bold rule. This warns the user that the columns or panels thus separated are
not appearing in an additive situation.
c. Medium rule as a negative s i g n a l . - T h e negative signal is essential in a case
such as found in paragraph 1328, example B. There, the first panel is a total
panel; t h e next three panels add to the first; t h e last panel is not to be included in
t h e addition. If the positive signal (the parallel total rule) is to have meaning, t h e
user must be warned exactly where its range of influence stops. T h e insertion of
t h e medium rule preceding the last panel does this. However, this important
usage will be negated or obscured if, in t h e same table er volume, t h e mediumweight rule also is used as an additive total rule. (See par. 1328d.)

1328. Medium r u l e . - T h i s discussion covers the use of the % point
medium rule used in letterpress and its corresponding rule (1}%point)
used in typewriter-offset planned for 50-percent reduction.
T h e medium rule is recommended for three primary purposes.
A fourth usage, to be avoided where possible, is its use as an additive
total rule.
236

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sro. 13-C

THE VERTICAL RULE

PAR.

1328

a. Panel or column r u l e . - W h e r e t h e table is divided by top spanners into
several independent segments, use t h e medium-weight rule for divide purposes;
t h a t is, to set off t h e panels from one another.* However, t h i s rule should not be
inserted among the members of a series of panels which add to a total at t h e left.
T h e same line of reasoning obtains as to t h e use of t h e medium-weight rules to
set off columns within spanners, or in a table without spanners.
b. Nonadditive total r u l e . - A nonadditive total column may be set off with
a medium column rule at its right if the column is considered sufficiently important.
Use of t h e medium rule for this purpose should be held to a minimum. Usually
such columns do not need t h e additional emphasis.
Example A. Right (simple use as a panel divider):
Central beating

Electric lighting

Cooking facilities

Radio

Year

Example B. Right (important use to indicate presence of an independent panel):
The State

Urban

Rural

Total farm

Age

Example C-1. Not recommended (use to separate members of additive series obscures meaning of
medium rule as used in example B; use hairline instead):
The State

Urban

Rural-nonfarm

Rural-farm

Age

Example C-2. Right (panels will add to a total, but the total panel is not shown in the table):
Native white

Foreign-born
white

Other races

Negro

Age

Example D. Not acceptable (over-use of panel-divide rule; use hairline instead):
County A

Year

County B

County O

County D

County E

County F

Fe
Fe
FeFe
Fe
Fe
Male male
Male male
Male male
Male male
Male male
Male male

4 A multiplicity of medium-weight rules to set off panels will defeat their purpose. On a document-size
page, the number of medium rules used for panel-divide purposes might well be restricted to 3; at most, 4.
On a census-size page, the maximum might be increased to 5, or even 6. If these numbers are exceeded, use
hairline rules instead (see par. 1326c). For page sizes, see par. 211.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

237

PAR.

1328

sEc. 13-C

HORIZONTAL AND VERTICAL RULES

Example E. Right (if one category is set off, all parallel categories mustEbeset off, even though one or
more are represented by single columns instead of by spanner groups):
1040

1930

1910
1920,

Age

total

c. To set off a nonadditive, or independent, column, or group of columns, at
the right of the t a b l e . - T h r e e examples of this use may be mentioned.
(1) A general case is where the table is evenly divided into a number and a
percent panel. Here, the medium rule is useful to improve vertical legibility
of the page.
Example F:
Number

Percent distribution

Age

Group A

Group B

Group O

Group A

Group B

Group O

(2) A common case is where a detailed distribution is given one major
category, but only the total number of units is shown (classified by the s t u b
entries) for another category. Usually, such a· " t o t a l number" column is
shown at the right of the table, preceded by a medium-weight rule. Note
in example G t h a t the medium rule is also desirable to terminate the influence
of t h e parallel total rule.
Example G:
Total
experienced
persons

Experienced persons by wage income groups
Under
$200

$200 to
$499

$500 to
$999

$1,000 to
$2,499

$2,500 to
$5,000
$4,999 and over

Number
of wage
workers,
total

(3) A special case is illustrated in examples H - 1 and H - 2 . Here the " N o t
reporting" column is separated considerably from its coordinate column
"Reporting on monthly rent." Use of the medium rule is essential to terminate
t h e influence of the parallel total rule which appears at the left. Also, it makes
clear t h a t the " N o t reporting" column is not coordinate with the columns
immediately preceding it.
d. As an additive total rule: Not recommended.-In rare instances, it may
be necessary to use the medium-weight rule as an additive total rule. This
may occur where spanners have been deleted because of lack of space, and the
parallel rule (rising to the same level) already has been used for other purposes.
T h i s situation most frequently arises where a subtotal is thus set off by a medium
rule, as in t h e examples below. This usage of the medium-weight rule should be
held to a minimum since it obscures its more important usage to set off independe n t panels and nonadditive columns.

238

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEo. 13-C

THE VERTICAL RULE

PAR.

1328

Example H-1. An extreme case. Not recommended:

State

Total,
all
units

Reporting on
Under
monthly
$15
rent

$15 to

$30 to

$29

$49

$50 to $75 and
over
$74

Not
rptg.

$20 to

$30 to

$29

$49

$50 to $75 and
over
$74

Not
rptg.

$20 to

$19

Example H-2. Possible alternative:
Total, Reporting on
all
units monthly
rent

State

Under

$15 to

$15

1329. Bold r u l e . - T h i s discussion covers both the bold rule used in
letterpress (1 point or 1% point) and its corresponding rule in offset
planned for a 50-percent reduction (2 point).
The bold rule should be avoided where possible. It is too black for
any use except to "shock" the reader into observation. In itself, this
tends to distract the user from the data.
In general, the bold rule should be reserved for cases where all
spanners are crowded out for lack of space; the use of parallel rules is
essential for additive totals; the medium-weight rule must be used to
indicate an independent column or a subtotal; and in addition it is
imperative that still another basic division of the boxhead be established. Thus, in the case illustrated below, all of the above types and
weights of rules appear, and all must rise to the top rule of the table.
This presentation is defective. However, any instance where the bold
rule is essential is likely to be equally defective.
Example A. U s e of boldface r u l e (table should be redesigned to avoid this complexity of rules) :
Total
units,

Type A

Report.ing on
X

Not re- Total
porting Tunits,
ype B

Reporting on

Not reporting

y

1330. Parallel r u l e . - T h i s is used for three main purposes:
a. To set off each partial measure in a fractional-measure table.-This usage
does n o t prevent its use also to set off an additive total rule, even though both rise
to the same level of box with no superior rule intervening. When the parallel rule
is used to set off a partial measure, it always precedes a new s t u b ; its significance
when so used is obvious to the user.
Example A. Use of parallel to set off both additive totals and partial measure of a half-measure table,
doubled up:

Area
Rand

Total
314


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Male
163

Female
151

Area
Welch..

Total

Male

1,299

789

Female

239

510

PAR.

1330

src. 13-C

HORIZONTAL AND VERTICAL RULES

b. To s e t off independent sections of the t a b l e . - N o t recommended. This
usage of t h e parallel rule is no longer as common as it once was. T h e medium or
bold single rule is suggested for t h e purpose in order to reinforce t h e use of the
parallel rule as signifying an additive total. Special cases of the use of the parallel
rule to set off independent sections are where (1) all spanners have been omitted
because of lack of space and (2) the medium-weight rule (rising to t h e top rule of
t h e table) has been used for nonadditive totals.
Example B. Not recommended. Better to use medium rule for panel divide and single hairline for
nonadditive totals.
Total A
units

Percent of Percent of
y
X

Total B
units

Median

Percent of Percent of
y
X

Median

c. To set off additive totals: Frequency d a t a . - W h e r e a total column represents t h e summation of t h e series of columns t h a t follows, the total column
normally carries a parallel rule at its right. T h e same is true of a total panel.
(1) N o r m a l c a s e . - H e r e , t h e total column (or panel) and t h e component
columns are strictly parallel in form.
Example C:
Urban

The State
Male

Total

Female

Total

Male

Rural
Female

Male

Total

Female

(2) Special c a s e s . - S e v e r a l special cases occur where t h e parallel rule is
appropriate even though the pure additive situation is not present. Some of the
more common are illustrated below. In each instance, t h e total panel provides
extra information not shown for all components.
Example D. Number and percent for total panel; only number of components:
T h e State
Number
Male

Rural

Male Female Total

Male Female

Percent
Total

Total

Urban

Female Total

Male Female

Example E. Total panels contain categories not distributed in component panels:
Urban, 1940

T h e State
1940
Total

Male

240

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

1930,

Female

total

1920
total

Total

Male Female Total

Rural, 1940

Male

Female

src. 13-C

THE

1330

PAR.

VERTICAL RULE

Example F. Total panel with full distribution but not all components with complete distribution.
However, the total, at least, should be shown for every component:
Urban

T h e State
Total

Male

Female

Rural-nonfarm

Male

Total

Female

Total

Male

Female

Ruralfarm,
total

Example G. Grand total column with full distribution for component groups:
The
State,
total

Rural-farm

Urban
Total

Male

Female

Total

Male

Rural-nonfarm
Female

Total

Male

Female

d. To set off additive totals: Derived data. The use of t h e parallel rule in
setting off "total" columns or panels for derived data requires special attention.
The parallel rule is not reserved for totals representing summation of the actual
derived data shown. In effect, this would restrict its use for derived d a t a to cases
where percent distributions are shown; and then only where t h e distribution adds
to 100.0 (or the equivalent, since the Census Bureau rounds, but does not
normally force percentages). Thus:
100. 0
100.0

50.5
25.0

20.5
45. 0

28. 9

30.0

The use of t h e parallel rule extends considerably further. Rule: T h e use of the
parallel rule to set off "total" columns or panels involving derived data depends
upon whether the frequencies for t h e classes distributed would add to t h e frequency for t h e total column if t h e frequencies were shown instead of t h e derived
figures. I t s use does not depend upon whether the derived data themselves add
to t h e value shown in t h e total column. This means that t h e parallel rule is used
in all cases with derived data just as it would be used if frequencies were substituted
for t h e derived statistics.
Example H. Parallel rules used for each panel, although additive feature differs:
Age by sex

Sex by age
Age

Total

Male

Female

Total

Female

Male

Total

100. 0

100. 0

100. 0

100. 0

50. 2

49. 8

Under 45 years __
45 and over_-

73. 3
26. 7

73. 1
26. 9

73. 6
26. 4

100. 0
100. 0

50.0
50. 6

50. 0
49. 4

Example J. Percentages (each cell a percent of a corresponding cell in another table):
Sex

All ages

Total

Male_.
Female__


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Under 45
years

45 years
and over

89. 8

65. 1

24, 6

45. 1
44. 6

32. 7
32. 5

12. 5
12. 2

241

PAR.

1330

sEo. 13-C

HORIZONTAL A N D VERTICAL RULES

Example K. Medians:
Male
Area
Total
United States_Urban_Rural-nonfarm.
Rural-farm_-

White

Female
Nonwhite

Total

White

Nonwhite

24. 3

24.5

22, 9

21, 6

21.8

19, 9

24. 7
23. 5
24. 3

24. 8
23. 6
24. 7

23. 2
22. 8
22. 7

21. 9
21.3
21.0

22. 1
21.4
21.2

20.0
19. 6
19. 8

1331. Combinations to be avoided.-Since both the parallel and
the medium single rule are used for a variety of purposes, confusing
combinations should be avoided.
a. Parallel r u l e . - W i t h i n a given level of box, only one parallel rule may be
used, unless subsequent appearances are separated from each other, and from the
first appearance, either by(1) A rule (irrespective of type or weight) rising to a higher level of box; or
(2) A medium or bold rule rising to t h e same level of box.
Exception: The sole exception to the above is t h e use of the parallel rule (in t h e
fractional-measure table) to signify an additive total and to precede t h e second
(and subsequent) appearance of t h e stub. This combination is not confusing because of t h e stub repetition.

b. Medium r u l e . - W h e n used with a given meaning within a given box level,
it should not be used again with a n y other meaning within the same box level,
unless(1) A rule (of a n y type or weight) intervenes which rises to a higher box
level; or
(2) T h e column arrangement is completely obvious to the most casual user.
Usage must be consistent.-If a medium rule is inserted in a n y portion of t h e boxhead, it must appear in all parallel situations elsewhere in t h e boxhead. Specifically, when used to set off coordinate panels, all table sections coordinate with
such panels must also be set off, even though this involves setting off two adjoining
individual columns.
Example A. If one category is set off, all coordinate categories must be set off':
1930

1940

Total

Male

Female

Total

Male

Female

1920,
total

1910

total

1332. Range of influence of vertical rules.-When inserting parallel
or medium-weight single rules, the table must be scrutinized carefully
in terms of range of influence of each such rule and for parallelism of
presentation.
By "range of influence" is meant the extent to which the feature
signalled by the given rule carries through the columns on the right.
Thus, a parallel rule indicating an additive total signals a situation
which, logically, will apply across the table until the final component

242

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEc. 13-C

THE VERTICAL RULE

PAR.

1332

part is reached. At that point, it is imperative that the rule after the
final component be distinctive to show that the end of the influence of
the additive parallel rule has been reached. (See par. 1327b.) On the
other hand, it is important that no intervening rule be such as to imply
that the end of the influence of the additive parallel rule has been
reached when, in fact, it has not.
The following observations may be helpful:
a. Parallel r u l e . - T h e range of influence of a parallel rule normally extends to
t h e right until it encounters(1) A medium rule rising to t h e same box level;
(2) Any rule (any t y p e or weight) rising to a higher box level;
(3) A parallel rule rising to t h e same box level (indicating t h e end of the
partial measure in a fractional-measure table) ; or
(4) T h e right-hand edge of the table-page.
b. Medium r u l e . - T h e range of influence of a single medium rule extends to
t h e right until it encounters(1) Another medium rule rising to t h e same level of box;
(2) A parallel rule rising to t h e same level of box where t h e parallel indicates a
major division of the boxhead (this use of t h e parallel rule is not recommended) ;
(3) Any rule (of a n y type or weight) rising to a higher level of box; or
(4) T h e right-hand edge of the table-page.
c. Rise of r u l e . - A n y rule (of any type or weight) rising-to a given level of box
curtails t h e range of influence of a n y other rule (of equal, inferior, or superior type
or weight) which rises to a lower level of box.

1333. Rule extension at bottom of table or table-page.-Vertical
rules are extended below the last line of entries, as indicated below:
a. Bottom of t a b l e . - V e r t i c a l rules are extended to meet t h e horizontal bottom
(end) rule of t h e table.
b. Bottom of table-page, where stub continues.-Where t h e bottom of t h e
table-page is not the end of the s t u b presentation (that is, where t h e s t u b continues to t h e next page), t h e horizontal end rule does not appear. (See par.
1316b.) Here, all vertical rules extend downward t h e depth of one additional
line (if letterpress) or one-half line (if offset) below t h e last line of entries, or
below t h e last stub line, whichever is lower. All vertical rules on a given page
should extend downward the same distance. They should never be permitted
to run '"ragged"; that is, to run down dissimilar depths below the last line of t h e
table-page.

1334. Tables in rules or boxes.-In Bureau of the Census practice,
tables are rarely enclosed in rules nor are they "boxed-in" with rules.
In general, the presence of vertical rules at the edge of the table, or of
the box in which the entire table (including title and footnotes) is
placed, distracts from the presentation in terms of the meaning of the
data by overemphasizing the mechanics of construction.

776554-49-17


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

243

PAR. 1334

All ages __

AGE

AND
T A B L E 1 . - P O P U L A T I O N , BY
AGE AND COLOR . • . .

Total

White

Nonwhite

100

75

25

55
45

45

Under 45-45 and over_

30

Age

Total

AU ages_-

10

Under 45--45 and over_

15

Example C. R i g h t :
TABLE

1 . - P O P U L A T I O N , BY
COLOR

Age

Under 45.45 and over.

244


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

13-C

Example B. Undesirable:

Example A. Undesirable:
T A B L E 1 . - P O P U L A T I O N , BY
COLOR . . .

Age

SEC.

H O R I Z O N T A L A N D VERTICAL R U L E S

AGE

AND

Nonwhite

Total

White

100

75

28

45

10

45

30

15

NonWhite white

100

75

55
45

45

10

30

15

Chapter 14
THE FIELD (1401-1448)
Sec. 14-A. General (1401-1402)
1401. Definition.-The field of the table is t h a t portion extending
from the bottom rule of the boxhead to the bottom of the table-page;
and from the column rule at the right of the stub to the right-hand
edge of the table-page; or, in fractional-measure tables, to the next
repetition or continuation of the stub. The field is the depository
of statistical or other information.
1402. Component parts defined.-In terms of construction, the
component parts of the field are comparatively few in number. The
more important of these are listed and described below.
a. Cell.-The basic unit of presentation. (See sec. 14-B.) T h e intersection of
a n y stub data-line caption with a n y column caption,
T h e description of a given cell consists of the combination of the stub caption
and the column caption, supplemented or qualified by related center heads and
colon or read-in lines in the stub, and the applicable spanner heads in the boxhead.
T h e cell entry may be informational (figure, word, phrase, reference symbol,
etc.) or noninformational (leaders). On rare occasions, it may be left blank
because it has been braced out of existence or for another special reason.
b. L i n e . - A horizontal row of cells with a common classification extending
across from a data-line caption in t h e stub. (Properly speaking, t h e " l i n e "
includes t h e caption.)
c. C o l u m n . - A vertical row of cells with a common classification extending
down from the descriptive entry in t h e column head. (Properly speaking, t h e
"column" includes t h e individual column head.)
A col-

umn

A line

61

A cell
72

28

81

154
75

d. Unit-indicator.-A specification of measure or presentation unit. Usually
it appears at the top of a column directly above t h e first d a t a line in the field.
Occasionally, however, it may appear elsewhere in t h e field, as in a " u n i t of
measure" column, or in a· field spanner. T h e unit-indicator consists of a statement
such as "dollars," "bushels," "acres," etc.; or it may indicate digit omission as


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

245

PAR.

1402

THE

sEc. 14-A

FIELD

" a d d 000," "thousands," etc. Where it appears at t h e top of a column, it is set
in italic, if letter press; it is placed within parentheses, if typewriter offset. (See

sec.14-C.)
Examples:
Acres

(Units)
29
64

Dollars

65
28
69
84

6, 254
1, 542

7, 890
6, 289

(Sq. m i . )

2,516

7,418

6,736

72

65

4,987

e. Field spanner.- A t y p e of spanner appearing recurrently in, a n d constituting
a component part of, t h e field. Functionally, it may be thought of as a part of t h e
s t u b transposed into t h e field because of space difficulties. T h e field spanner is
inherently undesirable, but it is necessary in certain cases. I t s major defect is
t h a t it breaks across t h e column rules and separates the column captions from
t h e entries they describe. (See sec. 14-D.) In t h e following example, t h e field
spanner is indicated by an arrow.
Example:
Other races.

New York City

All classes..

White __
Negro __
Other races_.

2

3

9

80
50

65

5

23
2

Bronx Borough
32
20
10
2

21
15
1

A common error is the erroneous inclusion of the uppermost of a series of
field spanners in the boxhead of t h e table. T h i s practice leads to serious confusion. (See par. 1434.)
f. B r a c e s . - A means of indicating t h e funnelling together of two or more d a t a
lines into fewer lines; and a subsequent reexpansion to a greater number of lines.
Braces are always placed at t h e left of t h e entries and at t h e right of t h e column
rule. (See sec. 14-E.)
Example;
White
Nonwhite

832
646

269, 324
114, 007

Sec. 14-B. The Cell (1411-1414)

1411. Definition.-The cell is the basic unit of presentation. It consists
of the intersection of any stub data-line caption with any column caption.
T h e description of a given cell consists of the combination of the stub
caption and the column caption, supplemented or qualified by related
center heads and colon or read-in lines in the stub, and the applicable
spanner heads in the boxhead.
1412. Function and purpose.-The cell provides placement for the
informational or noninformational entries defined by the combination
of the controlling stub line-caption and column caption.
1413. Types of entries.-Normally, each cell should contain an
entry of some kind. Cells left completely blank are subject to misinterpretation by the user. Blank cells also may cause difficulty in
the verification of transcription and computing operations. "Blanks"
246

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEc. 14-B

THE CELL

PAR.

1413

are particularly objectionable where the user is unable to tell whether
the cell was deliberately left blank or whether a clerical omission was
responsible.
The major types of cell entries are listed below. The presence of
any of these types of entries prevents the cell from being classified as
an empty or "blank" cell.
a. A specific observation or d a t u m . - U s u a l l y this is a specific figure. However,
it may be a date, a name, a class, a summary statement, or a code entry. It may
be accompanied by a footnote reference symbol.
b. A footnote reference symbol, in parentheses.-A footnote reference symbol
(in parentheses) m a y stand alone in a cell. It is normally required in a n y
instance where(1) D a t a are not available.
(2) D a t a are available but are suppressed by t h e Bureau of t h e Census
because of "disclosure"; t h a t is, their publication in t h e given context would
permit t h e reader to identify t h e specific person or firm to which they apply.
(3) A sample observation or d a t u m has been omitted because it is less t h a n
a given frequency or has too great a sampling error to warrant publication.
(4) D a t a are omitted for other specific reason; t h e reason is specified in
t h e footnote.
c. Leaders (letterpress) or three periods (offset) : General r u l e . - G e n e r a l l y
required in a n y instance where(1) D a t a are not applicable. This refers to a case where t h e combination of
line caption and column caption is such as to make an entry illogical. Example:
Intersection of a percent line and a percent column; or intersection of a median
line and a ratio column; etc.
(2) A derived entry (percent, median, etc.) is not shown because(a) T h e base is less t h a n a lower limit specified in t h e headnote, as Examples: Where percents are not shown where t h e base is less t h a n 100;
Where medians are not shown where t h e base is less t h a n 50.
(b) The value, when computed, is less t h a n a lower limit specified in a
headnote, asExample: Where percents are not shown where less t h a n 0.1 after rounding.

d. Specialized code entries.-In certain materials, specialized code entries
m a y be used to express one or more of t h e concepts listed above as requiring a
leader (letterpress) or three periods (offset). In such cases, t h e practice customary
for t h e given subject m a t t e r should be followed carefully. In this respect,
however, one general rule should be observed:
General rule: No code entry should appear in any table unless the meaning
of t h e given code is made clear in a readily accessible and observable point in the
report. Preferably, all such codes should be explained on t h e same page as t h a t
on which they appear.

1414. Braces in cells.-One or more braces may affect a given cell
in addition to any of the above entries. Where a brace appears in a
table, it invariably appears at the right of a column rule, preceding a.
given observation or cell entry. (See sec. 14-E.)


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

247

PAR.

1421

sEc. 14-C

Sec. 14-C. The Unit-Indicator (1421-1426)

1421. Definition.-The unit-indicator is a device designed to make
clear the exact unit of measurement to which the given statistics
refer. It is sometimes referred to, as in the G. P. 0. Style Manual, as
the "unit of quantity."
The unit-indicator and similar auxiliary devices may appear in the
field, stub, or boxhead. The nature of this device, restrictions on its
use, etc., are discussed in detail in sec. 9-F. The principles outlined
there, although the application is in terms of use of the unit-indicator
in the stub, apply generally to its use elsewhere in the table.
1422. Placement in the field.-In the field, unit-indicators normally
are placed at the top of the affected column on the line above that
allocated to the first data 'line of the table-page.
Example A:

Example B:
(Lower rule
of boxhead)

(Lower rule of boxhead)
1,000
1940
1941
1942-1943

1944

acres

6, 589

7, 483
7, 966
7, 307
6, 830

1,000

bushels

62, 570
113, 649
109, 745
82, 685
40, 224

1,000

dollars
35, 579
29, 940

32, 698
42, 168
40, 133

Dol. per
ton
4. 58
8.11

5.80

7. 30

7. 00

1423. Range of influence.-A unit-indicator placed at the top of a
column applies to all entries in that column on the given table-page.
I t s influence does not carry over to the next page; it must appear at
the beginning of each new table-page.
Three special cases deserve mention:
a. Divide tables.-Here, t h e table is set in decks, with t h e entire boxhead
repeated, or an entirely new boxhead shown, for each deck. (See par. 217.)
T h e unit-indicator must appear at t h e top of t h e field in each deck. T h e parallel
top rule of each succeeding deck actually represents a continuation of the top
r u l e of t h e table, hence t h e range of influence of t h e unit-indicators in the preceding
deck is terminated. A new appearance is necessary, therefore, as on a n y continued
page.
b. Fractional-measure t a b l e s . - T h e same rule applies as for divide decks and
continued pages. Each section of the fractional-measure table is, in effect, a
continued page, and the unit-indicators must appear again accordingly.
c. Field s p a n n e r s . - H e r e , again, t h e difference between t h e field spanner and
the boxhead of the lower deck (or decks) of t h e divide table must be kept closely
in mind.
(1) T h e presence of field spanners does not affect t h e range of influence of a
unit-indicator placed at t h e top of t h e column.
(2) By definition, at least two levels of field spanners must appear, if an
appear. T h e first level will appear below, but will not be a part of, t h e boxhead.
Here, t h e unit-indicator will appear at t h e top of t h e first d a t a line even though
this means it will appear below, a n d not above, t h e first level of field spanner.
In spite of this, its influence is maintained to t h e bottom of t h e table-page,


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEc.

14-C

T H E U N I T INDICATOR

PAR.

1423

irrespective of t h e presence of other field spanners c u t t i n g the column rules.
T h e only way to interrupt its influence is to insert a different unit-indicator in
t h e column.
(3) If t h e unit-indicator appears at the top of t h e column in a table with field
spanners, field spanners should not be used for t h e purpose of specifying unitindicator change below. Either t h e upper unit-indicator should be placed in a
field spanner, or t h e lower one should be removed from t h e field spanner a n d
run as a unit-indicator, as such.
(4) Where the field spanner includes or consists of a statement of unit, its
influence extends to t h e next field spanner or to the bottom of t h e table-page,
whichever appears first.

1424. Italics versus parentheses.-Where the material is in letterpress, the unit-indicator is set in italic and placed at the top of the
column. Unit-indicators are not enclosed in parentheses in letterpress
composition.
If the material is typed for offset and the unit-indicator is inserted
on the regular typewriter, italic is not available; therefore, parentheses
are used.
In each case the intent is to make the unit-indicator stand out in its
position at the top of the column.
1425. Capitalization.-Capitalization differs according to whether
the unit-indicator appears at the top of a column, as in par. 1422,
above; or whether it appears in a field spanner.
a. At top of column.-When at t h e top of a column, t h e unit-indicator is
given an initial cap. T h a t is, if it starts with a word, t h e starting word carries an
initial cap; succeeding words do not. This is t r u e in both letterpress and offset
material.
Example A-1. Right:
acres

Thous.
of dol.

55
32

7, 562
4, 138
7, 296

dollars

1,000
Acres

Thous.
of Dol.

754
629
385

55
32

7, 562
4, 138
7, 296

1,000

Dollars
754
629
385

Example A - 2 . Wrong:

b. In a field spanner.-If t h e sole purpose of t h e spanner is to specify t h e unitindicator, normal spanner rules apply; t h a t is, it should appear in lower case (This
is in lower case), not in caps and lower case (This Is in Caps and Lower Case), nor
in caps ( T H I S IS IN CAPS).
Where t h e unit-indicator appears parenthetically at t h e end of a field spanner,
it should be run in without a cap in both letterpress a n d offset.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Number
(thousands)

Value
(thousands
of dollars)

249

PAR.1426

sEc. 14-C

THE FIELD

1426. Style and abbreviation.-The style is as succinct as possible.
Include no more than that which is absolutely necessary to carry the
meaning. Normally, space is at a distinct premium. Abbreviate
freely, as long as the meaning is perfectly clear.
Certain liberties may be taken with expressions. Thus, "Thousands
of dollars" may be condensed to " T h o u s . dols.," "1,000 dollars" or
"1,000 dols.," as necessary. However, it is desirable that consistency
be maintained in such treatment, particularly within the same table.
Sec. 14-D. The Field Spanner (1431-1434)

1431. Definition.-The field spanner is a type of spanner appearing
recurrently within, and constituting a component part of, the field.
It is commonly confused with t h e complete boxhead which appears
over each deck of a divide table.
This confusion may be minimized if t h e field spanner is thought of
as a stub center head transposed into the field in cases wherea. No room is available to repeat t h e stub, and
b. T h e s t u b , if repeated, would be identical in every detail (except for t h e
center head represented by t h e field spanner) as t h e s t u b already shown.

Actually, the field spanner represents t h e uppermost spanner of a
boxhead, of which t h e remaining portion, identical with t h e boxhead
of t h e table, is not shown at each repetition. Thus, if t h e entire box
were repeated each time, t h e uppermost field spanner would be transposed to t h e top of t h e table to become t h e uppermost spanner of t h e
boxhead of the first deck, and t h e table would become a divide table.
Example A - 1 . Table with field spanners; compare with example A-2, divide table:

Country of birth

Total

Male

Female

Total

Male

The State

Urban

Rural-nonfarm

Rural-farm

All countries..

England_Scotland.-All other_-

All countries-

England.Scotland_All other_-

250

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Female

sEc. 14-D

THE

FIELD

PAR.

SPANNER

1431

Example A - 2 . Divide table; compare with example A-1, table with field spanners:

Urban

The State
Country of birth
Total

Male

Female

Total

Male

Female

All countries.-

England_-Scotland_-All other ..
Rural-nonfarm

Rural-farm

Country of birth
Total

Male

Female

Total

Male

Female

All countries ..

England_-Scotland_-All other __

1432. Preferred u s a g e . - T h e field spanner should be avoided except
where no other means is available for proper description of the data
concerned. Preferably, several conditions should be fulfilled before
its use is considered. (See also par. 1432c (2), below.)
a. T h e table should be one consisting of a series of blocks of data arranged
both in decks and panels, withb. All data in each deck having a common stub, but requiring a different center
head; a n d with all data in each panel having a common box head but requiring a
different uppermost spanner; a n d c. It is impossible or undesirable to repeat t h e stub for each panel ( fractionalmeasure table, or the entire box for each deck (narrow-divide table), because(1) Not enough room is available, or
(2) Each deck is so shallow, and each panel has so few columns, as to demand
too much space for repetition of t h e complete stub, or of t h e complete boxhead.

1433. Problems.-The following instances represent common problems and suggested solutions:
a. Where stub can be repeated in one-half, or one-third, measure without too
great sacrifice in space.-Here, shift the field-spanner statement to a center head
in t h e repeated stub.
b. Where the entire box can be repeated for each deck without too great sacriflee in s p a c e . - I n this case, repeat t h e box, convert each field spanner to an uppermost spanner, and thereby make t h e table a divide table.
c. Where the field spanners extend across the entire body of the table.-Here,
t h e suggested procedure varies according to t h e nature of t h e head statement.
(1) S t a t e m e n t of universe.-Place it as a center head in t h e stub if it can
be condensed into not more t h a n three lines. Otherwise, use t h e field spanner.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

251

PAR. 1433

sEc. 14-D

THE FIELD

(2) U n i t - i n d i c a t o r . - I f t h e statement specifies some t y p e of derived
measure, such as percent, median, average, etc., it usually will be satisfactory
to transpose it to t h e s t u b as a center head if t h e rest of the table is devoted to
frequencies. However, if a statement such as "millions of dollars," "thousands
of short tons," etc., is involved, it usually is wiser to leave it as a field spanner.

1434. Incorrect placement of field spanners.-When field spanners
are used, it is important that four common errors be avoided.
a. Upper and lower rules extending through s t u b . - E v e n where t h e spanner
head applies to the entire field, t h e upper and lower spanner rules should never
extend across t h e stub.
Example A - 1 . Wrong:

Year

Average

Jan.

Feb.

Mar.

Apr.

May

June

Total operating revenues (millions of dollars)
1918-1919--

410. 5
432. 0

285. 4
397. 2

290. 0
352. 4

336. 4
377. 4

371. 6
389. 5

379. 0
413. 9

395. 2
426. 1

38, 477
37, 307

Freight ton-miles (millions)
1927.
1928--

39, 559
39, 769

39, 233
36, 289

37, 259
35, 723

41,817
39, 486

37, 111
35,887

40, 116
39, 263

Jan.

Feb.

Mar.

Apr.

May

Example A-2. Right:
Year

Average

June

Total operating revenues (millions of dollars)
1918-1919--

410. 5
432. 0

285.4
397. 2

290. 0
352. 4

366. 4
377.4

371.6
389. 5

379. 0
413. 9

395. 2
426. 1

40, 116
39, 263

38, 477
37,307

Freight ton-miles (millions)
1927--1928

39, 559
39, 769

39, 233
36, 289

37, 259
35, 723

41,817
39, 486

37,111
35, 887

b. Head for first deck placed as an uppermost spanner of table boxhead.-This
is incorrect because(1) Any head appearing at t h e top of, and as a part of, t h e boxhead applies
to all decks down to t h e bottom of t h e table, or until t h e entire boxhead is
repeated (or a new boxhead presented) as in a divide table.
(2) T h e basic idea in using t h e field spanner is t h a t t h e table boxhead applies
to all decks; the variation is confined to t h e affected portion of t h e field. Where
t h e first level of field spanner is placed as an uppermost spanner of t h e boxhead,
it means that all of t h e boxhead applies to t h e first deck, but only a portion
of it is supposed to apply to all succeeding decks. This inconsistency is confusing to t h e user.
(3) T h e first deck is thereby deprived of a head parallel with those shown
for all succeeding decks.

252

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

sEc. 14-D

THE

FIELD S P A N N E R

PAR.

1434

Example B-1. Right:
Age

Total

Male

Female

Total

Male

The State

Urban

Rural-nonfarm

Rural-farm

T h e State

Urban

Female

Example B-2. Wrong:

Age
Male

Total

Female

Total

Rural-nonfarm

Male

Female

Rural-farm

c. Head for first deck placed as undercut spanner of table boxhead.-This
procedure is incorrect for the same reasons as stated in the preceding case. Either
all of the boxhead should apply to all decks, or none of it should. If all of it is to
apply, the field spanner for the first deck should not be a part of the boxhead. If
none of t h e boxhead is to apply to other decks, then the entire boxhead should be
repeated for each deck and the table made a divide table. However, where the
table is extremely tight, this technique is acceptable, though undesirable, if the
space saved makes it possible to squeeze the table on the page. In example C,
below, note t h a t no space is saved. However, if the s t u b box contained a 3-line
s t u b head, space-saving would result. In contrast, the procedure shown in example B - 2 is never acceptable.
Example C. Wrong (for "right" version, see example B-1, above):
Total

Male

Female

Total

Male

Female

Age


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

T h e State

Urban

Rural-nonfarm

Rural-farm

253

PAR.

1434

THE

sEc. 14-D

FIELD

d. 0 mission of head for first d e c k . - T h e usual case is where the-first deck represents a summary section of the table. Here, there is a tendency to assume that
the table title will carry the story, or t h a t the reader will take the summary
nature of the first deck for granted. In consequence, the field spanner is erroneously omitted for the first deck, even though such a head appears for all sueceeding decks. This is incorrect; if any deck carries a field spanner, then all
should carry one.
Example D-1. Wrong:
A G E OF T H E W H I T E P O P U L A T I O N , F O R C I T I E S OF 25,000 I N H A B I T A N T S OR
M O R E , BY C I T Y S i Z E - G R O U P S : 1920 To 1940

Census year

All ages U n d e r 5 5 to 14 15 to 19 20 to 24 25 to 29 30 to 34 35 to 44

1940
1930--1920

45 a n d
over

(What is t h e area covered by this first deck?)

In cities of 100,000 inhabitants or more
1940
1930-1920.
In cities of 25,000 to 100,000 inhabitants
1940
1930
1920

Example D-2. Right:
A G E OF T H E W H I T E P O P U L A T I O N , F O R C I T I E S OF 25,00) I N H A B I T A N T S OR
M O R E , BY C I T Y S 1 z E - G R O U P S : 1920 To 1940

Census year

Under
ages

5 to

15 to
19

20 to

25 to

30 to
34

Total
19401930--1920

In cities of 100,000 inhabitants or more
1940-1930--1920

In cities of 25,000 to 100,000 inhabitants
1940-1930--1920

254

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

35 to
44

45 a n d
over

sEc. 14-E

PAR.

BRACES

1441

Sec. 14-E. Braces (1441-1448)

1441. Definition.-The brace is a device for indicating the relationship of a given line to a group of lines, or of a smaller group to a larger
group, and vice versa. Specifically, it provides a means of indicating
the funnelling together of two or more data lines into fewer combined
lines and a subsequent reexpansion to a greater number of lines.
Braces should be avowed where possible.
a. Types of braces.-Two types of braces are used; they may be referred to as t h e
contracting brace (}) and the expanding brace ({). These terms are descriptive
of their function as t h e table is read from left to right.
Example:
Area A
Area B

646
832

990

114, 007
269, 324

1442. Principle of placement.-The following rules apply:
a. Point toward smaller g r o u p . - B r a c e s should be placed so t h a t t h e point is
toward t h e fewer number of entries and t h e spreaders embrace t h e larger number
of entries.
Example A-2. Wrong:

Example A-1. Right:

A
C

Example B-1. Right:
A
B

b
C

Example B-2. Wrong:

b

B

C

Example C-1. Right:
A

b

Example C-2. Wrong:
B
0

a
b

b. Placed at the right of column r u l e . - B r a c e s should be placed at the right of
t h e intervening column rule when t h e entries affected are in different columns.
Example D-1. Right:
White_
Nonwhite_--

832

269, 324
114, 007

Example D-2. Wrong (braces on wrong side of column rule):
White_
Nonwhite __

832
646

oo{

269, 324
114, 007

1443. Stub listing determined by maximum lines needed.-Where
braces are used, the stub listing must account for all lines on which
data are shown separately in any column. The stub should not be
restricted to the captions necessary for the entries which appear in the
first column. Otherwise, where the detail is expanded, the additional
lines may lack stub captions.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

255

PAR.

1443

sEc. 14-E

THE FIELD
Example A - 1 . Right (expansion lines explained):
White_
Nonwhite _

224

172
163

196
129

Example A - 2 . Wrong (expansion lines not explained):

All races_-

224

172
163

196

129

1444. Braces rarely appear singly.-Braces usually, though not
always, appear in groups of two or more, although not within the
same column. The only instance where a single brace (always a
contracting brace) appears on a given line within the field is where its
influence extends to the right-hand edge of the table-page (or to the
next repetition of stub in a fractional-measure table). Here, there is
no need for an expansion brace to appear.
1445. The contracting brace ( } ) . - T h i s is used wherever the figures
for two or more consecutive stub entries are available only in combination.
Example:
Hand sprayersPower sprayers __

3, 024

2, 123

a. It is placed with its spreaders pointing to t h e left and embracing t h e group
of lines for which t h e combination entry is to be shown (including overrun lines).
I t s point is directed toward t h e single cell entry on t h e right into which t h e preceding lines are being funnelled.
b. T h e spreaders should encompass t h e space occupied by t h e s t u b line-captions which describe t h e entries preceding t h e brace, not merely t h e space occupied
by t h e entries themselves. If the line captions overrun and t h e cell entries are
shown against t h e top line (see par. 1014a), t h e braces should be spread to include
t h e overruns clear across t h e table.
c. Cells which cease to be applicable because of t h e brace are considered to be
braced out of existence. They should always be left blank. They should not
carry d a t a , leaders, hyphens, or footnote references.
d. A footnote reference relating to the combination entry, or relating to any
one of t h e component lines thus consolidated, is placed against the combination
e n t r y . It is not inserted in t h e cells above or below it which have been braced out.
e. T h e influence of the brace extends to t h e right u n t i l an expansion brace is
encountered; until the edge of t h e table page is reached; or, in a fractional-measure
table, u n t i l a new stub is reached.

1446. The expanding brace ( { ) . - T h i s is used against figures in
those cases where a combination line created by a contracting brace
is reexpanded to two or more (but not necessarily all) of the original
stub lines.
Example:
Hand sprayers.Power sprayers..

256


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

3, 024

4, 146
2, 739

SEC.

14-E

PAR.

BRACES

1446

a. It is placed with its spreaders pointing to the right, embracing the group of
lines for which the entries are to be again shown separately. I t s point is directed
toward the single cell entry on t h e left which is being expanded.
b. T h e spreaders invariably encompass the space occupied by the line captions
(in the stub) which describe the entries t h a t follow the brace. Their spread is
not limited to the space occupied by the following entries, as such.
c. T h e line caption which appears on t h e given line in the s t u b automatically
applies to each row of entries thus again shown separately.
d. When a group of lines is braced to a combination entry, no line should be
included within the spreaders where the separate entry is known to be zero or
where no entry is applicable. Where it seems to be essential that the brace
include such a line, the combination entry should be footnoted to give the known
information unless the purpose of the brace is to prevent disclosure; that is, to
prevent the reader from identifying the specific person or firm to which the
statistics relate.

1447. Range of influence.-The influence of a brace is never interrupted by a column rule, any more than the range of a stub caption is
interrupted.
T h e sole exception to this is where a new series of stub captions
appears (new stub) as in a fractional-measure table. In this case, new
braces must be inserted, even though the new stub is identical with
the old, and even though identical line captions happen to line up in
the two (or more) stubs.
1448. Braces not used against column heads. Except in rare
instances, braces should not be used to consolidate entries for adjoining columns. T h e use of braces is best restricted to consolidation of
data appearing in lines. Where a consolidated entry is necessary for
adjoining columns, handle as follows:
a. Preferred m e t h o d . - P l a c e t h e entry in one column; footnote both cells.
T h e entry should normally be placed in t h e cell in which t h e major portion of t h e
combined entry would appear if available separately. Footnote t h e combined
entry; also insert the same footnote reference in t h e adjoining cell which has thereby
been deprived of its entry. This method is most suitable where only one or two
lines are involved.
Where a large number of lines are affected, particularly a consecutive block of
lines, t h e footnote might well be placed against the column heads, reading, for
example, "Figures for other races included with those for Negro for 1870 and
earlier years.
Example A-2. Undesirable:

Example A-1. Preferred:
Census year
19401900-1890-1880
1870.

White
12, 516
9, 248
6, 543
4, 123
1, 234

Negro

1
1

8, 742
5, 219
2, 824
1, 131
600

Other
races

White

651
384

12, 516
9, 248

100

6, 543
4, 123
1, 234

1

Other
races

Negro
8,742
5, 219

651
384

2, 824
1,131
600

100

Statistics for " O t h e r races" included with Negro

for 1890 and 1880.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

257

PAR.1448

sEc. 14-E

THE FIELD

b. Acceptable m e t h o d . - W h e r e an even number of columns is involved, as in
example A-1, break t h e column rules separating t h e columns. Then center t h e
combined entry (without a brace) under the group of columns. T h e breaking of
column rules is troublesome to t h e printer and therefore is costly where used to
excess.
This method is not practicable where an odd number of columns is involved
since t h e combined entry would fall directly into t h e middle column. Perhaps
t h e greatest justification for horizontal braces is where t h e entries in an odd number of columns must be consolidated to a single entry.
Example B-1. Acceptable, but troublesome to printer:
White

Census year

12, 516
9, 248
6, 543
5, 123

1940
1900
1890
1880
1870-

Example B-2. Wrong (if last two
entries are for all groups combined):

Other
races

Negro
8, 742
5, 219
3, 824
2, 131
1,600

B
742
493
508

1, 651
1, 384
1, 100

Example B-3. Acceptahle (troublesome to printer.
Also, note effect on caption spacing in stub):
Census year
1940
1900

White
12, 516
9, 248

1890
1880
1870-

258

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Negro
8, 742
5, 219

Other
races
1,651
1, 384

10,367
7, 254
4, 234

1, 600

1, 100

632
701
946
356
748

C
849
275
361

INDEX
Paragraph
Paragraph
Boxhead _
1209 Boxhead (see also individual component parts):
1209
Abbreviations used_-963
" D o . " not used for "ditto" __
701-713
Allocation as compared with stub_662
Footnote, use in_Banner head-_ 1202, 1261-1266
523
Headnote, use in_1210
Capitalization__ 347, 1109, 1114
States.
Captions running up __
1208
1426
Unit-indicator _
Captions should run across, not up __
1207
945
Additive lines _
1206
Colon head not u s e d 958
"All" vs. " T o t a l "
_ 1202, 1221-1233
Column head __
Allocation to s t u b a n d boxhead:
Component parts _
204, 1202
General relationship _
701
218
Continuous_-702
Influence of table purpose_712
Mechanical advantages of each_711
Principles, s u m m a r y _
Dates, use oL _
365
Relationship between tables in a series_703
Definition _
204, 1201
Space requirements of each_713
639
Footnote numbering sequence_Heads in field, not in box _
1203
Analytical space break (see also Reader break
_ 1211, 1313-1314
Horizontal rules_-and Space break):
1204
Levels of box_
1023
712
Definition_-Mechanical advantages__Insertion __
1024
711
Presentation principles_.Analytical table __
132
1241-1248
Read-in heads_" A n d , " use in time-reference segment of title __ 367
Space problem __
1205
Appendix tables, numbering of_423
Space requirements _
713
Arabic numerals (see also Asterisk, Capital
Spacing of captions __
1211
letters, Small letters, Table number, etc.):
Spanner head _
1202, 1251-1255
Reference symbols, use as _
633
Style and punctuation _
1209
_ 1202, 1271-1272
Table number, use a s -402-403
Undercut s p a n n e r - Area or space segment of title:
351 Braces:
" B y , " use of._347
Cities_1444
Appear singly only in rare instances_C i t y size-groups_348
1414
Cell entry, use with_Continental United States _
344
1445
Contracting brace_Counties_350-352
_ 1402, 1441
Definition_-.
_ 303,341
Definitions_1446
Expanding brace_1448
Not used against column h e a d s - " F o r , " use of.-351
Placement _
1442
350
Omission a n d curtailment_Range of influence _
1447
Placement_.342
1443
Stub, effect on _
Regions and divisions.
346 Brackets, use in headnote __
202, 523
"Selected" or "Specified," use oL _
352
States __
345 Broad table:
Terminology_
343
1208
Captions never run u p Territories a n d possessions _
344
Defined a n d described _
219
United States_
344
219
Placement on page_349 Broad-divide table.
Urban and rural_-220
Areas (see also Area segment of title, Cities,
216
Broad-measure defined __
Regions and divisions, etc.):
By
Distinction in banner head of standard
351
Cross-classifying term_1262
tables_1253
Spanner head, use in __
Spelling of names _
1117-1118
_ 336,351
Title usage_
Standard s t u b listing _
832, 1101-1116
351
Versus "For"_Asterisk:
Headnote explanation_
513 Capital letters (see also Arabic numerals,
Placement of footnote bearing s y m b o l . 656
Small letters, T a b l e number, etc.):
Reference symbol, use a s 633, 640
Reference symbols, use as __
633
Table
number, use as _
402-403
Banner head (see also Spanner head):
Capitalization:
Acceptable uses __
1262
Boxhead _
1210
Definition __
_ 1202, 1261
915-916
Center head in s t u b .
Improper uses_12G3
Column
head
or
caption
__
1226
Quasi-banner.1264-1266
523
Headnote_-Bearoff:
946
Line
caption,
d
a
t
a
line.
Definition_-1324
959
Line
caption,
total
and
subtotal_.In boxheads _
1211
Titles_
308
Blank cell:
1425
Unit-indicator_-Reference-symbol placement_-637
Types of entries_1413
Cell:
Block:
Braces in cells __
Definition_-203
Definition_
_ 205, 1402, 1411
Levels assigned in stub __
916
Function and purpose __
1412
Rules (horizontal), use of_-1313-1314
Grand-total cell__ 956, 1233
" R e a d In" rarely made to cell entries_--1244
Boldface t y p e (see also Italic t y p e and Type
Reference range of symbols.
641-642
specification):
Reference symbol placement ..
637
Horizontal rule, use w i t h _
1312
Types of entries_
1413
Line caption, total a n d subtotaL _
959
Not used to indicate coordinate leveL _
1032 Census d a t e s (see also Dates):
Boxhead, use in __
365
Stub, use in
1032
S t u b listing
-1121-1123
Table number, use i n 308
Boxed-in tables.
226, 1334
Time-reference segment of title, use i n . - 361-368
Abbreviations:

776554-49--18


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

259

INDEX
Paragraph
211
Census-size page, specification_1115
Census tracts, stub listing __
Center head (see also Subhead):
913
Comparison with colon and dash line_-_ 203, 812, 911
Definition
918-920
Level combinations_-915
Levels of center heads_916
Level distinction_
917
Omission in special cases_912
Purpose and function _
Range of influence _
914
955
Shifted to total lines _

Paragraph
Column head or caption-Contlnaed
Unit-indicator_Word order.-Column head, total and subtotal:
Definition.
Grand-total celL
Meanings of term "total" __
Spanners covering_
Column number, definition_-

1225
1224

1231
1233
1232
1254
1202

Combination of head levels. See Center head.
Comma (see also Colon, Dash, Punctuation,
Semicolon, etc.). use in time-reference seg363
ment of title_Common footnote or reference number. See
Recurrent references.
133
Complex tables__ 212,225
Composite forms of tables __
335
Composition table titles.
Contents, tables of (see also Leader work and
414
Tables of contents), tabular form
Continental United States. See United States.
819
Continued line, definition __

Citations of source. See Source note.
Cities (see also City size-groups and Metropolitan districts):
Area·segment of title, use in__ 347,350
·Independent cities _
1108
1109
Listing under State h e a d s 1111
Specification of area designation_1109
State name imperative, special cases_Stub listing _
1109
City size-groups (see also Cities and Metropol!tan districts):
348 Coordinate levels:
Area segment of title, use i n 1032
1110
Stub listing__Boldface type not used __
1002
Influence on indent_.903 Counties:
•"Class," use in stubhead_350-352
Classification level:
Area segment of title.-1002
1117
DefinitionSpelling of n a m e s - Influence on indent___
1002
1108
Stub listing _
Classification segment of title:
Cross-classification:
1253
336
"By," use of_-Boxhead, use of " b y " - 351
"Composition" table title used only once_335
" B y " versus "For"
Cross-classification:
902
Stubhead, indication i n - Limiting explanation__
333,
334,
336,
351
336
Title usage_Splitting of segment_333 Cross-references:
652
334
Variation in description_Footnotes at end of table_
512
303, 331
DefinitionHeadnote-Placement _
909
332
Stubhead or box __
Wording_307
Classification term substituted for " T o t a l " _
643 Dash, punctuation mark (see also Colon,
Comma, Hyphen, Punctuation, etc.):
•Codes as cell entries_930
1413
Confusion of internal with terminal dash
Coextensive counties and cities_
313
1108
Key-phrase title, use in
1114
Colon (see also Punctuation, and individual
Metropolitan district names, use i n marks):
926-927
"Read-in" significance_401
112
Introduces leader w o r k With table number_-121 Dash (or read-in) head, boxhead:
Introduces text tabulation_Key-phrase title, use in _
1243
313
Avoidance of misunderstanding_
1245
Time-reference segment of title, use in __
363
Avoided over total columns
1242
Colon beads, boxhead:
Clarifies paired distributions __
1241
Colon column beads not used_-1206
Definition_-Single subentry _
1248 1
1206
Incorrect u s e 1247
Colon line:
"Reading in" to terminal groups_1244
Comparison with center head _
913
"Read in" rarely made to cell entries1246
923
Comparison with dash line_Residuals exempted __
Definition
813, 922
1017 Dash (or read-in) line, stub:
Leaders not u s e d 913
Comparison with center head _
Single subentry__924
923
Comparison with colon line_1027
Space break insertion above, not b e l o w 931
Consecutive read-in lines permissible_917
Substituted for center head_Uses _
Dash, confusion of internal with terminal.- 930
925
813, 926
Definition _
1017
Leaders not u s e d Column (see also Boxhead, Column bead, and Field):
913
Not
a
heading
_
Definition__2()5, 1402
927
"Read in" defined _
Space allocation1324
928
Residual
subentries
__
Vertical rule, use of..
1327-1328
029
Single subentry_
Column head or caption (see also Boxhead,
1027
Space break insertion above, not below_Column):
932
Space
requirements
of
stub_1209
Abbreviations used_933
Test of read-in application.-1223
Additive and nonadditive columns_
1302
1448 "Dashes" versus "leaders"
Braces not used against column h e a d s 1302
"Dashes"
versus
"rules"
(see
also
Rules)
__
_ 1210, 1226
Capitalization
1206
Colon head not used_204, 1202, 1221 Data line (see also Ltne caption):
Definition942-943
Classes and types __
Limitations. __
1204
941
Definition_-1222
Prime and nonprime columns_948
T y p e specification _
641-642
Reference range of symbols966
637 Date column_
Reference symbol placement _
1227 Dates (see also Census dates, Time-reference
Rule (vertical), use of--segment of title, etc.):
365
Boxhead, use in __
1225
Single level_-1262
Banner head_
Space problem caused by wordiness __
1205
637
Reference symbol placement_1211
Spacing_833
Standard stub arrangement_
1209
Style and punctuation_-

260

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

INDEX
Paragraph
Definitions:
Additive data-lines_945
Analytical space break_
1023
Arca or space segment of title_-303,341
Banner head_
_ 1202, 1261
Bear-off__
1324
Block_
203
Boxhead_204, 1201
Braces_-. 1402, 1441
Broad table __
219
Broad-divide table_220
Cell.. _
205, 1402, 1411
Center bead or subhead_
203, 812, 011
Classification level_1002
Classification segment of title_-- 303,331
Colon line_-813,922
Column_
205, 1402
Column bead or caption __
204, 1202, 1221
Column head, total and subtotal.1231
Column number ..
1202
Continued line..
819
Dash (or read-in) bead __
1241
Dash (or read-in) line__ 813,926
Divide.211
Divide table. . .
211
Expansion table (described)_
434, 435
Field_-205, 1401
Field spanner ..
- 1203, 1402, 1431
Folio line_-208
Footnote.
206, 603, 605
Formal table..
131
Fractional-measure tables_215
General footnote_-605
Group or line number ..
821
Heading _
202
Headnote-. 202,501
Horizontal rule_1311
Indention1001
Keyed item (footnote identification).603
Leader work ..
111
Leaders:
General..
1011
Cell leaders.
1413
Stub leaders- 817, 1013
Line. . .
- 205, 1402
Line caption _
203, 941
Line caption, data line..
. 814,941
_ 815,951
Line caption, total and subtotal..
Master table (described) _
432
Measure. . .
211
Multipage box or stub.
1316
Multipage narrow table with continuous
boxhead __
218
• Narrow table._
216
Narrow-divide table. __
217
Nonadditive data-lines_945
Nonprime data-lines__942
One-page box or stub ..
1316
_ 1010, 1014
Overrun-Panel...
204, 1202
Parallel rule ..
_ 1312, 1322
Parallel table:
Normal type_221
Special type_222
Parallel-divide table.-223
Prefix l e t t e r s 421
Prime lines_-942
Quasi-banner head . . .
1202, 1264
Reader break_-1023
Reader ceJJ..
637
Reader column ..
1014
Reader entry __
637
"Read in"
927
Read-in head
1241
Read-in line . . .
813
Reem-rent reference. . .
503
Reference symbol...
631
Residual i t e m 1143
Rules, horizontal and verticaL.
1301
Source note _
207, 604, 621
Space break_
818, 1021
Spanner head _
. 204, 1202, 1251
Specific footnote _
603
• Standard-series tables __
411
505
Standard table forms_-


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Paragraph
Definitions-Continued
211
Structural types of tables_203,802
StubStub leaders . . .
817, 1013
Stubhead or box. __
. 203, 811, 901, 1202
Suffix letters _
431
Superior figures__633
Table number __
_ 202,401
Table title _
301
Tabular presentation __
101
Text tabulation_121
Time-reference segment of title_. 303,361
Title __
202
Totality table (described)_434
820
Tracer numbers.Undercut spanner __
1202, 1271
. 816,961, 1203, 1402, 1421
Unit-indicator.-Universe segment of title.
. 303,321
1321
Vertical r u l e Ditto, not used by Census Bureau_Divide, definition_Divide table (see also individual types of tables):
Boxhead rules, use o f Unit-indicator, use of_-Divisions, geographic (see also Regions and
divisions):
Listing with States_-Listing with regions _
"Do," term not used by Census Bureau _
Document-size page, specification.-"During" in time-reference segment of title _

963
211
1314
1423
1105
1103
211
367

Field (see also individual component parts):
. 1402, 1441-1448
Braces_._ 1402, 1411-1414
Cell_-_ 205, 1402
Component parts__205, 1401
Definition-_ 1402, 1431-1434
Field spanner ..
_ 1402, 1421-1426
Unit-indicator._Field spanner:
_ 1203, 1402, 1431
Definition_-_ 1313, 1314
Horizontal rule, use o f _
216
In narrow table.. _
Incorrect placement __
1434
1203
Part of field, not of box _
1432
Preferred usage-Problems.. _
1433
1425
Unit-indicator, capitalization of.-1423
Unit-indicator, use o f - . 201,208
Folio line. . .
Folio number. See Page number.
Footnote:
206,603
Definition_-_ 601, 605-606
General footnote _
_ 503, 612-613
Headnote relationship_Numbering sequence __
638-639
615
Omission and curtailment.. 523,662
Period always used at end_651-656
Placement and arrangement_615
Practices to avoid __
Presentation of footnote-type of statement
improper in stubhead _
908
Purpose__
611

Reference range of symbols_Reference symbols...
Restrictions on use_Scope of discussion_.
Source note_-Specific footnote
Style_-Tabular footnote.Title relationship _
To leader work_-To title. . .
Types_-Wording_-

641-643
631-640
. 606,614
602
. 601,604, 621-624
. 601,603, 611-615
662
665
613
113
. 502,613
601
661-665

Footnote numbering*(see also Reference symbols):
113
Leader work.
636
Multipage tables.
635
New series for each table_
638-639
Numbering sequence_Placement of symbol.
637

261

INDEX
Paragraph Indention:
Paragraph
Footnote numbering-Continued
Comparative i n d e n t 1003
Reference range of symbols__ 641-643
Definition_-1001
Indent spacing __
Symbols--633-634
1009
Total rells, columns, and lines
642
1010
OverrunsFootnote placement (see also Footnote num1002
Purpose and function_-1004
bering and Reference symbols):
"Reverse" versus "flush" for totals_-Selection of method _
1007
Leader work
113
1008
Location of footnotes
- 651-656
Spacing (vertical) effect on scheme_Titles_
Text and tables on same page_113
308
Total lines _
1004-1007
Footnote symbols. See Reference symbols.
Italic type (see also Boldface, Type specificaFor
tion, etc.):
351
Title usage_
1033
Deduct or decrease items_Versus " B y " __
351
953
Selected items of a distribution_Stub, use in __
Formal table:
1033
132
Analytical types_1203, 1424
Unit-indicator_-1313, 1315-1316 "Item" in stubhead_Bottom rule, use Of.903
Class of tabular presentation __
102
Complex and simple tables __
133 Leader work:
131
Definition_-111
Definition_-General purpose and special purpose table_-- 132
113
Footnotes to.-Heading (see also Table number, Title, and
112
General features __
Headnote)_
202
114
Indexes_-Parts (structural) of table (see also individual
112
Introduced by a colon_201-208
parts)
114
Tables of contents_Types (structural) of tables (see also indiLeaders:
-211-226
idual types)941
All leader lines __
_ 125,606
Without title_-1413
Cell entry_-1017
Colon or dash lines never take leaders __
Fourth-measure table. See Fractional-meas_ 817, 1011, 1013
Definition_-ure tables.
1325
Excessive use_-Fractional-measure tables:
1016
Flush to caption _
215
Defined and described_-941
Follow data-line captions_-1330
Parallel rule, use oL _
1015
Omission of period _
1423
Unit-indicator, use of_1014
Overrun item_-664
Fractions in footnotes_- 1011, 1013
Periods used in Offset_-Frequency tables:
1012
Purpose and function_-434-435
Numbering, using suffix letters __
817
Stub leaders_Title, when appearing in same table with
963
Unit-indicator in stub, use with_percent data _
372
Title, when appending percentage descrip1204
tion __
374 Levels of boxhead, limitations_-Levels of center heads. See Center head.
_ 205, 1402
Line, definition_-General footnote (see also Footnote), defined
and described
605-606 Line caption:
941-945
Classes and types._132
General purpose table_
_ 203,941
Definition __
Group or line numbers, defined and described_ 821
203
Part of stub_947
Punctuation_-Hairline rules. See Rule, horizontal, and
Line caption, data line:
Rule, vertical.
945
Additive and nonadditive lines __
Half-measure table. See Fractional-measure
946
Capitalization_tables.
814,
941
Definition
_
Heading of table (see also individual component
942-944
Prime and non prime lines __
parts), defined and described_202
947
Punctuation_-Headnote:
948
Type specification._As footnote to title
502
Component of headin
202
c o n i n u c d pases o r ' S i a
505 Line caption, total and subtotal:
1116
Area names, use oL _
202,501
Definition_-959
Footnote relationship __
- 503, 612-613
Capitalization and type specification_815,
951
Definition_501
Function_1005-1006
221
Parallel table_-Indention-952-955
523
Meanings of term " T o t a l "
Period never used at endPlacement of "Total" in combination with
Practires to avoid
506
957
Preprinted table forms __
other wording_
505
Presentation of headnote-type of statement
Reader-break insertion above, not below_- 1027
959, 1313-1314
improper in stubhead __
Rules (horizontal), use of _
908
958
"Total" versus "All" __
Uses _
956
Repetition, multipage table_505
821
Line
numbers
compared
with
tracer
numbers.
Restrictions on use_504
Specific uses_-511-513
432
Master table, described _
Style, including capitalization and punctua211
tion.
523 Measure, definition_-1114
Tables without titles_
606 Metropolitan a r e a s 1114
Time factor, specification of_364 Metropolitan districts, stub listing __
Title relationship _
1112
- 502,613 Minor civil divisions, stub listing _
1316
Use to curtail universe segment of title_
323 Multipage box or stub, definitionUse to qualify classification segment of title_ 334 Multipage narrow table with continuous boxWording_
-521-523
head:
Compared with parallel table __
218, 221
218
Horizontal rule. See Rule, horizontal.
Defined and described_-Hyphen:
Multipage table:
In column heads __
1316
1204
Bottom rule, use of_Metropolitan district names, use in _
1114
651, 655
Footnote placement_Rural-nonfarm and rural-farm, use in __
1113
636
Footnote treatmentTime-reference segment of title, use in ..
367
505
Headnote repetition_-

262

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

INDEX
Paragraph Period-Continued
Paragraph
Narrow table:
As the normal table type.
214
Headnote, not used at end_-523
Defined and described.-216
Omission before leaders _
1015
Deviations caused by space requirements
214
Overruns_947, 1014
Multipage, with continuous boxhead. See
Substitute for leaders in offset__ 1011, 1013
Multipage narrow.
Narrow-divide table:
Placement:
Compared with narrow table with field
Area or spaco segment of title. . .
342
spanners_217
Braces_-1442
Defined and described . . .
217
Broad table on page__
219
Narrow-measure, defined__
216
Classification segment of title_-332
Normal table. See Narrow table.
Footnotes_651-656
64
Note
Leaders, in stub with unit-indicator_963
Footnote usage__ 206, 605-606, 615
Reference symbols_-637
Headnote, usage avoided in .. _
Space breaks_506
_ 1024-1030
Number. See Numbering schemes, Table
Time-reference segment of title_362
number, Prefix letters, Suffix letters, etc.
"Total" in combination with other wording_
"Number" omitted as center head _
917
957
Unit-indicator in field.1422
Numbering schemes, footnotes:
Unit-indicator in stub.
963
639
Universe segment of title_Boxheads-322
Normal scheme for tables_-635
Symbols used.
633-634 Plural form versus singular in description . . .
307
To leader work_. 113,635 Prefix letters (see also Suffix letters, Table
number, etc.):
Numbering schemes, tables:
Census tracts, use in
Normal scheme_-406
1115
Prefix letters:
Definition_-421
General use.
Standard-series tables ..
422
Standard-series tables.
Text and appendix tables_-424
423
Text and appendix tables. __
Series of tables_-407
423
Standard-series tables_--411-414 Presentation unit. See Unit of presentation.
Suffix letters_-942
434-435 Prime lines_-Symbols used ..
402-403
Text and tabular section-407 Punctuation (see also Comma, Colon, Dash,
Period, Semicolon, etc.):
Numbers in footnotes_-664
Boxhead ..
1209
Headnote.
"Of," use in spanner head._
1253
523
Leader work.
One-page box or stub, definition_1316
112
Line caption, data line_"Other" versus "All other"_
1144
947
Text tabulation_Overruns:
122
Titles (Time-reference segment).
Definition_-. 1010, 1014
363
Indention_1010
Placement of cell entries with tracer numbers. 1014 Qualitative and quantitative classifications,
stub arrangement and listing
834-835, 1131-1133
Reader-break insertion.
1029
1264-1266
Tables with tracer numbers__ 947, 1014 Quasi-banner head_Tables with reader columns_1014
Unit-indicators_-963 Reader break (see also Analytical space break
and Space break):
Counting off lines, overruns and totals .. 1028-1029
Page number as a part of folio line.. _
208
Definition_1023
Page size, influence on table design_211
Insertion, coordinate and noncoordinate
Page sizes of Bureau of Census.211
items_1026-1027
Panel:
Principles of insertion_1025
Definition_
204, 1202
Vertical rule, use of_1326-1329 Reader cell:
Definition_-Parallel rule (see also Rule, horizontal, and
637
Reference symbol placementRule, vertical):
637
Definition_-_ 1312, 1322 Reader columns_1014
Drafting specification _
637
. 1314, 1323 Reader entry __
Horizontal rule, use as _
1314
Vertical rule, use as
1330 "Read in
Definition_927
Test of.Parallel table (see also Tracer nufnbers):
933
Cell entries for overrun captions_947 Read-in head, boxhead__ 1241-1248
Read-in line, stub_Cell entries, placement with caption over926-933
runs.
1014 Recurrent references:
Definition_Compared with multi page narrow table_- 218,221
503
Failure to footnote at all points of recurrence. 615
Headnote arrangement_
. 218,221
Footnote arrangement_Normal type, defined and described ..
655
221
Footnoting and symbol usage_Paging__ 213,221
640
Headnote explanation_Special type, defined and described_-222
513
Stub placement.
. 218, 221-222 Reference range:
Symbols_-Title arrangement ..
641-643
218,221
Unit-indicator_-Tracer numbers in stub ..
218, 221-222
964
Tracer numhers versus group or line numhers, use of_
820 Reference symbols (see also Footnote, Asterisk,
etc.):
Parallel-divide table
223
Cell entry_.
1413
Parentheses:
Definition_-631
Headnote, use in_523
Function and purpose_-632
Unit-indicator, use with __
1424
Letterpress versus typing__634
Parts of table. Su Structural parts of table
Multipage table, use in . . .
636
and individual parts by name.
New series for each table_635
Percent data, description in title
_ 371-374
Numbering sequence_638-639
Percent tables, numbering oL __
434-435
Placement.-637
Period:
Range of s y m b o l s 641-643
Abbreviation at end of line caption ..
947
Recurrent reference, use in.
Boxhead, not used in ..
1209
Types of symbols.-Footnote, always ends with period.
_ 523,662 Reference table. See General purpose table.


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

263

INDEX
Regions and divisions:
Area segment of title, use in_Standard stub arrangement_Stub listing_
Repository tables _
Residuals:
Dash head_-Dash line, use with _
Definition ..
Maintaining identity _
Stub listings . . .
Value varies__

Paragraph

Paragraph

- 3 4 6 , 350 Space break (see also Analytical space break
832
and Reader break):
1023
. 1103-1106
Analytical versus reader breaks_-. 818, 1021
132
Definition_-1022
Purpose and function .
1005
1245-1247
Reverse indent, used in .
. 1024-1030
928
Placement_-1211
1143 Spacing of column head captions__
1144
-1143-1144 Spanner head (see also Banner head, and indi1002

Roman numerals, use as table number _ 402-403
Rule, horizontal (see also Rule, vertical):
1032
Boldface type, use with_1201, 1211, 1313-1314
Boxhead_1311
Definition_-1315-1316
E n d or bottom rule of table, use as ..
1314
Parallel rule, use of. ..
1313
Single rule, use of_.1314
Top rule of table_
959
Total and subtotal lines, use with_1312
Types__-

vidual components of boxhead):
1253
Clarifying point of reference_204, 1202, 1251
Definition_-1255
Omission_1252
Purpose and function . . .
217
Top spanner in narrow-divide tables_-1254
Total columns. . .
132
Special purpose table_-Specific footnote (see also Footnote), defini603
tion_352
"Specified" use in area segment of title_.
1117-1118
Spelling of area names_-

Standard-series tables:
Rule, vertical (see also Rule, horizontal):
411
Definition . . .
1227
Column head or caption, use in_-. 413-414
Numbering scheme_-1331
Combinations to be avoided.414
Omission of tables in series_1321
424
Definition_-Prefix letters as table numbers, use of_1333
Extension at bottom of page__412
Purpose and use_-1330
Parallel rule_Suffixletters as table numbers, use of. .. . 433-435
1332 Standard metropolitan areas-Range of influence _
1114
Single rule:
States:
1327, 1329
Bold1109
Abbreviations_1326
Hairline or lightface347
Abbreviations used with city names_1327-1329
Medium__345,350
Area segment of title, use in ..
1325
Space allocation ..
Listing alphabetically or geographically_ 1105, 1107
1324
Space requirements . . .
Listing under division heads_-1103
1323
Specifications for ..
Metropolitan districts, use of names in ..
1114
1322
Types and weights_Required with certain cities_347
Standard stub arrangements_-832
Rules (see also Rule, horizontal, and Rule,
State name with cities, special c a s e s 1109
vertical):
Stub
listing
.
.
.
.
1103-1107
1302
"Dashes" versus ''rules"_
1301 Structural parts of table (see also individual
Definition-1314,
1323
ecifi1cations__Drafting sp 1
201-208
parts), component parts_- 1301, 1311-1316 Structural types of tables (see also individual
Horizontal rule_
1322
Mode of describing_
types):
226, 1334
Tables enclosed in rules_Classification not based on number of pages. 213
. 1301, 1321-1334
Vertical rule_-225
Composite forms ..
Running head:
211
Definition_-208
Conventional treatment_215-226
Described and compared ..
201
Not part of table_214
Narrow table as the norm_Omission of time-reference segment of title,
211-226
Specific types__368
use w i t h . 226, 1334
Tables in rules ..
Omission of universe segment of title, use
Tables without stubs_-324
with . . .
208 Stub (see also individual component parts):
Part of folio l i n e 345
Specification of State name, use as ..
Allocation as compared with boxhead . . . 701-713
414
Standard-series tables_
Areas, standard listing_-1101-1118
Rural. See Urban and rural.
Center head and subhead_-812, 911-920
Rural-farm and rural-nonfarm, use of hyphen. 1113
Classification term substituted for "total"__ 643
813, 921-933
Colon and dash (read-in) lines . . .
"Selected," use in area segment of title_352
918-920
Combining center heads__Semicolon (see also Comma, Colon, Punctua. 203, 803, 811-821
Component parts._tion, etc.) use in time-reference segment_
363
Continued line _
819
Series tables. See Standard-series tables.
215
Continuous-Simple table _
133
203,802
Definition_-Single rule (see also Rule, horizontal, and Rule,
1001-1010
Indention_vertical):
. 817, 1011-1018
Leaders.
Horizontal rule, use as
1313
814, 941-948
Line caption, data line_Vertical rule, use as--1322, 1324-1329
Line caption, total and subtotal_- . 815, 951-960
Singular versus plural form in description.
307
712
Mechanical
advantages_-Size-referent, omission or use_
348
Multipage narrow table with continuous
218
boxhead.Small letters (see also Arabic numerals, Capital
801
Organization of discussion_letters, etc.):
633
Reference symbols, use as.
221-222
Parallel table. . .
Suffixes to table numbers. . .
402-403, 431-435
711
Presentation principles_Table number, use as--402-403
Quantitative and qualitative classifications,
Table title, use in ..
308
standard listing___ 1131-1133
Reference range of symbols...
641-642
Source note:
. 207, 604, 621
Reference symbol placement_637
DefinitionGeneral principles_
622
Relationship between stubs of varying
1141-1144
Headnote explanation_detail...
512
1143
Unpublished manuscriptResidual items. . .
624
1325
Use_Space allocation._623
Space breaks_
. 818, 1021-1030
132
Source tables_-

264

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

INDEX
Paragraph
Paragraph Text tables:
Footnotes_713
635, 638, 654
Need for title_-124
831-835
Numbering _
811, 901-909
407
Summary tables _
132
-1121-1124
1031-1033 Text tabulation:
Definition_121
_ 816, 961-966
Disadvantages_124
Footnote placement when table at bottom
Stub leaders. See Leaders.
of page_Stubhead or box:
Footnote, use of_606
907
Amplifying title_General features __
122
908
Captions to be avoided_Introduced by a colon122
City, town, county specification _
Overuse and misuse.
123
902
Coverage_Release table without titlo _
125
Cross-references i n 203, 811, 901, 1202 Third-measure table. See Fractional-measure
Definition_-904-906
Order of listing_tables.
901 Time periods. See Census dates, and Dates.
Purpose_
903 Time-reference segment of title:
"Subject," "Class," or "Item," use oL _
Style (see also Capitalization, Punctuation,
"And," use of_367
Definition _
Wording, etc.):
_ 303,361
1209
"During," use of_Boxhead_-367
662
Hyphen, use of_-Footnote_367
523
Listing of dates _
Headnote_--364-368
308
Omission and curtailment_Title.
368
1426
Placement-Unit-indicator.-362
Punctuation__363
"To," use of_.367
Subentry not shown separately included in
953
total_-Title (see also individual component parts):
Subhead. See Boxhead, Center head, Stub,
Amplified by stubhead or box __
907
etc.
Area or space segment__ 303, 341-352
903
"Subject," use in stubhead_-.Arrangement of segments _
304
Subtitle (see also Title):
Capitalization and indent_308
502
Headnote as a limited form _
Classification segment_303, 331-336
314
Tables_-Comparison with stubhead_-902
Subtotal column. See Column head.
Component parts or "title segments" __
303
Subtotal line. See Line caption.
Coverage and brevity __
_ 302, 306, 336
Suffix letters (see also Prefix letters, Table
Definition_-202, 301
number, etc.):
Footnote relationship _
1115
Census tracts, use in _
613
431
Footnote to title not used_615
Definition_-Function_435
302
Examples of numbering_433
_ 502,613
Headnote relationship_Objectives in usage_Principles of table relationship and posiParallel t a b l e - 221
432
tion_
Parallel versus multipage narrow tables_
218
434
Relationship to master table_-Percent data, description of_-371-374
132
Summary tables_-Relationship
between
content
and
arrange633
Superior figures, footnote usage_ment_304
Symbols (see also Arabic numerals, Capital letTime-reference segment _
_ 303, 361-368
ters, Roman numerals, Reference symbols,
Title-writing methods_-311-314
402
etc.)
Treatment of segments_
305
Type size__309
Table number (see also Numbering schemes):
Universe segment__ 303, 321-325
401
Affixes and suffixes-Word division _
308
202
Component of heading_
Wording __
307
202, 401 Title segments_-Definition and purpose __
303
405 Title writing methods __
Duplication.
_ 311-314
406
Normal numbering scheme_404 "To," use in time-reference segment of t i t l e - 367
Omission_421-424 "Total," as a term:
Prefix letters_-411-414
Standard-series tables_
Meanings __
952-958, 1232
-431-435
Suffix letters_-Placement_-957
403 Totality table. See Master table.
Symbol combinations_402 Total cells, columns, and lines (see also Cell,
Symbols used
407
Text and tabular section_Column head, Line caption):
Table title. See Title, Area segment, Classi_ 954, 1232
Aggregate_fication segment, Time-reference segment,
Area names _
1116
Universe segment, etc.
Boldface _
_ 959, 1032
Captions_815
_ 955, 1232
Class in itself.-414
Table of contents, standard-series tables_Double-duty totals __
1254
226, 1334
Tables in rules or boxes_-Tables in series. See Standard-series tables.
Footnoting_642-643
Tables without stubs __
224
Grand-total celL
956, 1233
Tables without titles (see also Text tabulation):
Indention1004-1007
122-125, 606
Footnoting_Meanings, differences in _
_ 952-955, 1232
125
Release tables_Multiple uses _
956
Tabular footnotes, introduction required
665
Placement in captions_957
Tabular presentation:
Reader-break grouping, effect on_
1028
101
Definition_-_ 953, 1232
Result of summation_102
General classes_Spanner heads__1254
Style__Terminology. See Wording.
959
Territories and possessions:
"Total" versus "all"
958
1102 Total rule. See Rule, horizontal, and Rule,
Stub listing_Title, use in __
vertical.
Stub-Continued
Space requirements _
Standard stub arrangements _
Stubhead or box_-Time periods, standard listing_.Type specification__Unit-indicator _


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

265

INDEX
Paragraph Unit-indicator-Continued
Paragraph
Tracer numbers:
1313
Parentheses, use of_
Bottom boxhead mle, use of.1203
821
Part of field, not of box__
Compared with line or group numbers_-1203
820
Definition __
Placement in field_1422
Placement in stub _
1014
Leaders, use with_-963
947
Presentation improper in stubhead or box
Line caption overrun_908
Range of influence _
218
Not used in multipage narrow table_1423
221
Reference range_Parallel table_.
964
Restrictions on use __
962
Type specification (see also Boldface, and
_ 1252, 1253
Spanner head_-Specify units of measw-e clearly _
Italic):
965
Style and abbreviation_Center heads in stub..916
1426
Undercut spanner __
948
Line caption, data line_
1272
Universe segment of title __
Line caption, total and subtotal line_.
959
Stub, boldface and italic_-_ 1031-1033
Titles_
309 Universe segment of title:
Vertical rule
1323
Curtailment by use of headnote_323
Types of tables (see also Structural types of
Definition303, 321
Omission of portion thereof_tables and individual types)_-- _ 132, 137, 211-226
324
Placement_-322
Unit-indicator _
Undercut spanner:
325
Conditions of use _
1272 Unpublished data, source citation_624
_ 1202, 1271 Upright table. See Narrow table.
Definition_-Urban and rural:
Underscore (see also Rule, horizontal):
Center heads, typing__
Area segment, use i n 919
349
Stub listing_-Total lines (field only)-959
1113
_ 964,966
Unit column_Unit of presentation in stubbead _
908 Vertical rule. See Rule, vertical.
Unit of quantity. See Unit-indicator.
United States:
1224
Stub listing__1102 Word order of column heads _
344 Wording:
Title, use in_Area segment of title_
343
Basis of size classification __
34.8
Unit-indicator:
Classification segment of title _
307
1425
Capitalization_
Footnote_661-665
Colon line, indication in _
925
Headnote_--521-523
Column head or caption, use in _
1225
Percentage titles_-373
966
Columnar arrangement_-"Selected" versus "Specified"352
_ 816,961, 1203, 1402, 1421
Definition_Title_307
1433
Field spanner, use inTotal columns and lines_957
512
Headnote explanation_Use of "and," "during," "to," and hyphen
_ 1033, 1203
Italic type__(-) compared_367
1424
Italics versus parentheses..963
Leaders against_1108
963 Yellowstone National Park, listing_Line-up in s t u b -

266


https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

0