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) &le 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