The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.
Bulletin No. 1223 November 1957 Comparative Job Performance by Age: Large Plants in the Men’ s Footwear and Household Furniture Industries UNITED STATES DEPARTMENT OF LABOR James P. Mitchell, Secretary BUREAU OF LABOR STATISTICS Ewan Clague, Commissioner Bulletin No. 1223 Novem ber 1957 Comparative Job Performance by Age: L arge P la n ts M e n ’s F o o tw e a r H o u s e h o ld in th e and F u r n itu r e In d u s tr ie s UNITED STATES DEPARTMENT OF LABOR James P. Mitchell, Secretary BUREAU OF LABOR STATISTICS Ewan Clague, Commissioner For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington 25, D.C. - Price cents Preface The d i f f i c u l t i e s faced by old er workers in securing and retain in g em ployment c o n stitu te a national problem which i s becoming more p ressing as the proportion o f old er workers in the to ta l labor fo rce r i s e s . The U. S . Department o f Labor, in response to t h is challenge, has been conducting a broad program to examine the problem and to fin d means fo r overcoming i t . As one phase o f t h is program, the present study attempts to furnish inform ation on the r e la tio n sh ip s between age and work performance. Much o f the e a r lie r research in th is general area has shown q u ite c le a r ly th at many employers hold d e fin ite — and o ften unfavorable— a ttitu d e s toward employing o ld er workers. However, o b je c tiv e means o f determining the v a lid it y o f these views have g en erally been la c k in g . In other stu d ies \J conducted by the Department o f Labor, the statu s o f the o ld e r worker has been examined from the standpoint o f pension c o s ts , in surance p la n s, counseling and placement s e r v ic e s , and certain other s p e c ific a sp e c ts. The present study i s d irected toward a measurement o f actual o n -th e job performance o f old er production workers, as compared with the performance o f those in the younger groups. I t represents a continuation o f e a r lie r work in t h is f i e l d . A p i l o t study o f r e la tio n sh ip s betwen job performance and age was undertaken by the Bureau o f Labor S t a t i s t i c s during 1 9 5 5 -5 6 , and the r e s u lts were published as BLS B u lle tin 1203, Job Performance and Ages A Study in Measurement, September 1956. The p i l o t study was planned as a means o f developing o b je c tiv e techniques which would be u se fu l in comparing the performance o f production workers in d if f e r e n t age groups and was p rim arily m ethodological in nature. The present study has as i t s purpose the accumulation o f more extensive data in t h is area which w ill permit the drawing o f more d e fin it iv e conclusions than were p o s s ib le from the p i l o t work. This study was conducted in the Bureau o f Labor S t a t i s t i c s ' D iv isio n o f P rod u ctivity and Technological Developments under the d ir e c tio n o f Jerome A. Mark, a s s is te d by Bernard N. Rein, Stanley F. M i lle r , and Margaret L . M o e lle r. The Bureau wishes to thank those firm s whose cooperation made th is work p o s s ib le . They generously made th e ir records a v a ila b le to the Bureau's rep resen ta tiv es and furnished h elp fu l suggestion s. \J Older Workers Under C o lle c tiv e Bargainings Part I t H irin g, R etention, Job Termination, BLS B u ll. 1 1 9 9 -1 ; Older Workers Under C o lle c tiv e Bargaining: Part I I : H ealth, Insurance, and Pension P lan s, BI£ B u ll. 1 1 9 9 -2 ; Pension Costs in R elation to the H iring o f Older Workers; Older Worker Adjustment to Labor Market P ra ctice s? An A nalysis o f Experience in Seven Major Labor Markets, BES No. R. 1 5 1 ; Counseling and Placement Services fo r Older Workers, BES No. E. 1 5 2 ; How to Conduct an Earning-Opportunities Forum in Your Community. - iii - CONTENTS Page Summary o f f i n d i n g s ......................................................................................................................... 1 Findings ........................................................................................................................................... 1 S c o p e ......................................... 3 Concepts and m e t h o d s .................................................................................................................... ^ Output per man-hour .................................................................................................................... k L im itations o f comparing employedpieceworkers . . ............................................ 5 Attendance ....................................................................................................................................... 6 C ontinuity o f s e r v i c e ....................................................................................... 7 S t a t i s t i c a l procedures ...................................................................................................... . . 8 Establishment o f d ir e c t comparison groups ................................................................ 8 Combinations o f d ir e c t comparison groups ............................................................... 9 A d d ition al combinations ..................................... . ............................................................. 10 In d ivid u al v a r ia tio n ................................................................................................................ 11 S t a t i s t i c a l sig n ific a n c e te s tin g ......................................................................... 11 F i n d i n g s ..................................................................................................................................................... 12 Output per m an -h ou r......................................................................................................................... 12 Industry t o t a l s .............................................................................................................................. 12 Higher and lower paid o c c u p a tio n s .................................................................................... 17 Machine and hand o c c u p a t i o n s ............................................................................................. 21 A t t e n d a n c e ...................................................................................................................................... 2 6 C ontinuity o f s e r v i c e ..................................................................................................................... 30 TABLES 1. 2. 3. Indexes o f output per man-hour fo r incentive workers in 15 men's footwear establish m en ts, by sex and age g r o u p ........................................................... Indexes o f output per man-hour fo r incentive workers in 11 household fu rn itu re establish m en ts, by sex and age g r o u p ...................................................... 13 Indexes o f output per man-hour fo r men in cen tive workers in higher and lower paid occupations in men's footwear establish m en ts, by 13 15 age g r o u p ..................................................................................................................................................19 1+. Indexes o f output per man-hour fo r men incentive workers in higher and lower paid occupations in 11 household furniture establish m en ts, by age g r o u p ................................................................................................................................... Indexes o f output per man-hour fo r women incentive workers in higher and lower paid occupations in men's footwear estab lish m en ts, by age g r o u p ............................................................................................................................................20 Indexes o f output per man-hour fo r women incentive workers in higher and lower paid occupations in 11 household fu rn itu re establish m en ts, by age g r o u p .................................................................................................................................. 20 Indexes o f output per man-hour fo r men incentive workers performing machine and hand operations in men's footwear establish m en ts, by age g r o u p ............................................................................................................................................ 2 b Indexes o f output per man-hour fo r men incentive workers performing machine and hand operations in 11 household fu rn itu re establish m en ts, by age g r o u p .................................................................................................................................. 2 b 19 5. 6. 7- 8. 15 15 ▼ - CONTENTS—Continued Page T ables— Continued 9. 10. 11. 12. 13. 1*4-. 15 . 16. 17. 18. 19. 20. 21. 22. 23. 2*4-. Indexes o f output per man-hour fo r women in cen tive workers p er forming machine and hand operations in 15 men's footwear e s ta b lis h ments, by age g r o u p ........................................................................................................... Indexes o f output per man-hour fo r women in cen tive workers p er forming machine and hand operations in 11 household fu rn itu re e sta b lish m en ts, by age g r o u p ................... .... .......................................................................... Indexes o f attendance fo r production workers in 11 men’ s footwear establish m en ts, by sex and age g r o u p .................................................................. Indexes o f attendance fo r production workers in 11 household fu r n i ture estab lish m en ts, by sex and age g r o u p ........................................................ Indexes o f attendance fo r production workers in 11 footwear e sta b lish m en ts, by se x , method o f wage payment, and age g r o u p ................... Indexes o f attendance fo r production workers in 11 household fu r n i ture establish m en ts, by se x , method o f payment, and age group . . . Indexes o f attendance fo r production workers in higher and lower paid occupations in footwear estab lish m en ts, by sex and age g r o u p ............................................................................................................................................ Indexes o f attendance f o r production workers in higher and lower paid occupations in household fu rn itu re estab lish m en ts, by sex and age group ......................................................................................................................... Indexes o f attendance fo r production workers in 11 footwear e sta b lish m en ts, by se x , years o f s e r v ic e , and age g r o u p ............................ .... Indexes o f attendance fo r production workers in 11 household fu r n i ture establish m ents, by se x , years o f s e r v ic e , and age group . . . Attendance ra tes fo r production workers in 11 men's footwear and fu rn itu re establish m en ts, by se x , method o f payment, length o f s e r v ic e , and occupational pay l e v e l .......................................................................... Indexes o f co n tin u ity o f service fo r production workers in 11 f o o t wear establish m en ts, by sex andage g r o u p ............................................................. Indexes o f con tin u ity o f service fo r production workers in 11 house hold fu rn itu re establish m en ts, by sex and age g r o u p ............................ Indexes o f con tin u ity o f service fo r production workers in 11 f o o t wear establish m en ts, by sex, years o f s e r v ic e , and age group . . . Indexes o f con tin u ity o f service fo r production workers in 11 house hold fu rn itu re establish m en ts, by se x , years o f s e r v ic e , and age group ........................................................................................................................................... C o n tin u ity -o f-se r v ic e ra tes fo r production workers in 11 men's f o o t wear and fu rn itu re establish m en ts, by sex and length o f s e r v ic e . 25 25 29 29 31 32 11 11 11 11 33 3^ 35 36 37 39 39 *40 *4-1 h2 CHARTS 1. Percent o f workers age *4-5 and over with output per man-hour greater than the average fo r age group - ^ ................................................................. 35 - vi - l 6 CONTENTS— Continued Page C h a rts— Continued 2. 3* 4. 5- Percent o f o ld e r workers in h igh er p aid occu pation s w ith output per man-hour g r e a te r than the average f o r age group 3 5 - ^ .................................. 22 Percent o f o ld e r workers in lower paid occu pation s w ith output per man-hour g r e a te r than the average f o r age group 3 5 - ^ .................................. 2 3 Percent o f o ld e r workers in machine occu pation s w ith output per man hour g r e a te r than average f o r age group 3 5 - ^ ......................................................27 Percent o f o ld e r workers in hand occu pation s w ith output per man-hour g r e a te r than average f o r age group 3 5 - ^ ............................................................... 28 APPENDIXES I. II. Q u estionnaire and w o r k s h e e t s ............................................................................................ ^3 D e r iv a tio n o f f o r m u l a s .......................................................................................................... 5^- - v ii - Summary o f Findings This survey o f job performance and age represents one segment o f the U. S . Department o f Labor’ s program in the area o f old er worker stu d ies and i s an extension o f the p i l o t work undertaken in 1 9 5 5 -5 6 . U t ili z in g the methods developed during the p i l o t work, t h is survey centered on the measure o f the r e la tio n sh ip s between age and output per man-hour, attendance, end con tin u ity o f se rv ice in la r g e establishm ents manufacturing men’ s footwear and household fu r n itu r e . Output data were included f o r over 5 ,1 0 0 production workers in 26 establishm ents— 15 in the footwear industry and 11 in the fu rn itu re in d u stry . Comparable output per man-hour data fo r workers in 4 footwear establishm ents included in the p i l o t in v e stig a tio n were incorporated with the data obtained in the current in v e stig a tio n in order to furnish a broader base fo r the fin d in g s . Attendance data were obtained fo r about 9 ,4 0 0 workers and c o n tin u ity -o f-s e r v ic e data f o r almost 1 0 ,0 0 0 in 22 establishm ents— 11 in each in d u stry. Findings With respect to output per man-hour th is study r e v e a ls, in gen eral, th at fo r both men and women the indexes o f output per man-hour are p ro g ressiv ely higher from the under-25 group to the 2 5 -3 4 age group; indexes fo r subsequent age groups show a gradual d eclin e which i s somewhat more marked fo r the 5 5 -6 4 age group. F in a lly , at the 65-and-over l e v e l , a sharper d eclin e occu rs. How ever, the d iffe r e n c e s between age groups through the 5 5 -6 4 group are gen erally small 6*ith the exception o f only 1 group o f women in 1 in d u stry, they are with in 8 percentage p oin ts o f the base, 3 5 -4 4 age group) and the d iffe r e n c e s be tween average output per man-hour o f adjacent groups (a ls o with 1 exception) are p a r tic u la r ly sm all. At the same tim e, there i s a wide range in output w ithin each o f the age groups, so th at the output o f any given ind ivid u al i s l i k e l y to d i f f e r sharply from the average o f h is age group. V ariations in the output per man-hour o f persons in the same age group are greater than d iffe r e n c e s between the average output o f d iffe r e n t age b rack ets. Consequently, su b sta n tia l proportions o f workers in o ld er age groups perform b e tte r than the average fo r younger groups. As an i l l u s t r a t i o n , in the 4 5 -5 4 age group. 47 percent o f the women workers (in both the footwear and fu rn itu re p la n ts) have higher p ro d u c tiv ity than the average woman worker in the 3 5 -4 4 age group. This wide v a r ia b i li t y w ithin age groups coupled with the small d iffe r e n c e s between age group averages in d ica te s that conclusions about individual performance cannot be based s o le ly on chrono lo g ic a l age. Analysis o f r ela tio n sh ip s between age and output per man-hour when the workers are c la s s if ie d and compared by type o f operation and pay l e v e l-h a n d operators versus machine op erators, higher versus lower paid occupations— re v e a ls the same b asic pattern as shown by the industry t o t a l s . - 1 - The in feren ce to be drawn from the fin d in g s concerning output per man hour i s that in d ivid u al c a p a b ilitie s are the most important fa c to r s to be considered in s e le c tin g workers f o r s p e c ific jo b s and th at any attempt to e s ta b lis h formal age l i m i t s fo r various c la s s e s o f jo b s , even where the job content i s c le a r ly d efin ed , f a i l s to take in to account the wide d iv e r s ity o f a b i l i t y l e v e l s shown by persons o f the same age. D iffe ren ces in attendance ra te s between one age group and another are extremely sm a ll. In the 11 footwear p la n ts f o r which attendance data were compiled, the indexes o f attendance among the 6 age groups vary by l e s s than 1 percent f o r the men and by 1 . 2 percent fo r the women. In the fu rn itu re p la n ts they vary by l e s s than 4 percent f o r both men and women. The absence o f any trend in r e la tio n to age and the apparently random manner in which the in s ig n ific a n t d iffe r e n c e s between age groups are d is tr ib u te d , suggest th a t age as a f a c t o r r e la tin g to a w orker's attendance can be ignored. No age attendance p attern s emerge from the indexes o f workers c la s s i f i e d according to method o f payment, len g th o f s e r v ic e , and occupational pay l e v e l , although in a few cases the d iffe r e n c e s between age groups are somewhat la r g e r than the d iffe r e n c e s f o r the industry t o t a l s . In the two in d u strie s the women workers, regard less o f age, show a s l i g h t l y low er attendance re te than the men workers. At the same tim e, both men and women in the footwear p la n ts have higher attendance ra te s than those in the fu rn itu re p la n ts . For both in d u strie s hourly paid workers have lower attendance r a te s than in cen tive workers and workers with 2 or more years o f se r v ic e have higher ra te s than those with l e s s than 2 y e a r s ' s e r v ic e . F in a lly , f o r the most p a r t, workers in higher paid job s have a b e tte r attendance record than those in lower paid jo b s . In tern s o f con tin u ity o f serv ic e (th e proportion o f workers who remained on the jo b during a 1 -y e a r p e r io d ), the study revealed th at the percentage o f workers who remain i s high est f o r ages 45 through 6 4 . As might be expected, both the youngest and o ld e s t groups have the highest percentage o f sep aration s. Yet in the footwear p la n ts the in flu en ce o f retirem ent on the o ld e s t groups i s not as great as the in flu en ce o f other fa c to r s causing separations among the youngest group. Examination o f the con tin u ity indexes o f workers grouped by in d u stry, sex , and len gth o f se rv ic e rev ea ls the same p attern in d icated above except th at among workers with l e s s than 2 y e a r s' s e r v ic e , the d eclin es in the ra tes fo r the o ld e s t and youngest groups are greater than those fo r the corresponding groups o f workers with 2 or more years o f s e r v ic e . I t i s apparent th at the rela tio n sh ip between job performance and age i s by no means sim ple. For each o f the three asp ects o f job performance examined a d iffe r e n t p attern emerges. For output per man-hour, although there are changes on the average associated with age, there i s wide v a r i a b i li t y about the averages; in the case o f attendance, no rela tio n sh ip i s found; and fo r con tin u ity o f s e r v ic e , with the exception o f the o ld e s t age group where the in flu en ce o f re tirem ent i s f e l t , as age in creases the indexes gen erally in c re a se . 2 Scope This study represents an outgrowth o f a p i l o t survey o f r e la tio n sh ip s between jo b performance and a g e . Whereas the e a r lie r survey was methodological in approach, the present work concentrated on expanding the coverage so th a t more d e f in i t i v e conclusions could be drawn. The number o f workers on which data were obtained was increased and the coverage was expanded geographically to include p la n ts in the Middle West and Southeast regions as w ell as the Middle A tla n tic and New England re g io n s. This report p resents data fo r the in d u strie s producing wooden household fu rn itu re (upholstered and unupholstered) and men's footwear (excep t ru bber). These in d u strie s were selec ted fo r study because they show a d is tr ib u tio n o f men and women workers throughout a l l age groups which i s sim ila r to manufac turing as a whole and because they involve wide use o f in cen tiv e systems o f payment. This l a t t e r c h a r a c te r istic i s e s s e n t ia l, fo r i t fu rn ish es a means to measure individual output. The f a c t that management and union sources in these two in d u stries had expressed in te r e s t in the problem o f job performance and age was a contributing fa c to r in th e ir s e le c tio n f o r study. Because o f the n e c e ssity o f making output per man-hour comparisons on ly among workers o f the same sex engaged in the same occupation, i t was important to s e le c t la r g e establishm ents fo r t h is study to secure’ a s u f f ic ie n t number o f o b se rv a tio n s. The sample was drawn from a r o s te r o f a l l la r g e firm s having in cen tiv e payment plans in the 2 in d u strie s— u su a lly firm s se le c te d employed a t l e a s t 300 workers paid on an in cen tive b a s is . The establishm ents may not n e c e ssa r ily be representative o f a l l la r g e firm s in the in d u s tr ie s , f o r some establishm ents were eventually excluded because o f an unw illingness to co operate or a la c k o f su ita b le record s. In order to fu rn ish a broader base fo r the fin d in g s , comparable output per man-hour data f o r workers in U footwear establishm ents included in the p i l o t in v e stig a tio n were incorporated with the data f o r output per man-hour. - 3 - Concepts and Methods As a r e s u lt o f the p i l o t in v e s tig a tio n , three in d ic a to rs were found to be su ita b le fo r representing an in d iv id u a l's job perform ance-output per man-hour, attendance, and con tin u ity o f s e r v ic e . They were selec ted because they afforded o b je c tiv e measures and were d ir e c t ly a v a ila b le from p lan t records Output Per Man-Hour Probably the most important aspect o f an in d iv id u a l's job performance i s h is p ro d u c tiv ity , f o r the ra te at which a worker produces acceptable output i s a key in d ica tio n o f how w ell he w i l l be able to meet production requirements In t h is study, p ro d u ctiv ity was defined as the in d iv id u a l' s p h ysical volume o f production per hour— h is output per man-hour. For the most p a r t, output per man-hour was measured by average str a ig h t-tim e hourly piecework earnings. As described in the report on the p i l o t study, 2 / t h is in d ica to r was chosen as the b e st a v a ila b le measure o f an in d iv id u a l’ s p h ysical output. Although the lim ita tio n s o f th is measure were recognized, i t was found to be su ita b le f o r the purpose at hand. To d erive t h is measure, i t was necessary th at the earnings data exclude payments f o r other than production work. Thus, bonuses fo r length o f s e r v ic e , makeup payments, 2 / and premium payments fo r overtime and holiday work were elim inated from the earnings fig u r e s f o r each worker. In a d d itio n , a l l time work hours and earnings were removed from the fig u r e s on in d iv id u a ls who worked on both piecework and timework during the survey p erio d . In some c a se s, output per man-hour was measured in terms o f the number o f standard time u n its produced per hour worked. These data were maintained in certa in p lan ts fo r group in cen tive workers as w ell as those paid on an in d iv id u a l p ie c e -r a te b a s is , so that i t was p o s s ib le to expand the coverage to include these workers. The observation period fo r output per man-hour data ranged from U to 12 weeks in the various p la n ts . This was a compromise between a very long period which would tend to even out a ty p ic a l in flu en ces o f a temporary nature, and a very short p erio d , which would permit the in c lu sio n o f a la r g e r number o f in d iv id u a ls . Only those employees who had worked during a l l , or nearly a l l , o f the survey p eriod , were included. 2 / Job Performance and Age, BLS B u ll. 1203 ( p p .8 - 9 ) . 2 / The term "makeup payments" r e fe r s to the ad d ition al amounts paid to workers whose raw piecework earnings f a l l below some le g a l or guaranteed minimum. Eliminated from the p rod u ctiv ity sample were pieceworkers in occupations f o r which management imposed production l i m i t s in order to maintain q u a lity standards. Output measures based on earnings in these jo b s would not have represented production ra tes determined by the individual worker. Beginners and employees who were paid according to group in cen tiv e plans ( f o r whom no in d ivid u al output data were a v a ila b le ) were a lso excluded from the output per man-hour sample. Beginners were id e n tifie d through d isc u ssio n s with foremen and by examining records o f len gth o f serv ic e in the p a r tic u la r jo b . In most occupations, employees with l e s s than 3 months' serv ic e on the jo b were considered beginners. k in d t -a tic B g -a f .SfflmBE-rlflg fin ploggfl. f.is&gwgr.kegjg Although th is study includes data on the p ro d u ctiv ity o f a small number o f group in cen tiv e workers, the great m ajority o f the in d ivid u als studied were paid according to piecework system s. Consequently, many o f the lim ita tio n s c f any study which examines only pieceworkers are relevan t to th is survey. The performance o f pieceworkers, i t i s recognized, may not be e n tir e ly represen tative o f a l l production workers. S k ille d craftsmen are u s u a lly not paid according to production by the p iece and most u n sk ille d workers are paid on a time b a s is . However, the bulk o f production workers are employed in the sem isk illed type o f jo b s , many o f which are covered by piecework p la n s. I t i s q u ite p o ssib le that the fa c to r s determining p ro d u c tiv ity o f piece workers may be d iffe r e n t from those determining the p ro d u c tiv ity o f in d u stria l workers as a whole, and that these d iffe r e n c e s may not apply uniform ly to a l l age groups. For example, p ie c e -r a te job s may, on the average, p la ce greater emphasis on speed, a g i l i t y , and other p hysical a ttr ib u te s which are gen erally assumed to d eterio ra te with age. I f t h is i s so , then data based on the per formance o f pieceworkers would tend to show the old er in d iv id u a ls in a l e s s favorable l i g h t than would be the case i f a v a r ie ty o f payment plans had been included. N evertheless, there i s ju s t i f i c a t i o n fo r a study confined to piece workers, sin ce in many establishm ents the p ie c e -r a te system i s the dominant method o f payment. I t may a lso be argued th a t because only employed persons were included in the study, the old er workers who were s t i l l present in the occupations studied a c tu a lly represented a selected group, since many workers o r ig in a lly in these occupations had l e f t fo r other job s which were not included in the study. However, there are a c tu a lly two types o f s e le c tio n operating here, which may be expected to cancel each other to some extents the excep tion ally superior workers may be assumed to have gone in to b etter paying occupations; the marginal workers who could not maintain the minimum standards required fo r th e ir jobs would a lso have l e f t to enter other occupations. Thus, the o ld er workers' average output ra te s would be influenced by the removal o f these two extremes, and th erefore would be comparable with those o f the younger groups. - 5 - There are no l if e t im e job h is t o r ie s o f workers in the in d u strie s studied which show the exten t to which workers enter e ith e r as time or p ie c e -r a te workers and move from one type o f work with i t s associated pay procedure to another over th e ir l i f e span. N either i s there evidence o f the extent o f movement from production to supervisory or entrepreneurial work on the one hand, and to cu stod ial job s on the o th e r . Therefore, i t was not p o s s ib le , w ithin the scope o f the present study, to determine the extent to which these s itu a tio n s p rev a il in the p la n ts surveyed. Since th e ir e f f e c t s tend to be o f f s e t t i n g , the net r e s u lt i s not believed to be very g r e a t. Attendance Although very im portant, an in d iv id u a l's rate o f production i s not the on ly aspect o f h is jo b performance. The amount o f time he can be depended upon to be on the jo b and maintain h is output— h is attendance ra te— i s a lso s ig n if ic a n t . For purposes o f t h is study, the attendance rate was defined as the r a tio o f days worked to days scheduled. To compute th is r a te , an employee's absences were subtracted from the t o t a l number o f days on which he was scheduled to work during the observation p eriod , and t h is y ield ed the t o t a l days worked. This fig u r e was then divided by the number o f days scheduled. I f a worker per formed any work a t a l l during a day, he was not considered absent th a t day. Days on which a worker was away because o f l a y o f f s , h o lid a y s, shutdowns, or regu lar vacations were excluded from both the days worked and days scheduled. Attendance rather than absenteeism was selected because i t i s believed th a t the proportion o f to ta l scheduled time an employee i s on the jo b i s o f greater importance as a performance in d ica to r than the proportion o f to ta l time he i s o f f . Small d iffe r e n c e s in absenteeism r a te s would tend to exag gerate the r e la t iv e performance o f d iffe r e n t workers. For example, a worker who was absent 2 days in 100 scheduled workdays cannot r e a l i s t i c a l l y be considered twice as dependable as a worker absent U days in 1 0 0 . I t would seem more appropriate to r e la t e the two workers in terms o f th e ir attendance r a t e , i . e . , 98 days present out o f 100 as compared with 9 6 . Unlike the output per man-hour measure, the attendance concept need not be r e s t r ic t e d to workers paid according to any one s p e c ific arrangement. Data are presented fo r timeworkers and group bonus workers, as w ell as fo r piecework employees. For attendance, as in the case o f output per man-hour, the observation period must be lo n g enough to avoid a ty p ic a l s itu a tio n s . The period selec ted fo r obtaining attendance data in the various p la n ts ranged from 12 weeks to 1 y ea r, depending on the nature o f the reco rd s. Parts o f that period coincided with the observation period fo r output per man-hour. - 6 - C ontinuity o f Service A th ird aspect o f job performance examined in th is survey was the rate o f co n tin u ity o f se rv ic e — the proportion o f to ta l workers who remained on the job during a 1 -y e a r p eriod . This in d ica to r— the converse o f the separation ra te— was included as a measure o f the r e la tiv e s t a b i l i t y o f the various age groups. The con tin u ity ra te was computed by d ivid in g the number o f workers in a given group who were s t i l l employed, i . e . did not q u it, get discharged or r e t i r e , a t the end o f the observation period by the number in the group at the beginning o f the p erio d . Employees who were separated, as a r e s u lt o f production l a y o f f were not considered as sep aration s. Accessions during the p erio d , were excluded from both the numerator and denominator. A ll workers fo r whom the necessary data could be obtained were included in th is measure regard less o f payment method. As in the case o f attendance, con tin u ity o f serv ic e rather than sepa ra tio n s were used because small d iffe r e n c e s in the separations rates would tend to exaggerate d iffe r e n c e s between age groups. - 7 - S t a t i s t i c a l Procedures The s t a t i s t i c a l methods applied in th is study were designed, f i r s t , to i s o l a t e the in flu en ce o f age from the many other fa c to r s which a f f e c t a person18 jo b performance and second, to combine measures drawn from small groups o f workers in to la r g e r aggregates. The i n i t i a l measures drawn from small groups, could not fu rn ish meaningful r e s u lt s , but th e ir combination perm itted the drawing o f s t a t i s t i c a l con clu sion s. Establishment o f D irect Comparison Groups Workers were f i r s t grouped according to fa c t o r s , other than age, which might a f f e c t ind ivid u al job performance, such as s p e c ific occupation, method o f payment, and len g th o f s e r v ic e . Within each o f these groups, comparisons were made o f the average performance o f workers in various age groups. The non-age fa c to rs which determined these groups varied f o r each o f the in d ic a to r s (output per man-hour, attendance and con tin u ity o f s e r v ic e ), although fo r a l l three in d ica to rs the groups were lim ite d to workers o f the same sex in the same p la n t. The purpose here was to insure that age-performance observations were made only among workers having in common those c h a r a c te r is tic s which would have an inportant bearing on the performance t r a i t being measured. I f t h is had not been done, any apparent d iffe r e n c e s in the performance o f age groups m ight, in r e a l i t y , be only a m anifestation o f the unequal d is tr ib u tio n o f these c h a r a c te r is tic s among the age groups. For example, i t would be mis le a d in g to compare the average hourly earnings o f two age groups composed o f d iffe r in g proportions o f men and women, sin ce t h is would introduce important fa c to r s other than age. For output per man-hour, d ir e c t comparisons were lim ite d to workers o f the same sex who were regular operators (not beginners or p a rt-tim e workers) performing the same job or se r ie s o f operations in the same p la n t. In parison non-age payment (higher the case o f attendance, the b a sis fo r e sta b lish in g the d ir e c t com groups was somewhat d if f e r e n t . I t was believed th a t the fo llo w in g c h a r a c te r is tic s would in flu en ce attendances Sex; p la n t; method o f (time or in c e n tiv e ); len gth o f se r v ic e ; and l e v e l o f occupation paid or lower p a id ). Workers paid a t an hourly rate are l i k e l y to have d iffe r e n t attendance m otivations from those under an in cen tive system. S im ila r ly , employees with varying le n g th s o f serv ic e may have d iffe r e n t attendance m otivation s. F in a lly , i t i s probable that the fa c to r s in flu en cin g the attendance o f workers in lower paid jobs d i f f e r from those a ffe c tin g the attendance o f workers in higher paid jo b s . Attendance comparisons by age, ignoring these d iffe r e n c e s , might then r e f l e c t changes associated with these fa c to r s rather than with age. T herefore, the groups ‘were estab lish ed so that a l l o f the above fa c to r s were constant w ithin a given group. - 8 - For the c o n tin u ity -o f-s e r v ic e measure, the groups were estab lish ed ac cording to the fo llo w in g c h a r a c te r is tic s : Sex} p lan t} and len gth o f s e r v ic e . The le n g t h -o f-s e r v ic e c la s s if ic a t io n divided workers in to those who had been employed in the p lan t 2 years or more and those with l e s s than 2 y e a r s ' s e r v ic e . Most probably, separations are greater among r e la t iv e ly new employees than among a group o f workers with extended s e r v ic e . To c l a s s i f y each occupation as higher paid or lower p a id , the fo llo w in g procedure was adopted: S p e c ific average hourly earnings c r it e r ia were d eter mined sep arately by region , sex, in d u stry, and method o f payment (in c en tiv e or timework), which were then used to d istin g u ish between the two pay l e v e l s . These c r i t e r i a were derived from data obtained in wage surveys conducted by the Bureau, and represented averages o f a l l employees in each category, re gardless o f p la n t . The actual occupations found in the current study were then designated as higher paid i f the average o f the hourly earnings o f workers in th at category equaled or exceeded the predetermined c r ite r io n , and lower paid i f i t did n o t. Each worker was then c la s s if ie d by age in to 1 o f 6 groups— under 2 5 , 2 5 -3 4 , 35-44-, 4 5 -5 4 , 5 5 -6 4 , and 65 and over. For each in d ic a to r , the performance o f age groups w ithin the b a sic com parison groups was measured by c a lc u la tin g indexes fo r each age group average, u sin g the average o f the 3 5 -4 4 age group as the base. Thus, a measure was derived f o r each age group r e fle c t in g the r e la t iv e performance o f th a t given age group as compared with the corresponding 3 5 -4 4 age group. By transforming the average hourly earnings data o f in cen tive workers into indexes fo r output per man-hour, the p o s s i b i li t y o f d is to r tio n from v a ria tio n s in absolute pay l e v e l s in d iffe r e n t operations or in d iffe r e n t p lan ts was elim inated, and the indexes could be viewed as independent o f the p lant in which the individual worked. Combinations o f D irect Comparison Groups To obtain r e s u lts which would represent la r g e r numbers o f in d ivid u als than were a v a ila b le in the basic d iv is io n s mentioned above, these groups were com bined in to c la s s e s in which the e f f e c t o f fa c to r s other than age was believed to be f a i r l y uniform fo r a l l age groups. In the case o f output per man-hour, the indexes fo r workers in s p e c ific occupations were grouped according to ( l ) sex , (2 ) higher or lower paid occupations, (3) hand or machine occupations, and (4) in d u stry. The b a sis fo r c la s s ify in g occupations in to higher or lower paid categories has already been d escrib ed . The hand-machine d is tin c tio n was drawn by d efin in g a machine job as one which required the use o f a powered d e v ic e . Most o f these d esignations were assigned a fte r d iscu ssion with p lan t o f f i c i a l s as to the p a r tic u la r equipment used in various occupations. The machine-hand d is tin c tio n was made in order to separate jobs which p lace somewhat d iffe r e n t requirements - 9 - on workers. Machine jo b s c a ll fo r s k i l l s said ta le n ts d iffe r e n t from those needed f o r hand op eration s, and the s k i l l s p ecu lia r to machine operations or to hand operations may in flu en ce the p ro d u ctiv ity o f various age groups d i f f e r e n t l y . By making t h is d is t in c t io n , the in flu en ce o f these fa c to r s in age group comparisons was minimized. In the case o f attendance, the p lan t indexes fo r workers in higher and in low er paid job s were combined across p la n ts in to la r g e r groups o f workers c l a s s i f i e d according to ( l ) sex, (2) higher or lower paid occupations, (3) in cen tiv e or timeworkers, ( ) len gth o f se r v ic e , and t'5) in d u stry. 4 S im ila r ly , fo r con tin u ity o f se r v ic e , the p lan t indexes fo r workers with d iffe r e n t le n g th s o f serv ice were combined across p lan t l i n e s • Here, they were c l a s s i f i e d according to ( l ) se x , (2) length o f s e r v ic e , and ( ) in d u stry . 3 To obtain these combinations, each o f the age group indexes fo r a d ir e c t comparison group was combined with a weight which r e fle c te d i t s r e la t iv e importance in the p a rtic u la r combination group. For example, the output per man-hour age group indexes fo r s p e c ific occupations were combined with weights r e f le c t in g the contribution o f each s p e c ific occupation to the output per man-hour measure f o r an occupational group ( e .g . higher paid machine j o b s ) . The weight took in to account the number o f workers in the age group as w ell as the number o f workers in the base group (age 3 5 -4 4 ) fo r each occupation «r*d took the form NaNb/(Na + N^), where Na was the number o f workers in the age group and the number in the base group. L j For attendance and con tin u ity o f se r v ic e , the same form o f weights was a p p lied , except th at the weight was m u ltip lied by a constant fo r each p la n t in order to account fo r d iffe r e n c e s in the periods fo r which these data were c o lle c t e d . In the case o f attendance, t h is constant was the number o f scheduled workdays in the p lan t during the survey p e rio d . For con tin u ity o f s e r v ic e , the fig u r e was the len gth o f the observation period in the p la n t. Additional Combinations Besides the indexes fo r the basic comparisons and the i n i t i a l l y combined groups, more generalized in d ica tio n s o f the r e la t iv e performance o f the various age groups were derived . By combining the occupational group indexes fo r output per man-hour, a separate measure fo r workers in each industry by s e x and occu p ation al c la s s (hand, machine, higher p a id , lower paid) was obtained. S im ila r ly , by subsequently combining these indexes, a separate measure fo r men and fo r women workers in each industry was d erived . For attendance and c o n tin u ity , sim ila r combinations were made to obtain indexes fo r each industry by s e x . To derive these indexes, each o f the component indexes was assigned a constant weight fo r a l l age groups. With th is weighting system, the f in a l lj The d erivation o f th is formula i s shown in appendix I I (pp. 5 4 -5 6 ) . - 1 0 - index was not influenced by d iffe r e n t proportions o f the component groups in one age c la s s as compared with another. Thus, the in flu e n ce o f fa c to r s other than age was constant fo r a l l age groups. The weight used in a l l cases was the t o t a l number o f workers o f a l l ages w ithin the component group. Ind ivid ual V ariation In ad d ition to p resenting the indexes fo r an age group a t the various stages o f combination, the study a lso includes a measure o f the extent o f in d iv id u a l v a r ia tio n w ithin each age group. This measure in d ica te s whether an index derived fo r a p a r tic u la r age group r e f l e c t s the performance o f in d iv id u a ls whose scores show considerable uniform ity or whether, on the other hand, t h e ir scores vary widely about the average fo r the group. A d isp ersion measure was derived fo r the output per man-hour combinations o n ly . D ispersion measures fo r attendance were not derived , since they have lim ite d u s e fu ln e s s . The d is tr ib u tio n o f in d ivid u al attendance ra te s i s skewed, with most scores very d o s e to the maximum score o f 1 0 0 . I t was not fe a s ib le to d erive a d isp ersion in d icator fo r employment con tin u ity because o f the nature o f the measure. The d isp ersion measure employed was the c o e ffic ie n t o f v a r ia tio n , which was expressed as a percent o f the group sco re. As with the age group indexes, the c o e ffic ie n ts o f v a ria tio n fo r the s p e c if ic occupations were combined to obtain a measure fo r the occupational group. These measures in turn were combined to obtain more general in d ica tio n s o f the d isp e rsio n . In each combination, weights were used to r e f l e c t the r e la t iv e importance o f the component measure in the combination. In each case, the weight applied was the number o f workers in the component group, l e s s one. S t a t i s t i c a l S ig n ific a n ce T estin g In the case o f output per man-hour, procedures were employed fo r te s tin g f o r s t a t i s t i c a l sig n ific a n c e the d iffe r e n c e between the age group in d exes. Each age group index fo r an occupational group and combinations o f these groups was compared with the corresponding base group index o f 1 0 0 . The standard error o f the d iffe r e n c e between the two indexes was obtained and i t s standard score computed. With few excep tion s, i t was found that these d i f feren ces were s t a t i s t i c a l l y s ig n ific a n t a t the 5-percent le v e l — that i s , these indexes were s ig n ific a n t ly d iffe r e n t from 100 in the sense th a t, i f there were r e a lly no d iffe r e n c e between an age group and base group, a d iffe r e n c e as great as t h is would be obtained l e s s than 1 time in 20 on repeated sampling^ The indexes found to be not s ig n ific a n t were those o f the women employees aged 4 5 -5 4 and the men aged 2 5 -3 4 in the footwear p la n ts and women aged 4 5 -5 4 and under 25 and men under 25 and 2 5 -3 4 in the fu rn itu re p la n ts . The d i f f e r ences between these indexes and 100 were so small that even with the s iz a b le samples they were not found to be s t a t i s t i c a l l y d iffe r e n t from 1 0 0 . - 11 - ?lP<foagg The fin d in g s presented here r e f l e c t the experience o f production workers in c e rta in la r g e establishm ents in the footwear and household fu rn itu re indus tr ie s . Data on one or more o f the performance in d ic a to r s— p ro d u c tiv ity , attendance, or con tin u ity o f serv ic e — are presented f o r 1 3 ,3 5 0 workers em ployed in 26 establish m en ts, 15 o f which were in the men's footwear industry and 11 in the wood and upholstered fu rn itu re in d u stry. (In the footwear in d u stry , data on attendance and con tin u ity o f serv ic e were obtained from 11 p la n t s .) The data were c o lle c te d f o r s u b s ta n tia lly fu ll-p r o d u c tio n p eriod s during 1 9 5 5 -5 6 . Output Per Man-Hour D irect comparisons o f output per man-hour were lim ite d to those workers whose in d ivid u al contribution to t o t a l production could be measured, and who were performing the same o p era tio n s. Consequently, i t was not p o ssib le to c o lle c t p ro d u c tiv ity inform ation f o r a l l workers in the establish m en ts. How ev e r, output per man-hour measures were obtained f o r 5 ,1 4 7 production workers— 1 ,6 5 1 in the fu rn itu re establishm ents and 3 ,4 9 6 in the footwear p la n ts . The m ajority o f these workers, as mentioned above, were pieceworkers, but there were some workers fo r whom ind ivid u al output data could be d erived , even though they were paid according to a group in cen tive p la n . I t was not p o s s ib le , however, to include any hourly rated workers in t h is group. Separate r e s u lts are presented fo r men and women employed in the 2 indus t r i e s (t a b le s 1 and 2 ) , and the r e s u lts are fu rth e r subdivided in to earnings l e v e l groups (ta b le s 3 , 4 , 5, and 6 ) , and according to hand and machine occupations (ta b le s 7 , 8 , 9 , and 1 0 ) . Industry T o ta ls . From an examination o f the combined indexes representing a l l occupational l e v e l s , two broad conclusions emerge. F i r s t , there i s a c o n siste n t p attern o f re la tio n sh ip between age and output— a r i s e in the indexes from the under 25 age group to the 2 5 -3 4 group, follow ed by a gradual d e c lin e which i s somewhat more marked fo r the 5 5 -6 4 group. F in a lly , a t the 65-en d-over l e v e l , a sharper d eclin e occu rs. Second, i t should be noted th at d iffe r e n c e s between age groups are g en era lly sm a ll. With only one excep tion, a l l o f the output indexes, through the 5 5-6 4 year group are w ithin 8 percentage p o in ts o f the base group indexes. D ifferen ces between adjacent 1 0 -y e a r age groups are (a ls o with one exception) r e la t iv e ly sm a ll. These observations apply to both the footwear and fu rn itu re in d u s tr ie s , and to both men and women workers. The one exception ind icated above occurs in the case o f women fu rn itu re workers aged 5 5 -6 4 , with the index d e c lin in g to 8 5 .6 . - 1 2 - Table 1 . Indexes o f output per man-hour f o r in cen tive workers in 15 men's footwear establishm ents, by sex and age group age group jo— Women Men Age group Index C o e ffic ie n t o f v a ria tio n (percent) Number of workers Index C o e ffic ie n t o f v a r ia tio n (percent) 98 9 3 .8 1 7 .9 111 9 4 .4 1 7 .1 1 0 0 .3 1 0 0 .0 1 6 .3 1 3 .8 292 1 0 2 .8 589 1 0 0 .0 1 7 .5 1 5 .2 9 7 .7 1 4 .1 534 9 8 .8 1 5 .6 Number of workers Under 25 9 2 5 -3 4 . . 278 3 5 -4 4 . . . . 484 460 55-64 . . 322 9 2 .5 1 4 .5 219 9 4 .1 1 3 .1 75 8 1 .1 1 6 .6 34 8 8 .0 2 0 .7 4 5 -5 4 65 and over Table 2 . Indexes o f output per man-hour fo r in cen tive workers in 11 household fu rn itu re establishm ents, by sex and age group (Age group 35-4A=100) Women Men Age group Number of workers Under 25 • Index C o e ffic ie n t o f v a ria tio n (percent) liumber of workers Index C o e ffic ie n t o f v a r ia tio n (percent) 9 8 .5 1 6 .3 22 I d .4 1 8 .8 1 0 1 .5 1 0 0 .0 1 5 .1 1 1 .8 79 97 107. A 1 0 0 .0 1 9 .4 1 7 .8 2 5 -3 4 3 5 -4 4 . . . . 214 436 372 4 5 -5 4 . 218 9 6 .1 1 1 .0 63 9 8 .7 16.0 96 9 4 .5 9 3 .6 1 1 .8 33 1 8 5 .6 1 8 .6 . 5 5-6 4 . . 65 and ovei 20 1 1 .6 0 /) ( 1 /) 1 / Fewer than 5 observations were considered in s u ffic ie n t fo r d erivin g the in d exes. 13 - In the footwear p la n ts , the men show a s l i g h t l y more pronounced tendency to d e clin e in performance, with advancing age, than do the women. For age group 4 5 -5 4 , the index fo r the women i s 9 8 .8 whereas the men in t h is group record an index o f 9 7 . 7 . S im ila r ly , a t age 5 5 -6 4 , the 2 indexes are 94*1 and 9 2 * 5 , r e s p e c tiv e ly . In the fu rn itu re p la n ts , a p attern o f d iffe r e n c e i s not apparent. In the 4 5 -5 4 age group, the women1s output i s s l i g h t l y b e tte r than the men’ s (an index o f 9 8 .7 as compared with 9 6 . l ) ; the reverse i s true f o r the 5 5 -6 4 age group— an index o f 9 4 .5 fo r the men as again st 8 5 .6 f o r the women. I t i s p o s s ib le th at t h is l a t t e r age group i s one in which the r e la t i v e l y heavier p h ysical demands o f occupations in the fu rn itu re in d u stry, as compared with those in the footwear p la n ts , p lace the old er women in a l e s s advantageous p o s it io n . The jobs in which the women are commonly employed in the fu rn itu re industry are sewing covers, tacking u p h olstery, packing and wrapping c h a ir s, and hand sanding. In comparison with the usual footwear occupations fo r women— s titc h in g , p a stin g , cementing, sk iv in g , fo ld in g , and d ressing— the fu rn itu re occupations in volve la r g e r p ie c e s on which to work and operations which n e c e ssita te strength in ad d ition to s k i l l . Even in the 5 5 -6 4 age group o f women fu rn itu re workers, however, performance i s w ithin 15 percentage p oin ts o f the base group index (th e 3 5 -4 4 age group). Although the 2 5 -3 4 age group, in both the footwear and fu rn itu re p la n ts , and fo r both men and women, c o n stitu te s the age group showing the peak per formance index, the su p erio rity o f t h is group over those aged 3 5 -4 4 i s gener a l l y s l i g h t . In the case o f women fu rn itu re workers, the index fo r t h is age group i s 1 0 7 .4 , but in none o f the other 3 categories does i t exceed 1 0 2 .8 . In the 65-and-over age group fo r which data are a v a ila b le , the indexes are a l l considerably lower than those o f the base groups (3 5 -4 4 y e a r s ). In the footwear p la n ts , output o f the women workers age 65 and over holds up b e tte r than th at o f the men, with an index o f 8 8 .0 as compared with 8 1 .1 fo r the men* Among the fu rn itu re workers, the number o f women 65 and over was in s u f f ic ie n t to permit the presentation o f a meaningful index, but the men showed an index o f 9 3 .6 . The foregoin g comments r e fe r to the average indexes o f the age groups. However, in an actual employment s itu a tio n , there are always s p e c ific in d i vidual workers in volved, so that inform ation regarding the average performance l e v e l s o f la r g e groups i s , by i t s e l f , o f lim ite d v a lu e . What i s necessary i s some measure o f how c lo s e ly the performance scores o f the ind ivid ual workers conform to t h is group average. In t h is study, the measure used to provide t h is inform ation i s the coef f i c i e n t o f v a r ia tio n , which r e f le c t s the d iffe r e n c e between the age group average and the individual sc o re s, j j / *>/ The c o e ffic ie n t i s calcu lated by d iv id in g the standard d eviatio n by the mean and in d ica te s the r e la tio n sh ip between the value o f the mean and d istan ce from the mean w ithin which a s p e c ifie d proportion o f the observations w ill be i f the ind ivid ual observations included in the mean are d istr ib u te d in a normal frequency curve. For example, i f the index o f an age group were 90 and the c o e f f ic ie n t o f v a r ia tio n 10 p ercent, then about tw o-th ird s o f the ind ivid ual indexes would be between 8 1 .0 and 9 9 .0 (th e se lim it s being the mean p lu s and minus 10 percent o f the mean). - 14 - W ith resp ect to the v a r ia b ilit y o f in d ivid u al age groups, th ere appears to be no c o n siste n t re la tio n sh ip between age and the c o e ffic ie n t o f v a r ia tio n . Although fo r both sexes and in both the footw ear and fu rn itu re in d u s tr ie s , th ere i s a s lig h t tendency fo r the average c o e ffic ie n t s o f v a ria tio n o f the medium age groups (coverin g the ages 3 5 -5 4 ) to be sm aller than those in the younger and old er groups, th is p attern i s not very marked and i t i s u n lik e ly th a t any broad conclu sions on th is p o in t would be j u s t i f i e d . W ithin the in d ivid u al age groups, t h is v a r ia b ilit y i s q u ite la r g e and c a rr ie s s ig n ific a n t im p lic a tio n s. As can be seen from chart 1 , su b sta n tia l proportions o f the workers age 4 5 -5 4 have output indexes which are higher than the average index o f the 3 5 -4 4 age group. This proportion v a rie s from 47 percent fo r th e women workers in both footw ear and fu rn itu re p la n ts to 39 percent in the case o f men fu rn itu re w orkers. 6 / Even in the age group 5 5 -6 4 th ere i s a s ig n ific a n t prop ortion o f workers w ith output indexes higher than the average o f the 3 5 -4 4 age group. In the case o f men w orkers, they coraprise 30 percent o f the employees in the footw ear p la n ts w ithin t h is age group and 34 percent in the fu rn itu re p la n ts . For women w orkers, the percentages are 35 and 2 1 , r e s p e c tiv e ly . These r e s u lts in d ica te th a t an evalu ation o f in d iv id u a l workers i s fa r more im portant than any general id ea s concerning the r e la tio n sh ip between age and p ro d u c tiv ity . In view o f the sm all d iffe r e n c e s in the averages o f the age groups and the wide v a r ia b ilit y in in d ivid u al performance r a te s found in th is stu d y, i t i s c le a r th a t an employer should emphasize th e development o f e ffe c t iv e means fo r determ ining the in d ivid u al c a p a b ilitie s o f p a rtic u la r w orkers, rath er than attem pt to apply conclu sions based on the ch ron ological age o f the workers in v o lv ed . I t may be noted th at the women workers show s lig h t ly higher average c o e f f ic ie n t s than the men. In the footw ear p la n ts , the average fo r men i s 14*6 percent and fo r women, 15 *6 p ercen t. In the fu rn itu re p la n ts , the men showed an average c o e ffic ie n t o f v a ria tio n o f 1 3 .9 p ercen t, whereas the women averaged 1 7 .9 p e rc e n t. W hile the women show a higher v a r ia b ilit y in both in d u s tr ie s , the d iffe r e n c e between the men and women workers i s g reater in the fu rn itu re p la n ts . §/ These p rop ortion s were derived u sin g the index and the average coef f ic ie n t o f v a ria tio n w ith the assumption th a t the form o f the d is tr ib u tio n o f the output per man-hour scores fo r in d iv id u a ls was norm al. There i s evidence supporting the view th at th is w ill be the case in output per man-hour sc o r e s . See In d ivid u al P ro d u ctiv ity D iffe ren ces S e ria l No. R. 1 0 4 0 , February 1940 (p p. 18 and 19) • The actu al prop ortion s were computed and fo r the most p a rt they were very c lo s e to those based on the normal d is tr ib u tio n . - 15 - Chart 1. PERCENTOFWORKERSAGE45 ANDOVERWITHOUTPUT PER MAN-HOUR GREATERTHANTHEAVERAGE FORAGE GROUP35-44 WOMEN P ercen t Age F ootw ear P la n ts G rou p I M ON 4 5 - 5 4 Furniture .P la n ts I A ge F o o tw ea r P la n ts G rou p 5 5 - 6 4 Furniture .P la n ts UNITED STATES DEPARTMENT OF LABOR BUREAU OF LABOR STATISTICS The b a sis fo r th ese d iffe r e n c e s i s not c lea r* I t i s p o s s ib le th a t the economic and dom estic situ a tio n o f women employees shows le s s u n iform ity than th a t o f men, and th a t t h is , in tu rn , r e s u lts in im portant d iffe r e n c e s in m otiv a tio n . The ages o f the ch ild ren , fo r example, should have more e ffe c t on the m other's employment requirem ents than on the f a t h e r 's . In g en era l, i t should be pointed out th at women tend to move in and out o f the la b o r fo rc e more freq u en tly than men. Their movement i s in response to a wide range o f in flu e n ce s— both economic and s o c ia l. As noted p re v io u sly , the current study in d erivin g the p ro d u c tiv ity indexes did not consider any fa c to r s other than volume o f output per man-hour. Charac t e r is t ic s such as perseveran ce, q u a lity o f ou tp u t, and m aturity o f judgment are not r e fle c te d in the in d exes. These a re , o f cou rse, im portant y a rd stick s to be employed in determ ining the o v e r a ll e ffe c tiv e n e s s o f an employee. A l though they are im precise and s u b je c tiv e ly evaluated c h a r a c te r is tic s , variou s other stu d ies 7 / in d ica te th a t o ld er workers would have the advantage in th ese r e s p e c ts . T herefore, bearing in mind the nature o f the age-output re la t io n s h ip -s m a ll d iffe r e n c e s between the several age groups and great v a ria tio n among the in d iv id u a ls w ithin age groups— i t would be appropriate fo r an employer to consider the advantages o f d ispensin g w ith s p e c ific age lim it s in h ir in g , and thus provide h im se lf with a much broader f i e ld from which to make in d ivid u al s e le c tio n s . Even the p ra c tic e o f s e ttin g age lim it s fo r jo b s which are considered e s p e c ia lly rigorous would appear to have doubtful m e rit, sin ce the fin d in g s in the fu rn itu re p la n ts , where p h ysical cap acity does appear to be an im portant fa c t o r , in d ica te th at su b sta n tia l percentages o f workers age 5 5 -6 4 performed b e tte r than the average o f those age 35—4 4 . Higher and Lover Paid Occupations. As in d icated in the se c tio n on sta t i s t i c a l methods, the s p e c ific occupation indexes in th is study are combined in to broader groupings to fu rn ish more general conclu sions than are p o s s ib le from the lim ite d numbers o f workers in in d ivid u al occu p ation s. Such groupings are based on c h a r a c te r is tic s oth er than age which are b eliev ed to in flu en ce the measure under co n sid era tio n . In the case o f output per man-hour, the groupings are determined by the average earnings le v e ls o f the occupations and by the d is tin c tio n between hand and machine occu p ation s. 2 / See, fo r example, V a lte r G ershenfeld, Job S p e c ific a tio n s and Employment P o lic y (p p. 1 7 1 -1 7 2 ) and James H. M ullen, The C osts o f R etaining Older Vbrkers in Industry (p . 1 9 7 ). (In Proceedings o f the Second Conference on the Problems o f Making a L ivin g W hile Growing O ld .) Commonwealth o f Pennsylvania and Temple U n iv e rsity , 1 9 5 3 . W illiam H. Bowers, An Appraisal o f Workers C h a ra cte ristics as R elated to Age (in Journal o f A pplied P sychology, October 1 9 5 2 , V o l. 3 6 , No. 5 , p p .2 9 6 -3 0 0 ). - 17 - The summarization o f r e s u lts sep arately fo r employees in higher and low er paid occupations i s based on the assumption th at th ese two groups may rea ct d iffe r e n t ly , in terms o f th e ir output per man-hour, to the changes a sso cia ted w ith age. I t i s g e n era lly assumed, fo r example, th a t higher paid jo b s tend to in clu d e the more complex op eration s in which s k i l l and experience are im portant fa c to r s in determ ining perform ance, whereas the low er paid jo b s tend to be sim pler and more r e p e tit iv e , p o s s ib ly p la cin g a greater premium on p h ysical a b i l i t i e s as con trasted w ith mental ones or where s k ill and experience re quirements are lo w er. The s p e c ific content o f many jo b s v a rie s from p la n t to p la n t. Some o f the common higher paid jo b s in the fu rn itu re p la n ts , fo r example, are uphol s te r e r s , assem bler8 , boring machine o p era to rs, and shaper o p era to rs. In the footw ear p la n ts , examples o f higher paid jo b s are the ou tsid e le a th e r c u tte r s , p u llo v er o p e ra to rs, top s t it c h e r s , and goodyear s t it c h e r s . Lower paid jo b s in the fu rn itu re p la n ts in clu d e cuslons f i l l e r s and glue damp operators and in the footw ear p la n ts , cem enters, heel trim m ers, lin in g and le a th e r d o s e r s , and fo ld e r s . As shown in ta b le s 3 and 5 , the same general p attern i s follow ed by indexes fo r the higher and low er paid occupations in the footw ear p la n ts as i s follow ed by the indexes represen tin g the industry t o t a l s . For both the men and women w orkers, in both higher and low er paid occu p ation s, indexes covering the age groups from 2 5 -5 4 d isp la y o n ly minor v a r ia tio n from one age group to sm other, and a l l are d o s e to the base group index o f 1 0 0 . For most o f the c a teg o ries m entioned, the under 25 group, as w d l as the 2 o ld e st groups (5 5 -6 4 and 65 and o v e r ), have low er indexes than the m iddle groups. This again i s c o n siste n t w ith the in d u stry to ta l fin d in g s . There i s l i t t l e evidence th a t the r d a tio n s h ip between age and output per man-hour d if f e r s g r e a tly between the higher sind low er paid occupations in the footw ear p i suits, except in two age groups. In the 65-an d-over group, both men smd women in the higher paid jo b s show a much more pronounced d e d in e than do th ose in low er psdd jo b s . Comparison o f the v a r ia b ilit y o f output among footw ear workers in higher smd low er paid occupations re v e a ls no d iffe r e n c e o f any s iz a b le p ro p o rtio n s. The average c o e ffic ie n t o f v a ria tio n ( a l l age groups combined) i s alm ost id e n tic a l fo r the upper smd low er paid occupations fo r workers o f both se x e s. F urther, the p attern o f v a r ia b ility -a g e r d a tio n s h ip s i s irre g u la r and u n d e a r ; the tendency fo r the c o e ffic ie n ts o f v a ria tio n to be sm a llest among the medium age groups i s le s s evident than i s the case in the c o e ffic ie n ts fo r the industry t o t a ls . Among the workers in the fu rn itu re p la n ts (ta b le s 4 and 6 ) , the indexes fo llo w a somewhat d iffe r e n t tendency. There i s a s ig n ific a n t d e d in e in the performance o f men workers in the higher paid jo b s in the 45-an d -over groups, whereas the corresponding d e d in e among men in the low er paid jo b s does not appear u n til the next higher age group ( 5 5 -6 4 ) . For women, on the other hand, - 18 - Table 3 . Indexes o f output per man-hour fo r men in c e n tiv e workers in higher and low er paid occupations in 15 men’ s footw ear estab lish m en ts, by age group Lower paid Higher paid Age group Number of workers Under 25 ♦ 59 Index C o e ffic ie n t o f v a ria tio n (p ercen t) 9 4 .8 1 8 .7 Number of workers 39 Index C o e ffic ie n t o f v a ria tio n (p ercen t) 9 0 .4 1 7 .4 2 5 -3 4 . • 212 1 0 0 .5 1 6 .5 66 9 9 .5 1 5 .6 3 5 -4 4 . • 353 1 0 0 .0 1 3 .9 131 1 0 0 .0 1 3 .2 1 5 .3 4 5 -5 4 . • 386 9 7 .4 1 3 .9 74 9 8 .8 55-6 4 . • 271 9 2 .2 1 4 .9 51 9 3 .3 1 2 .9 65 and over 50 7 9 .1 1 7 .6 25 9 7 .9 1 5 .2 Table 4 . Indexes o f output per man-hour fo r men in c e n tiv e workers in higher and low er paid occupations in 11 household fu rn itu re estab lish m en ts, by age group (Age group 35-44=100) Lower paid Higher paid Ace group C o e ffic ie n t o f v a ria tio n (percent) Number of workers C o e ffic ie n t o f v a ria tio n (p ercen t) Number of workers Index Under 25 . 182 9 9 .1 1 6 .8 32 9 6 .5 1 1 .7 2 5 -3 4 . . 361 1 0 0 .9 1 4 .3 75 1 0 3 .3 1 8 .8 3 5 -4 4 . . 279 1 0 0 .0 1 2 .8 93 1 0 0 .0 7 .3 4 5 -5 4 . . 146 9 4 .9 9 .3 72 9 9 .9 1 3 .6 5 5 -6 4 . . 60 9 4 .7 8 .7 36 9 3 .9 1 4 .0 10 9 4 .3 10 9 1 .5 1 1 .6 65 and over 0 /> 1 / Data in s u ffic ie n t to d e r iv e measure. - 19 - Index Table 5 . Indexes o f output per man-hour fo r women in cen tiv e workers in higher and low er paid occupations in 15 men’ s footw ear estab lish m en ts, by age group ( Age__grou.p. 35-44=lQQ) Lower paid Higher paid Age group Number of workers Index C o e ffic ie n t o f v a ria tio n (p ercen t) Under 25 * 72 9 2 .7 1 7 .2 39 9 7 .6 1 5 .3 25-3-4 . • 202 1 0 3 .6 1 3 .5 90 1 0 1 .1 2 1 .0 3 5 -4 4 . • 377 1 0 0 .0 1 7 .8 212 1 0 0 .0 1 6 .6 4 5 -5 4 . • 357 9 9 .1 1 5 .7 177 9 8 .1 1 4 .0 5 5 -6 4 . • 131 9 3 .3 1 6 .4 88 9 5 .5 1 3 .9 22 8 4 .0 2 3 .3 12 9 5 .4 4 .4 65 and ovei Table 6 . Number of workers Index C o e ffic ie n t o f v a ria tio n (p ercen t) Indexes o f output per man-hour fo r women in cen tiv e workers in higher and low er paid occupations in 11 household fu rn itu re estab lish m en ts, by age group (Age group 35 -4 4 = 100) Higher paid Age group Lower paid Number of workers Index C o e ffic ie n t o f v a ria tio n (p ercen t) Under 25 • 16 1 0 6 .8 1 8 .0 2 5 -3 4 . • 61 1 0 5 .9 1 8 .8 3 5 -4 4 . • 55 1 0 0 .0 1 8 .0 4 5 -5 4 . • 34 9 8 .6 5 5 -6 4 . • 13 8 4 .1 rnmmm -- 65 and over Number of workers Index C o e ffic ie n t o f v a r ia tio n (p ercen t) 6 9 2 .5 2 0 .4 18 1 0 9 .7 2 1 .9 42 1 0 0 .0 1 6 .6 1 6 .0 29 9 8 .8 1 6 .0 1 4 .4 — 20 8 8 .1 2 0 .3 1 ( 1 /) ( 1 /) 1 / Fewer than 5 observation s were considered in s u ffic ie n t fo r d erivin g the in d exes. 20 - th ere i s no im portant d iffe re n c e between the p attern s o f the higher and low er paid occu pation s, except fo r the under 25 group. The index fo r t h is group i s con sid erably higher than the base group (1 0 6 .8 ) in higher paid occu pation s, w hile the reverse i s tru e in the low er paid jo b s . In th is la t t e r category, the under 25 group o f women workers shows an average index o f 9 2 .5 . With regard to In d ivid u al v a r ia b ilit y , the r e s u lt s in the fu rn itu re in d u stry are sim ila r to those seen in the footw ear in d u stry . Average coef f ic ie n t s o f v a r ia tio n ( a l l age groups combined) are v ir t u a lly id e n tic a l fo r higher and low er paid occu p ation s. A ls o , c o e ffic ie n t s fo r s p e c ific age groups show no d isc e rn ib le tendency to be re la te d to a g e. In both in d u s tr ie s , as may be seen from ch arts 2 and 3 , the magnitude o f the in d iv id u a l v a r ia tio n s i s s u b s ta n tia l, in d ic a tin g th a t in each o f the occupational c a te g o rie s many workers w ill be found whose performance indexes are higher than the average o f the base group. Machine and Hand O ccupations. Findings are presented sep a ra tely fo r workers in machine and hand occupations in order to a sc e rta in whether d if f e r ences between the requirem ents imposed by the two typ es o f jo b s a f f e c t the variou s age groups d is s im ila r ly . Machine occupations are d efin ed as those in which a powered d evice i s u t i l i z e d . The d is tin c tio n between the two types o f jo b s in v o lv es the notion th a t the p h ysical demands o f machine occupations are l i k e l y to be d iffe r e n t from those o f hand occu p ation s. R esu lts o f t h is a n a ly sis are shown in ta b le s 7 through 1 0 , from which no d e a r p attern em erges. For example, although the men footw ear workers in the 2 o ld e s t age groups (5 5 -6 A and 65 and over) do b e tte r in the machine jo b s , r d a t i v e to th e ir re sp e ctiv e base groups than in the hand occu p ation s, the reverse i s true in the case o f women workers in t h is in d u stry . Among the men fu rn itu re w orkers, th e average index o f the 5 5 -6 4 age group i s higher fo r those in hand occu pation s, w hile fo r those in the 65-an d -over bracket th e machine workers have a higher in d ex. Examination o f the c o e ffic ie n ts o f v a ria tio n re v e a ls some lim ite d d if f e r ences between the hand and machine occu p ation s. For example, the c o e ffic ie n t s o f v a ria tio n (fo r a l l age groups combined) in both in d u str ie s are h igh er fo r hand than fo r machine occupations among the men. For women workers the reverse i s tr u e , w ith employees in the machine occupations showing the g rea ter v a r i a b i l i t y . This su ggests th a t no g e n e ra liza tio n i s warranted concerning the eq u a lizin g e ffe c t o f the machinery used by in d u str ia l w orkers. V h ile t h is e f f e c t may be a s ig n ific a n t fa c to r in some occu p ation s, i t i s evident from th is study th at i t i s by no means u n iv e r s a l. Comparisons o f the a g e -v a r ia b ility p attern s between the hand and machine occupations reveal no d is t in c t tren d s. Although th ere appears to be a s lig h t tendency fo r the c o e ffic ie n t o f v a ria tio n to be higher among the younger and o ld er age groups than in the m iddle groups, the d iffe r e n c e i s rath er sm a ll, as was noted in the comparison between higher and low er paid occu p ation s. 21 - Chart 2. PERCENTOFWORKERSAGE45 ANDOVERINHIGHERPAID OCCUPATIONS WITHOUTPUT PERMAN-HOUR GREATER THAN THEAVERAGEFORAGE GROUP35-44 MEN WOMEN P ercen t "F o o tw e a r P la n t s A ge G roup 4 5 - 5 4 Furniture .P la n t s "F o o t w e a r P la n t s Age G roup 5 5 - 6 4 -< Furniture .P la n ts UNITED STATES DEPARTMENT OF LABOR BUREAU OF LABOR STATISTICS P ercen t Chart 3. PERCENTOFWORKERSAGE45 ANDOVERINLOWERPAID OCCUPATIONSWITHOUTPUTPERMAN-HOURGREATER THANTHEAVERAGEFORAGEGROUP35-44 WOMEN P ercen t Footw ear Plants Age G ro u p -< 4 5 -5 4 V- Footw ear P lants A ge Group 5 5 -6 4 Furniture Plants W B m S i. -< Furniture Plants V- i UNITED STATES DEPARTMENT OF LABOR BUr .AU OF LABOR STATISTICS T a b le 7 . In d e x e s o f o u tp u t p e r m an-hour f o r men in c e n t iv e w o rkers p e rfo rm in g m achine and hand o p e ra tio n s in 15 m en's fo o tw e a r e s ta b lis h m e n ts , by age group (A ge group 35-4.4=100) M achine Age group 2 5 -3 4 3 5 -4 4 . . . . 4 5 -5 4 . , 5 5 -6 4 . . 65 and o v e r T a b le 8 , In d e x C o e f f ic ie n t o f v a r ia t io n (p e rc e n t) Number of w o rkers 228 9 3 .4 9 9 .6 1 6 .9 1 6 .2 17 50 1 0 5 .4 2 0 .3 1 6 .5 419 1 0 0 .0 424 296 9 7 .4 9 3 .0 1 3 .2 1 4 .0 65 36 1 0 0 .0 9 9 .8 1 7 .0 1 6 .0 1 4 .6 26 8 8 .6 1 3 .5 63 8 1 .8 1 6 .6 12 7 5 .8 1 6 .6 Number of w o rke rs U nder 25 • 81 In d e x 9 7 .0 C o e f f ic ie n t o f v a r ia t io n (p e rc e n t) In d e x e s o f o u tp u t p e r m an-hour f o r men in c e n t iv e w o rkers p e rfo rm in g m achine and hand o p e ra tio n s in 11 h o u seh o ld f u r n it u r e e s ta b lis h m e n ts , by age group (A ge group 3 5-44=100) M achine Age group U nder 25 • 2 5 -3 4 . . 3 5 -4 4 4 5 -5 4 Hand . . . . 5 5 -6 4 . . 65 and o v e r Number of w o rke rs 82 199 242 182 72 17 Hand Number of w o rkers In d e x 1 0 0 ,3 1 0 0 .0 9 8 .6 1 1 .0 1 2 .9 8 .9 1 1 .3 132 237 130 36 1 0 0 .5 1 0 3 .2 1 0 0 .0 9 2 .6 1 8 .4 1 6 .4 1 5 .5 9 .0 9 4 .4 9 4 .7 1 1 .2 9 .0 24 9 4 .7 1 3 .3 3 tt/ ) (1 / ) In d e x 9 7 .0 C o e f f ic ie n t o f v a r ia t io n (p e rc e n t) C o e f f ic ie n t o f v a r ia t io n (p e rc e n t) 1 / Few er than 5 o b s e rv a tio n s were co n sid e re d in s u f f ic ie n t f o r d e r iv in g th e in d e x e s . T a b le 9* In d e x e s o f o u tp u t p e r m an-hour f o r women in c e n t iv e w o rkers p e rfo rm in g m achine and hand o p e ra tio n s in 15 m en's fo o tw e ar e s ta b lis h m e n ts , by age group (A ge group 35-44=100) Hand M achine Age group Number of w o rke rs 76 223 398 U nder 25 • 2 5 -3 4 . . 3 5 -4 4 . . 4 5 -5 4 . . 5 5 -6 4 . . In d e x 9 5 .7 C o e f f ic ie n t o f v a r ia t io n (p e rc e n t) Number of w o rkers 1 7 .8 1 9 .0 In d e x 394 167 1 0 2 .4 1 0 0 .0 9 3 .9 9 3 .6 1 5 .9 1 6 .1 35 69 191 140 1 3 .8 52 9 1 .1 1 0 3 .7 1 0 0 .0 9 8 .5 9 5 .2 29 8 5 .7 2 0 .7 5 9 3 .9 65 and o v e r C o e f f ic ie n t o f v a r ia t io n (p e rc e n t) 1 4 .2 9 .0 C l/ ) 1 3 .2 1 3 .6 9 .5 1 / D ata in s u f f ic ie n t to d e riv e m easure. T a b le 1 0 . In d e x e s o f o u tp u t p e r m an-hour f o r women in c e n t iv e w o rke rs p e rfo rm in g m achine and hand o p e ra tio n s in 11 h o u seh o ld f u r n it u r e e s t a b lis h m e n t , by age group jA g e gro u p., ,2 5 -^ = 1 0 0 ) Machin<5 Age group Hand Number of w o rkers In d e x C o e f f ic ie n t o f v a r ia t io n (p e rc e n t) 9 1 0 5 .0 2 5 .6 Under 25 . In d e x C o e f f ic ie n t o f v a r ia t io n (p e rc e n t) 13 9 4 .8 1 6 .1 2 2 .7 52 1 0 4 .2 1 7 .8 1 7 .8 1 7 .7 26 17 1 0 0 .0 1 0 1 .0 1 5 .8 6 .8 1 (1 / ) 0/) 2 5 -3 4 . . 27 3 5 -4 4 4 5 -5 4 . . . . 71 46 1 0 9 .3 1 0 0 .0 9 7 .3 5 5 -6 4 . . 32 8 8 .9 1 8 .6 0/) 0/) 65 and o ve r 1 Number of w o rkers — — — 1 / Few er than 5 o b s e rv a tio n s were co n sid e re d in s u f f ic ie n t f o r d e r iv in g th e in d e x e s. - 25 - Even though th e y a re somewhat ir r e g u la r , th e c o e f f ic ie n t s o f v a r ia t io n in a l l age gro u p s f o r both hand and m achine o ccu p a tio n s ten d to be la r g e . In d iv id u a l v a r ia b ilit y i s a f a c t o r o f m a jo r s ig n if ic a n c e among a l l th e gro u p s o f w o rk e rs s tu d ie d , and th e p erfo rm an ce o f an in d iv id u a l s e le c te d from one o f th e se gro u p s i s l i k e l y to d if f e r m a rke d ly from th e a v e ra g e o f h is g ro u p . As i s e v id e n t from c h a r ts 4 and 5 , s u b s t a n t ia l p ro p o rtio n s o f w o rk e rs in th e h ig h e r age b ra c k e ts f o r both c la s s if ic a t io n s p e rfo rm b e t t e r th an th e a ve ra g e f o r age group 35-44* Th e re was no c o n s is t e n t d iffe r e n c e in th e se p ro p o rtio n s f o r m achine o p e ra tio n s a s c o n tra s te d w ith th o se engaged in hand jo b s . T h is w ide g e n e ra l v a r i a b i l i t y com bined w ith th e sm a ll d iffe r e n c e s in th e a v e ra g e in d e x e s , a s w e ll a s th e ab sen ce o f any c o n s is t e n t d iffe r e n c e s in th e p a tte rn s f o r hand and m achine o c c u p a tio n s , w ould in d ic a t e th a t th e re i s no v a lid b a s is f o r s p e c ify in g d e f in it e age lim it a t io n s f o r c e r t a in c la s s e s o f jo b s . A tte n d an ce A tte n d an ce d a ta were c o lle c t e d f o r 9 ,4 4 0 w o rkers o f whom 4 ,7 8 0 w ere in 11 m en 's fo o tw e a r p la n t s and 4 ,6 6 0 were in 11 f u r n it u r e p la n t s . D epending on th e p la n t s ' re c o rd s , d a ta w ere o b ta in e d f o r p e rio d s ra n g in g from 3 m onths to 1 y e a r . The sam ple f o r t h is in d ic a t o r was la r g e r th a n th a t f o r o u tp u t p e r m an-hour b e cau se h e re i t was p o s s ib le to in c lu d e w o rk e rs p a id on an h o u rly b a s is a s w e ll a s th o se p a id a c c o rd in g to an in c e n t iv e syste m . A s can be seen in t a b le s 11 and 1 2 , w hich show th e a tte n d a n ce f in d in g s f o r th e in d u s t r y t o t a ls s e p a ra te ly by s e x , th e d a ta r e fu t e c u rr e n t id e a s th a t th e re a re s t r ik in g d iffe r e n c e s between age gro u p s a s to r e g u la r it y o f a tte n d a n ce a t w ork. The d iffe r e n c e s between age g ro u p s, h e re , a re e x tre m e ly s m a ll. Among th e fo o tw e a r w o rkers, th e se d if fe r e n c e s do n o t even exceed 1 p e rc e n t f o r th e men and 1 .2 p e rc e n t f o r th e women. In th e f u r n it u r e p la n t s , th e in d e x e s f o r a l l age gro u p s f a l l w ith in 9 8 .6 and 1 0 2 .4 p e rc e n t o f th e base group (a ra n ge o f 3 .8 p e rce n t) f o r men and w ith in 9 7 .1 and 100 f o r women (2 .9 p e r c e n t ). T h e re i s no c o n s is te n t r e la t io n s h ip between a tte n d a n ce and a g e . W ith re g a rd to in d iv id u a l v a r ia t io n s , th e d iffe r e n c e s in in d iv id u a ls ' r a t e s from th e group a ve ra g e in a l l c a se s were in s ig n if ic a n t . The ab sen ce o f any p a tte rn o f th e in d e x e s by a g e , a s w e ll a s th e e x tre m e ly sm a ll v a r ia t io n between age g ro u p s, s u g g e s ts th a t age a s a f a c t o r r e la t in g to w o rk e rs' a tte n d a n ce can b e ig n o re d . A lth o u g h th e re a re no c o n s is te n t d iffe r e n c e s in the p a tte rn o f th e in d e x e s by age f o r men o r women in e it h e r in d u s t r y , th e re a re some d iffe r e n c e s in th e o v e r a ll m easures f o r th e se g ro u p s. The men w o rke rs in both in d u s t r ie s have s l i g h t l y h ig h e r a ve ra g e a tte n d a n ce r a t e s (num ber o f d ays worked p e r 100 sched u le d w orkdays) th an do the women w o rk e rs. In th e fo o tw e a r p la n t s th e r a te f o r men i s 9 8 .3 and f o r women, 9 7 .0 . The r a t e s in th e f u r n it u r e p la n t s a re 9 6 .8 and 9 4 .5 , r e s p e c t iv e ly . A Lso , a s can be se e n , both the men and women in th e - 26 - Chart 4 . PERCENTOFWORKERS AGE45 ANDOVERINMACHINE OCCUPATIONS WITHOUTPUT PERMAN-HOUR GREATER THANTHEAVERAGE FORAGE GROUP35-44 WOMEN MEN P ercen t P ercen t Age Group "Footwear Plants -< 4 5 -5 4 Age Group 5 5 -6 4 Furniture .Plants "Footwear Plants < Furniture .Plants UNITED STA T ES DEPARTM EN T OF LABOR BUREAU OFLABOR STATISTICS m m w W ® m m m Chart 5 . PERCENTOFWORKERS AGE45 ANDOVERINHANDOCCUPATIONS WITHOUTPUT PERMAN-HOUR GREATERTHAN THEAVERAGEFORAGEGROUP35-44 WOMEN MEN P ercen t Age Group 4 5 -5 4 Age Group 5 5 -6 4 "Footwear Plants -< Furniture .Plants "Footwear Plants -< Furniture .Plants UNITED STA T ES D EP A R TM EN T OF LABO R BUREAU OFLABOR STATISTICS P ercen t T a b le I X . In d e x e s o f a tte n d a n ce f o r p ro d u c tio n w o rkers in 11 m en's fo o tw e a r e s ta b lis h m e n ts , by se x and age group (Age group 3 5-44=100) Women Men Number of w o rke rs In d e x Number of w o rkers In d e x U nder 25 . . . . . . . . . 251 9 9 .8 206 1 0 0 .5 2 5 -3 4 3 5 -4 4 ........................................ ........................................ 381 447 420 724 1 0 1 .0 1 0 0 .0 4 5 -5 4 ....................................... 530 9 9 .9 1 0 0 .0 1 0 0 .1 804 1 0 1 .2 5 5 -6 4 ........................................ 65 and o v e r . . .................. 442 396 1 0 1 .2 54 1 0 1 .2 Age group T a b le 1 2 . 9 9 .9 9 9 .7 125 In d e x e s o f a tte n d a n ce f o r p ro d u c tio n w o rke rs in 11 h o u seh o ld f u r n it u r e e s ta b lis h m e n ts , by se x and age group LM s. r e pu p .. 35 t 4 A ^ lQQ) Women Men Number of w o rkers In d e x Number of w o rke rs In d e x U nder 2 5 . ............................... 2 5 -3 4 ........................................ 562 1 ,1 2 2 1 0 1 .3 1 0 2 .4 54 140 9 9 .9 9 8 .9 3 5 -4 4 4 5 -5 4 ....................................... ........................................ 1 ,0 1 0 770 1 0 0 .0 1 0 0 .9 197 148 1 0 0 .0 9 9 .8 5 5-6 4 ....................................... 433 1 0 1 .5 71 9 7 .6 149 9 8 .6 4 (1 / ) Age group 65 and o v e r * • • * • • • 1 / Few er th an 5 o b s e rv a tio n s were co n sid e re d in s u f f ic ie n t f o r d e r iv in g the in d e x e s . - 29 - fo o tw e a r in d u s t r y show h ig h e r a tte n d a n ce r a t e s th an do th e co rre sp o n d in g g ro u p s in th e f u r n it u r e in d u s t r y . B u t a l l o f th e se d iffe r e n c e s a re sm a ll— in no c a se e x ce e d in g 2 .5 d a ys ta ke n o f f p e r 100 d ays sch e d u le d . B ecause i t was b e lie v e d th a t th e atte n d a n ce m o tiv a tio n s m igh t d i f f e r f o r w o rk e rs u n d e r v a r io u s system s o f paym ent, w ith v a r y in g le n g th o f s e r v ic e , o r in h ig h e r o r lo w e r p a id jo b s , and th a t th e se m ig h t a f f e c t age gro u p s d if f e r e n t ly , se p a ra te in d e x e s f o r th e se c la s s if ic a t io n s were d e riv e d . T a b le s 13 and 14- show age group in d e x e s f o r in c e n t iv e w o rke rs and f o r tim e w o rk e rsj t a b le s 15 and 1 6 , f o r w o rk e rs in h ig h e r p a id jo b s and in lo w e r p a id jo b s ; and t a b le s 17 and 1 8 , f o r w o rk e rs w ith l e s s th an 2 y e a rs* s e r v ic e and w ith 2 y e a rs o r more o f s e r v ic e . The in d e x e s f o r each o f th e se gro u p s show no te n d e n cy to v a ry c o n s is t e n t ly w ith a g e . N or do th e in d e x e s show o th e r th an random d iffe r e n c e s in t h e ir p a tte rn s b y age g ro u p . F o r th e m ost p a r t , th e v a r ia t io n s between acre gro u p s in th e se c la s s if ic a t io n s , ju s t a s in th e in d u s t r y t o t a ls , a re s m a ll, b u t th e re a re a few gro u p s w hich do show r a th e r la r g e d if f e r e n c e s . F o r exam ple, th e in d e x e s f o r both men and women w ith le s s than 2 y e a rs* s e r v ic e in th e f u r n it u r e p la n t s v a r ie d b7 1 6 .2 and 1 0 .3 p e rce n ta g e p o in t s , r e s p e c t iv e ly . A lth o u g h no a g e -a tte n d a n ce p a tte rn s emerge f o r th e v a r io u s c la s s if ic a t io n s , some d if fe r e n c e s between th e o v e r a ll r a t e s can be o b se rv e d . As shown in ta b le 19 b o th men and women in c e n t iv e w o rke rs in th e two in d u s t r ie s have h ig h e r a t ten d an ce r a t e s than th e h o u rly p a id w o rk e rs. S im ila r ly , th e w o rke rs w ith 2 y e a rs o r more o f s e r v ic e in a l l c a se s show h ig h e r r a t e s th an th o se w ith le s s th an 2 y e a r s ' s e r v ic e and, w ith one e x c e p tio n , w o rkers in h ig h e r p a id jo b s in both in d u s t r ie s have b e tte r a tte n d a n ce than th o se in lo w e r p a id jo b s . The d if fe r e n c e s betw een th e r a t e s ,w h ile n o t la r g e , a re c o n s is t e n t and s u g g e st th a t th e a tte n d a n ce m o tiv a tio n s f o r th e se v a rio u s c la s s if ic a t io n s do d i f f e r . The ab sen ce o f any a g e -a tte n d a n ce p a tte r n , how ever, in d ic a t e s th a t w h ile th e e f f e c t o f th e c la s s if ic a t io n on th e se m o tiv a tio n s m igh t v a r y , th e y do n o t a f f e c t age gro u p s d if f e r e n t ly . C o n t in u it y o f S e r v ic e D ata on c o n t in u it y o f s e r v ic e were re co rd e d f o r a t o t a l o f 9 ,8 9 8 w o rke rs— 5 ,1 1 2 in th e fo o tw ear p la n t s and 4 ,7 8 6 in the f u r n it u r e p la n t s . T h is in d ic a t o r , th e co n v e rse o f th e s e p a ra tio n r a t e , m easures th e r e la t iv e p ro p o rtio n s o f work e r s in each age group who rem ained on th e jo b d u rin g a su rv e y p e rio d o f 1 y e a r. I t i s in c lu d e d to r e f le c t th e r e la t iv e jo b s t a b i l i t y o f th e v a rio u s age g ro u p s. The number o f w o rkers who rem ained was d eterm in ed b y s u b t r a c t in g th e number o f w o rk e rs who q u it , r e t ir e d , o r w ere d is c h a rg e d ( in c lu d in g m ilit a r y and m e d ica l s e p a ra tio n s b u t no la y o f f s ) from th e number o f w o rkers em ployed a t th e b e g in n in g o f th e o b s e rv a tio n p e rio d . The r e s u lt s o f th e stu d y show th a t th e p e rce n ta g e o f w o rkers who rem ain on the jo b i s h ig h e s t f o r ages 45 th ro u gh 6 4 . (See t a b le s 20 and 2 ]J Both th e yo u n g e st and th e o ld e s t age gro u p s have the h ig h e s t p e rce n ta g e o f se p a ra t io n s . The d e c lin e in th e in d e x e s f o r th e 6 5 -a n d -o v e r group la r g e ly r e f le c t s th e in flu e n c e o f re tire m e n t. F o r th e fo o tw ear p la n t s , t h is in flu e n c e a p p a re n tly - 30 - T a b le 1 3 . In d e xe s o f a tte n d a n ce f o r p ro d u c tio n w o rkers in 11 fo o tw ear e s ta b lish m e n ts , by s e x , method o f wage paym ent, and age group ______________________________ (Age group 3 5-4 4 =100)____________________________ Men In c e n t iv e Age group Number of w orkers Under 2 5 ...................... 168 2 5 -3 4 ....................................... 3 5 -4 4 Tim ework Number of w o rkers In d e x In d e x 83 87 9 9 .9 294 9 9 .7 1 C 0 .1 ....................................... 348 1 0 0 .0 99 1 0 0 .0 4 5 -5 4 ....................................... 426 1 0 0 .3 104 9 9 .7 5 5 -6 4 ....................................... 306 9 9 .9 136 9 9 .8 . ...................... 64 9 9 .8 61 9 9 .5 65 and o v e r 9 9 .3 Women In c e n t iv e Age group Tim ework Number of w o rkers In d e x ...................... 138 1 0 1 .1 68 9 7 .4 2 5 -3 4 ....................................... 301 1 0 0 .6 119 1 0 2 .3 3 5 -4 4 ....................................... 536 1 0 0 .0 188 1 0 0 .0 4 5 -5 4 ....................................... 562 1 0 1 .0 242 1 0 1 .9 5 5 -6 4 ....................................... 278 1 0 1 .3 118 IC O .8 39 1 0 1 .2 15 1 0 1 .1 U nder 2 5 . . . 65 and o v e r . ................... - 31 - Number of w orkers In d e x T a b le 1 4 . In d e x e s o f a tte n d a n ce f o r p ro d u c tio n w o rkers in 11 household f u r n it u r e e s ta b lish m e n ts , b y s e x , method o f paym ent, and age group i i g e g rgup._3S=44=10p). Men Tim ework In c e n t iv e Age group Number of w o rkers In d e x Number of w o rkers In d e x . .................. 425 103. .4 137 1 0 1 .2 2 5 -3 4 ........................................ 852 1 0 2 .8 270 Id .8 3 5 -4 4 ........................................ 74L 1 0 0 .0 2(B 1 0 0 .0 4 5 -5 4 ........................................ 509 1 0 1 .7 26L 9 9 .4 5 5 -6 4 ....................................... 229 1 0 2 .0 204 1 0 0 .5 . ...................... 40 9 6 .7 109 1 0 0 .5 U nder 25 . . . 65 and o v e r Women In c e n t iv e Age group Number of w o rkers U nder 2 5 * In d e x Tim ew ork Number of w o rkers In d e x ............................... 35 9 7 .5 19 1 0 3 .9 2 5 -3 4 ....................................... 96 9 8 .6 44 9 9 .5 3 5 -4 4 ....................................... 142 1 0 0 .0 55 1 0 0 .0 4 5 -5 4 ....................................... 103 46 9 9 .4 9 5 .9 1 0 0 .6 2 (1 / ) 45 25 2 5 5 -6 4 ....................................... 65 and o v e r . . . . . . . 1 0 0 .0 (1 / ) 1 / Few er then 5 o b s e rv a tio n s w ere co n sid e re d in s u f f ic ie n t f o r d e r iv in g th e in d e x e s . 32 - Table 1 5 . Indexes o f attendance fo r production workers in higher and lower paid occupations in 11 footwear establish m ents, by sex and age group (Age group 35-44=100) Men Lower Higher Age group Number of workers Index Number of workers Index ................................. 84 9 9 .7 167 9 9 .8 2 5 -3 4 .......................................... 212 1 0 0 .0 169 9 9 .8 3 5 -4 4 .......................................... 252 1 0 0 .0 195 1 0 0 .0 4 5 -5 4 .......................................... 312 1 0 0 .0 218 1 0 0 .3 5 5 -6 4 .......................................... 193 9 9 .1 249 1 0 0 .7 65 and over................................. 34 9 9 .1 91 1 0 0 .3 Under 2 5 . Women Lower Higher Age group Number of workers . 2 5 -3 4 Index Number of workers Index 109 1 0 0 .7 97 1 0 0 .1 .......................................... 257 1 0 0 .4 1 0 1 .9 3 5 -4 4 .......................................... 458 1 0 0 .0 163 266 4 5 -5 4 .......................................... 517 1 0 0 .5 287 1 0 2 .4 5 5 -6 4 .......................................... 221 1 0 0 .5 175 1 0 2 .3 32 1 0 0 .4 22 g- , ■.... ......—■ 1 0 3 .1 Under 2 5 . . . 65 and over ................... . . . . . . . - 33 - 1 0 0 .0 Table 1 6 . Indexes o f attendance fo r production workers in higher and lower paid occupations in 11 household fu rn itu re establish m ents, by sex and age group (Age group 3 5 -4 4=l .QP ). Men Under 25 • ............................ Lower Higher Age group Number of workers Index Number of workers Index 301 1 0 0 .3 26L 1 0 2 .7 2 5 -3 4 .......................................... 651 1 0 0 .6 471 1 0 5 .1 3 5 -4 4 ........................................... 583 1 0 0 .0 427 1 0 0 .0 4 5 -5 4 ........................................... 434 1 0 0 .3 336 1 0 1 .9 5 5 -6 4 ........................................... 206 9 9 .8 227 1 0 4 .0 63 1 0 0 .8 86 9 6 .6 65 and over ....................... .... Women Higher Age group Under 25 ...................................... Lower Number of workers Index 41 9 9 .8 13 9 9 .9 Number of workers Index 2 5 -3 4 .......................................... 109 9 9 .2 31 9 8 .4 3 5 -4 4 ........................................... 124 1 0 0 .0 73 1 0 0 .0 4 5 -5 4 ........................................... 90 1 0 0 .2 58 9 9 .1 5 5 -6 4 .......................................... 34 9 7 .2 37 9 8 .5 2 ( 1 /) 65 and over . . ................... 2 ( l/) L----- - ------------ l / Fewer than 5 observations were considered in s u ffic ie n t fo r derivin g the in d exes. Table 1 7 . Indexes o f attendance fo r production workers in 11 footwear establish m ents, by sex, years o f s e r v ic e , and age group (Age group 35-44=100) Men 2 y e a r s' service or more Age group Under 25 ................................. • • • • • • # # • 55“ ^4 65 and over . . . . . . Less than 2 y e a r s1 serv ice Number of workers Index Number of workers Index 108 9 9 .6 143 1 0 0 .9 330 9 9 .8 51 1 0 1 .1 410 1 0 0 .0 37 1 0 0 .0 500 1 0 0 .0 30 1 0 2 .1 411 9 9 .6 31 1 0 2 .8 118 9 9 .8 7 9 7 .5 Women Age group Under 25 ................................. 2 y e a rs' serv ic e or more Less than 2 y e a r s' serv ice Number of workers Index Number of workers Index 97 1 0 0 .4 109 1 0 1 .3 2 5 -3 4 ...................................... 337 1 0 1 .3 83 9 8 .7 3 5 -4 4 ...................................... 627 1 0 0 .0 97 1 0 0 .0 4 5 -5 4 ...................................... 729 1 0 1 .1 75 1 0 1 .8 5 5 -6 4 ...................................... 363 1 0 1 .0 33 1 0 3 .3 51 1 0 1 .0 3 65 and over . . . . . . 0 /) \J Fewer than 5 observations were considered in s u ffic ie n t fo r d eriving the indexes. - 35 - Table 1 8 . Indexes o f attendance fo r production workers in 11 household fu rn itu re establishm ents, by sex, years o f se r v ic e , and age group (.Age group. .35r:44=lQ0l Men Age group Less than 2 y e a rs' service 2 y e a r s' serv ice or more Number of workers Index Number of workers Under 25 . . . . . . . 235 9 9 .5 327 2 5 -3 4 ................................. 851 1 0 0 .3 271 1 0 7 .4 1 1 0 .1 3 5 -4 4 ................................. 870 1 0 0 .0 140 100.0 4 5 -5 4 ................................. 693 9 9 .8 77 1 0 6 .5 5 5 -6 4 ................................. 404 9 9 .0 29 1 1 6 .2 144 9 7 .3 5 1 1 4 .2 65 and over . . . . . Index Women 2 y e a rs' se rv ic e or more Less than 2 y e a r s' serv ice Age group Number of workers Index . Number of workers Index Under 25 • ........................ 22 1 0 1 .0 32 9 6 .0 2 5 -3 4 ................................. 98 1 0 1 .1 42 9 0 .9 3 5 -4 4 ................................. 150 1 0 0 .0 47 1 0 0 .0 4 5 -5 4 ................................. 136 9 9 .6 12 1 0 1 .2 5 5 -6 4 ................................. 67 9 7 .3 4 65 and over ( 1 /) — . . . . . 4 C l/) — 1 / Fewer than 5 observations were considered in s u f f ic ie n t fo r derivin g the indexes. 36- Table 1 9 . Attendance ra tes \J fo r production workers in 11 men's footwear and 11 fu rn itu re establish m ents, by sex, method o f payment, len gth o f se r v ic e , and occupational pay le v e l Industry Men's footwear G la s s ific a tio n Furniture Men Women Men Women £8*2 £2*0 £6*8 £4*2 9 8 .4 9 8 .0 9 7 .2 9 6 .4 9 6 .9 9 6 .7 9 5 .4 9 3 .0 Less than 2 years ........................ 9 7 .4 9 5 .5 9 5 .3 9 4 .0 2 yeers and more 9 8 .4 9 7 .3 9 7 .3 9 4 .6 9 8 .5 9 8 .1 9 7 .4 9 6 .4 9 7 .1 9 6 .5 9 4 .2 9 5 .2 T otal ........................................................ Method o f navment Incen tive . • ................................. Time* • • • • • • • • • • • « Lensrth o f service • ................... Occuoational Dav le v e l Higher paid ..................................... Lower paid . . ............................. 1 / Number o f days worked per 100 days scheduled. - 37 - i s not as great as the in flu en ce on the youngest age group o f other fa c to r s causing sep aration s. The indexes fo r the youngest age group are markedly lower than those fo r the o ld e s t age group. Although the c o n tin u ity -o f-s e r v ic e ra te s f o r age group 5 5-6 4 are gen erally higher than those fo r the 65-and-over group, there i s a rather sharp d e clin e in the number o f workers in the l a t t e r group. I f the d is tr ib u tio n s o f workers by age shown in ta b le s 20 and 21 are gen erally rep resen tative o f the actual s it u a tio n over a period o f y ea rs, then the d eclin e in the number o f workers in the o ld e s t age group r e f l e c t s in part the continuing in flu en ce o f separations in the 5 5 -6 4 age group without replacement by workers in the same age group. Even though the separation rate i s low est fo r th is age group, the fewer accession s (r e s u ltin g from fewer people in t h is age group reentering the la b o r fo rce once they le a v e as w ell as a fa ilu r e on the p art o f employers to h ire workers in th is age group) le a d s to the sharp d eclin e in the number on the job a t 6 5 . In ad d ition to the fin d in gs by sex and ind ustry, r e s u lts are shown sepa r a te ly f o r those employees with l e s s than 2 y e a r s' se rv ic e and fo r those with 2 or more years o f se r v ic e . This l a s t d is tin c tio n was made on the assumption th at a r e la t iv e ly newly hired group o f workers may show a c o n tin u ity -o f-s e r v ic e p attern d iffe r e n t from a group which had worked in the p lan t a longer p eriod , and that th is d iffe r e n c e , i f not taken in to account, might in flu en ce the actual age-performance r e la tio n s h ip s . Examination o f ta b le s 22 and 2 3, which present con tin u ity indexes by len g th o f s e r v ic e , rev ea ls fo r the most part the same p attern as the industry to ta ls . The highest indexes were generally recorded f o r age groups 45 through 6 4 , and the low est fo r the youngest and o ld e st age groups. However, the d i f feren ces in the rates between age groups are gen era lly la r g e r f o r workers with l e s s than 2 y e a r s' s e r v ic e . Although the age-performance patterns fo r the workers c l a s s i f i e d by len gth o f service are sim ila r , the average c o n tin u ity -o f-s e r v ic e rates do show wide d iffe r e n c e s . Table 24 shows the c o n tin u ity -o f-s e r v ic e ra tes fo r a l l workers stu d ied , grouped by sex, in d u stry, and len gth o f s e r v ic e . (Note th at these are actual rates rather than in d ex es.) In gen eral, the workers with 2 or more years o f service show considerably higher ra tes them those with l e s s than 2 y e a rs' s e r v ic e . A lso , the d iffe r e n c e between these two groups i s more pronounced fo r men than i t i s fo r women. Among the women fu rn itu re workers, the ra te s fo r the two groups are very d o s e . On the other hand, the r e su lts show th at the c o n tin u ity -o f-s e r v ic e ra te s fo r both men and women (over a ll age groups) in the two in d u strie s do not d i f f e r a p p reciably. The ra tes are id e n tic a l in footwear p la n ts and show a d iffe r e n c e o f s l i g h t l y over 1 percent fo r the fu rn itu re p la n ts . - 38 - Table 2 0 . Indexes o f con tin u ity o f serv ice fo r production workers in 11 footwear establishm ents, by sex and age group Women Hen Age group Number of workers Index Number of workers Index Under 25 • • ........................ 245 8 3 .9 238 8 8 .2 2 5 -3 4 ...................................... 418 9 6 .7 438 9 7 .6 3 5 -4 4 . . . . ................... 478 1 0 0 .0 749 1 0 0 .0 4 5 -5 4 ...................................... 589 1 0 1 .5 838 1 0 2 .1 5 5 -6 4 ...................................... 509 1 0 0 .4 393 1 0 1 .4 158 9 0 .4 59 65 and over iJ . . . . . . 1/ 9 7 .5 Includes only employees with more than 2 y e a rs' s e r v ic e . Table 2 1 . Indexes o f continuity o f service fo r production workers in 11 household fu rn itu re establishm ents, by sex and age group __ (Age group 35-44=100 j Women Men Age group .... Under 2 5 . ........................ .... Number of workers Index 561 8 7 .7 58 6 1 .1 Number of workers Index 2 5 -3 4 ...................................... 1 ,1 7 9 9 4 .1 141 9 3 .0 3 5 -4 4 ...................................... 1 ,0 1 7 1 0 0 .0 206 1 0 0 .0 4 5 -5 4 ...................................... 783 1 C 1 .3 152 5 5-6 4 ...................................... 440 1 0 3 .7 168 8 7 .6 65 and over \J 0 . . . . . 1 0 2 .3 75 1/ 1 0 5 .1 6 1/ 4 7 .7 Includes only employees with more than 2 y e a r s' s e r v ic e . - 39 - T a b le 2 2 . In d ex es o f fo o tw e a r c o n tin u ity o f e s ta b lis h m e n ts , ___________________________________ (A g e s e r v ic e fo r b y y e a rs grou p sex* p r o d u c tio n o f w ork ers s e r v ic e , and in age 11 grou p 35-44=100)_________________________________ M en 2 A ge y e a r s ’ s e r v ic e o r m ore L e ss th a n 2 y e a r s ' s e r v ic e grou p N um ber H um ber o f | o f In d ex In d ex w ork ers w ork ers • 12 0 8 4 .0 12 5 8 3 .0 2 5 -3 4 ................................... 364 9 8 .8 54 8 0 .3 3 5 -4 4 ................................... 447 1 0 0 .0 31 1 0 0 .0 4 5 -5 4 ................................... 560 1 0 1 .2 29 1 0 3 .9 5 5 -6 4 ................................... 483 1 0 0 .0 26 1 0 3 .9 65 o v e r 1 4 8 9 5 .1 1 0 5 4 .1 U nder 25 a fid • * * • . • • W om en 2 A ge y e a r s ' s e r v ic e o r m ore o f In d ex w ork ers 25 th a n 2 y e a r s ' s e r v ic e N um ber Som ber U nder L ess grou p • • • • n • o f In d ex w ork ers o 8 7 .7 1 2 8 9 1 .3 2 5 -3 4 ................................... 353 9 6 .3 85 1 0 4 .3 3 5 -4 4 ................................... 642 1 0 0 .0 107 1 0 0 .0 4 5 -5 4 ................................... 7 6 1 1 0 0 .0 77 1 1 3 .0 5 5 -6 4 ................................... 363 1 0 0 .0 30 1 0 8 .6 65 o v e r 55 9 7 .5 4 and 1 / th e F ew er * • • th a n 5 o b s e r v a tio n s w ere c o n s id e r e d in d e x e s * - 40 - in s u ffic ie n t (1 /) fo r d e r iv in g T a b le 2 3 . In d ex es o f h o u s e h o ld and age c o n tin u ity fu r n itu r e o f s e r v ic e fo r e s ta b lis h m e n ts , p r o d u c tio n by s e x , w ork ers y e a rs o f in 11 s e r v ic e , grou p (Age group 35-44=100) M en 2 A ge y e a r s ' s e r v ic e o r L ess m ore th a n 2 y e a r s ' s e r v ic e grou p N um ber N um ber o f In d ex In d e x o f w ork ers w ork ers 258 8 8 .8 303 8 2 .7 923 9 5 .0 256 8 9 .9 • 900 1 0 0 .0 1 1 7 1 0 0 .0 • 717 1 0 1 .2 66 1 0 1 .4 5 5 -6 4 ............................ 417 1 0 1 .2 23 1 1 4 .4 65 o v e r 1 6 2 9 5 .0 6 5 5 .3 U nder 25 . . . 2 5 -3 4 • • s e 3 5 —4 4 • • • • 4 5 —5 4 • • e e and . e • • W om en \ 2 A ge grou p y e a r s ' s e r v ic e o r m ore L ess th a n 2 y e a r s ' s e r v ic e num ber N um ber o f In d ex w ork ers o f In d e x w ork ers _ U nder 25 • ♦ • • 33 5 9 .1 25 7 0 .0 2 5 —3 4 • • • • # 108 9 3 .6 33 9 0 .4 3 5 -4 4 • • • # • 1 6 1 1 0 0 .0 4 5 1 0 0 .0 4 5 —5 4 • • e e e 1 4 0 1 0 5 .1 1 2 1 0 2 .3 5 5 -6 4 • • e • • 71 1 0 5 .1 4 65 o y e r • # 6 4 7 .7 and 1 / th e F ew er th a n 5 o b s e r v a tio n s w ere in d e x e s . - a c o n s id e r e d 0 / ) — in s u ffic ie n t — fo r d e r iv in g Tafale 2 4 . C o n tin u ity -o f-se r v ic e ra tes 1 / fo r production workers in 11 men’ s footwear and 11 fu rn itu re establish m ents, by sex and length o f serv ice Industry ---------------- K;n ’ s Fool iwear Length o f service Total ................................................................. • Furniture Men Women Men Women 9 5 .2 9 5 .2 9 0 ,4 9 1 .6 Less than 2 y e a r s ’ service . . . . 8 0 .8 8 4 .4 7 4 .8 9 0 .4 2 y e a r s ' serv ic e and more . . 9 6 .4 9 6 .4 9 4 .0 9 1 .6 . . . \J Number o f workers per 100 who remained on the job during a survey period o f 1 y e a r . APPENDIX I : Questionnaire and Worksheets UNITED STATES DEPARTMENT OF LABOR Bureau o f Labor S t a t i s t i c s Washington 2 5 , D. C , Job Performance and Age Industry Plant Address Parent company__________________ ___________________________ Address__________ O f f i c i a ls interview ed: (Cross out "C o ." or "P la n t") Name Ti t l e (C o .) (P lan t) Name Ti t l e (C o .) (P lan t) Survey made by___________ _________________________________________ Dates_____ Dates The data submitted on t h is schedule w i l l be seen only by sworn employees o f the Bureau o f Labor S t a t i s t i c s , The data w i l l not be released in ary form which permits id e n tific a tio n with any s p e c ific company without w ritten permission,, A, General Information 1, What are the p rin cip a l products manufactured in t h is p lan t and t h e ir p rice range ?__________ ____________________________________ ____ 2, During normal production periods approximately (a ) how many production workers are employed? , (b ) how many o f these workers are paid according to in d iv id u a l in cen tive ra te s? • THE FOLLOWING QUESTIONS REFER TO PRODUCTION WORKERS ONLY B. Plant Records 3* For what periods (during the f i r s t h a lf o f 1956) was t h is p lan t operating on a f u l l production schedule? ( I f these periods vary by department, s p e c ify ) U. For what departments do you maintain records which w i l l y ie ld data on output per man? (As measured by incentive earnings or p h y sic a l u n its ) 5. B r i e f ly d escribe these records 6. Which o f the fo llo w in g are u t i l i z e d in determining piece ra tes or other in d iv id u al incentive r a t e s : a* Time stu d ies Yes r j no b* Employee-management conference Yes r j n or c• Other ( spec i f y )_______________________________ o ’ j 7. For what incentive jobs have maximum production lim it s been e sta b lish e d ( l i m i t s such as maximum number o f u n its paid fo r per hour)? 8. Do you maintain sp e c ia l records o f employee absences?r 9. I f y e s , d escribe________________________________________________________________________ Yes / / No / J 10. For employees who have l e f t durin g the p ast y e a r , do you record the date on which they l e f t , Yes / / No /~ ~ 7 » and the reason (q u it , lack o f work, e t c 0)? Yes / / No / ~ f 11. Are employees on leave o f absence because o f i l l n ess or other reasons marked on your records as separated? Yes / / No f~ 7 I f no, describe how they are marked - u C. Placement Procedures Do your procedures fo r p lacin g production workers include the follow in g* 12. I n i t i a l p h ysical examination Yes / ~ 7 No / / 13. Periodic p h ysical examination Yes / No / / / I f y e s , give time in te r v a l between p h y sica l checkups___________________________ 1U. 1$. Aptitude te s tin g Yes / 7 No / / A n alysis o f p h ysical job requirements; a. W ritten d escrip tio n o f p h ysical requirements ( such as working p o s itio n , p h y sic a l e x e r tio n , v is io n , e t c .) Yes / / No / / b. Inform al evaluation Yes / 7 No / / 16. Other, s p e c ify 17. Have any measures been taken to review and redesign jobs held by old er workers? Yes / j No / / I f y e s , give examples, and i f p o ssib le in d icate reasons fo r in i t ia t in g these measures 18. Have there been instances o f r etra in in g or reassignment o f older workers? Yes / / No / / I f y e s , give examples, including reasons fo r these s t e p s . D« Earnings Adjustments 19. Do you compute, fo r in d ivid u al worker s , t h eir average hourly piecework earnings over certa in periods? Yes / / No / 7 I f y e s , in d icate p e rio d s. 20. How i s the vacation pay fo r incentive workers determined?____________ 21. Overtime rates become e ffe c t iv e a ft e r ____ hours per day, 22. What i s the premium rate fo r overtime work____, Sunday or holiday work 23. I f these rates o f extra pay do not apply to piecework, sp ecify the premium rates on piecework fo r overtime and holid ays__________________________________ 2U. Dates o f vacation periods o f p la n t_________________ - A5 - _____ hours per week. Job Performance and Age General In stru ctio n s fo r Completing Questionnaire The questionnaire and worksheet on Job Performance and Age are to be used to c o l l e c t data pertain in g to p ro d u c tiv ity , absenteeism, and separations o f in d iv id u a l production workers in the p lants stu d ied . These data are to be c o lle c te d together with employees* ages and other p ertinent personal in fo r mation. The questionnaire r e la te s to general inform ation, plant record s, placement procedures, and earnings adjustments (se c tio n s A through D ). The worksheet i s used fo r recording persbnnel information and data on output, attendance, and separations o f an employee. The recommended procedure i s to f i l l out the questionnaire through sectio n D during the i n i t i a l interview with the plant o f f i c i a l , then to record the in d ivid u al employee data on the worksheets. Question n aire In stru c tio n s are included only fo r those items not considered s e l f explanatory. I f a d d itio n a l space i s needed fo r answers, use back o f sh e e ts . 1. P rin cip al Products — In d icate p rin c ip a l products manufactured in tern s o f broad product c la s s e s , such as Men's Goodyear Welt Shoes, wooden c h a ir s . Designate p rice range at r e t a i l p ric e s fo r an in d ica tio n o f general q u a lity range o f the p la n t 's production. 2. Employment — a . Enter approximate number o f production and r e la te d workers in t h is p la n t. These include workers (up through the working fo r e man l e v e l ) engaged in fa b r ic a tin g , p rocessin g, assem bling, in sp ectio n , re c e iv in g , storage, handling, packing, warehousing, shipping (but not d e liv e r y ), maintenance, r e p a ir, ja n it o r a l, watchman s e r v ic e s , product development, a u x ilia ry production fo r p la n t 's own u se, recordkeeping and other serv ic es c lo s e ly associated with these production op eration s. b . Number o f workers on in d ivid u al (not group) incentive pay. 3. Production Period — Obtain dates fo r 8 weeks (not n e c e ssa r ily consecutive) when p lant was operating at f u l l production. U ,5 . Output Per Man Records — P ayroll records which y ie ld s tr a ig h t piecework earnings o f production workers separate from timework earnings w i l l furnish output per man fig u r e s . I f such p a y ro ll records are not a v a ila b le , production records from which in d iv id u a l str a ig h t-tim e piecework earnings can be derived are s a t is f a c t o r y . S traigh t-tim e piecework earnings can be computed from records which show the q u a n titie s produced at given piece rates in given numbers o f hours worked by in d ivid u al workers. In other c a se s, output data may be a v a ila b le in terms o f time standards• - -4 6 - In some c a ses, output data may be recorded a lso fo r timeworkers. Inquire whether these data are a v a ila b le * 6. 8 ,9 . Incen tive Pay Rates - - The purpose o f th is question i s to ascertain whether in cen tive ra tes r e f l e c t d ifferen c es in the time required to perform the same operation on d iffe r e n t models. I f they do, workers performing the same operation can be compared regardless o f d iffe re n c e s in t h e ir product mix, Absenteeism Records — In d ica te whether any sp e cia l records maintained by the establishm ent w i l l furnish information on time scheduled and time worked. Scheduled days and days worked may have to be obtained from time cards d ir e c t ly . Determine whether time cards are a v a ila b le fo r f u l l pro duction periods o f a t le a s t 8 weeks, p referably during the f i r s t h a lf o f 1956. 10. Describe nature and lo c a tio n o f such records. 11. I f i l l or absent employees are not marked as "s e p a r a te d ," determine what entry i s used, fo r instance "temporary term ination" or no e n tr y . 16. Other Placement Procedures — Indicate any steps not mentioned above which are employed in p lacin g production workers. 20. Vacation Pay — Record how the hourly rate during paid vacations i s determined f o r p ie c e -r a te and other ind ivid ual incentive workers. 2 1 ,2 2 . 2U. Premium Rates — I f the premium rates vary by job c l a s s i f i c a t i o n , l i s t them on back 'o f the sh e e t• Vacation Periods — I f vacation periods vary by department, l i s t the department and relevant p eriod s. - -47 - CLO CK NUM BER P LA N T 1 1 1 1 1 1n DEPT PLANT MM D A T E CODE D E P A R T M E N T CODE NAME OF E M P LO Y E E f s R B o r n k AGE A S P S T a te r = t L------------------ 1 FI HAL PPS o c c u p a t io n PREU H M A CH IN ER Y U SED O C C U P A T IO N C O O ES P E R IO D END \N6 T O P E R A T IO N D E S C R I P T I O N P IE C E W O R K E A R N 1NOS _______ HO OPS_________ S T R A IG H T T 1 MEVIORK O V E R T ! ME TO TAL T\M C * 3" * 8 4 n II i 1. c AVfcNRff’E ' U r i iK= 1 P r l Sgp . H O U R LY E A R N IN 6 S L - * !M=3. T s 3 . JANTO CP DATE years’ Se r v i c e 1 1 YEAR^ 9C61MHE1 E% PER rem arks: | v iV if RATE 7 ft TO TAL ;^ / / / // _ A D JU ST M EN T .A D J U S T E D TOTAL '/ / / // ' 's s t f 's t t '''s /// /'V V /f /* / /f /f.ss / / /f/f S' * ' *r S * sSsS/S S ^ ,ss LEARN T IM E ABSEN TEEISM " f C o c 'K ___NO- 1>'AYS SC HE. P . PAY W EEK V/EEK OF YEAR 1 z 3 4 S <0 7 « 4 10 l i !Z M O N D AY T u esd a y W ED N ESD A Y TH U RSPA X F R ID A Y I»1 IS lb 17 IS 14 20 2.1 22 23 24 2S 26 d a Y r y y rr 1 n z t 24 3o 31 i t 33 34- 35 3b 37 3S 34 40 41 4 * 43 44 45 46 47 4* 44 SO 51 52 s lAMS-iill MONDAY TU ESD AY ; We d n e s d a y THURSDAY FR ID A Y P ates h ired DATES SEPARATED X; "t's s / /-/ \ " /> / /✓ V /, / '/ / '/ s "^✓ '//, ' */ '' s S '/ ' '/ ' s s tt«cM "REMARKS- > IDATE. F\«VT H'PfED Se r a r a i e d % Job Performance and Age In stru ctio n s fo r the Worksheet Clock Somber — Record d o c k number l i s t e d on the p a y ro ll or the w orker's tim ecard. I f th is number d i f f e r s from the number on the personnel record s, note t h is in "Remarks" space. P lant Code — Record p lan t code from q u estion n aire. Dent. Code — Record department number as l i s t e d by the p la n t. I f p lan t does not number departments, use a rb itra ry code and record code on p lan t qu estio n n a ire. Name n f Employee — Record f i r s t and l a s t name. M = 1 . F = 2 — In d icate sex by t h is code. Date Born Age — Record l a s t two d ig it s o f year o f b ir t h . Do not f i l l in a t p lan t u n less convenient. AS) ) — Do not f i l l PS) in a t p la n t. PPS — P relim . Prod. Sample •— Check i f workers belong to a prelim inary p ro d u c tiv ity sample group. Such a group c o n sists o f a t l e a s t two in cen tive workers o f the same sex performing the same o p era tio n s. For groups which f a i l to f u l f i l l t h is con d ition , do not take down p ro d u c tiv ity inform ation. Occupation T i t l e — L i s t occupation t i t l e (and grade, i f used) o f the worker ( e . g . , "c h a ir assem bler") . Prelim inary Occupation Code — Where a v a ila b le , record occupation number as used by establish m en t. Otherwise, leave blank. Final Occupation Code — Do not f i l l in a t p la n t. Operation D escrip tion — Describe the s p e c ific operation or operations performed, including where relevan t the names o f the products turned o u t. T his should be in s u ffic ie n t d e ta il to determine whether the employee's a c t i v i t y i s comparable with th at o f o th e r s. I f standard operation numbers are a v a ila b le , l i s t them in t h is space. This inform ation may be obtained from production records (p a r tic u la r ly operation numbers) and from the foreman. In doubtful ca ses, confirm from the appropriate p lan t o f f i c i a l th at a l l workers in an occupation a c tu a lly perform the same types o f op era tio n s. 50 - Machinery Used — L is t the type o f machines or to o ls u sed . W rite in "autom atic fe e d " i f the machine employs an autom atic feed and discharge mechanism fo r the work p ie c e s , so th a t the in d ivid u al work p ie c e s do not have to be in se rted and removed by hand. AHE — Average Hourly Earnings o f in cen tiv e workers o n ly . I f average hourly p ie c e -r a te earnings fo r the p ro d u c tiv ity period have already been computed by the company, record in t h is c e l l . O therw ise, le a v e blan k . P = 1 . L = 2 — Do not f i l l in a t p la n t. H o r M — Hand or Machine — In d ica te by " 1 " or " 2 " whether worker i s prima r i l y a hand or a powered machine o p era to r. ( " 1 " = Hand, " 2 " = machine; P = 1 . T = 2 — P iece or Time Work — In d ica te by " 1 " i f worker i s paid by p ie c e ra te and " 2 " i f worker i s paid on a tim e b a sis (h ou rly r a t e ) . I f a worker i s on both p ie c e and time work during the rep o rtin g p erio d , in d ica te by " 1 " ( i . e . , l i s t as p ie c e -r a te w orker). Sep. Sample — Do not f i l l in a t p la n t. Esrtgfle. pgfflprfc.JwnlBga*.. JOBTg — J*> not f i l l in i f the p la n t c a lc u la te s average hourly piecework earnings fo r a f u l l production period o f A weeks or more. For the rep o rtin g p eriod p r ia te company o f f i c i a l , during the f i r s t h a lf o f a v a ila b le , u se a minimum on p r o d u c tiv ity , s e le c t , w ith the aid o f the appro a f u l l production period to t a lin g 8 weeks, p re fe ra b ly 1956 (n ot n e c e ssa r ily c o n se c u tiv e ). I f t h is i s not o f U weeks fo r the p eriod stu d ie d . Period Ending — Record in spaces 1 to 8 number o f the month and day and on ly the l a s t d ig it o f the year fo r the ending d ates o f th e relev a n t p e rio d . I f p ro d u c tiv ity inform ation i s a v a ila b le on the b a sis o f a 2-week pay p e rio d , record p ro d u c tiv ity inform ation in every other row, i f i t i s m onthly, record on every fou rth row. Straigh t-T im e Piecework Earnings - - Record s tr a ig h t-tim e piecework earn in gs. Include d o lla r s and cen ts w ith .0 0 i f no c e n ts. O vertjpg — Record overtim e piecework earn in gs. Do not a d ju st to s tr a ig h ttim e l e v e l s . Where records have overtim e earnings already adjusted to s tr a ig h t-tim e r a te s , in d ica te in "Remarks" sp ace. Record both d o lla r s and c e n ts . T o ta l Hours - - Record the t o t a l number o f hours worked during the tim e p eriod stud ied to n earest quarter hour. Mark quarter hours as d ecim als, i . e . , .2 5 or .5 0 . I f to ta l piecework hours are a v a ila b le , record in th is space and in d ic a te t h is fa c t in the "Remarks" space. 51 - — Record the to ta l number o f hours fo r which the worker was paid on a tim e b a sis (h ou rly r a te ) during the period stu d ied . I f to ta l piecework hours are recorded in the to ta l hours column, le a v e t h is column blank. C. P . Date — Current P a y ro ll — Record the month and day o f the pay period n earest to the tim e o f the p la n t v i s i t , i f the employee was on the p a y r o ll. I f the employee was not on the p a y r o ll, le a v e blan k . T his inform ation i s not needed i f complete data on separations are conveniently a v a ila b le . Y ears* S ervice — Do not f i l l in a t the p la n t. Taapff* Exp erien ce — Determine whether inform ation i s a v a ila b le on experience on sim ila r jo b s in the same company, or in previous employment and record . Beginner — Do not f i l l in i f years o f experience are obtain ed . I f years o f experience are n ot a v a ila b le , obtain id e n tity o f beginners from the forem an. frearnjnp Tjjqe — Obtain from production manager or foreman inform ation on average le a rn in g period required fo r an inexperienced employee to a tta in average e ffic ie n c y in h is occu pation . For each group o f employees working a t the same occu pation , t h is inform ation needs to be entered on one card o n ly . In d ica te whether tim e i s sta ted in months or y e a rs, e . g . , "6 m o ." or nU y e a r s ." Tlmeunpk — Record here the wage ra te paid to employees on timework. This in clu d es workers p a r tly on in cen tiv e work and p a r tly on timework. For incen t iv e workers fo r whom a wage ra te i s not a v a ila b le , determ ine whether the p la n t m aintains a record o f th e ir ra te o f pay per minute (o r hour) o f standard tim e and en ter in t h is sp ace. Absenteeism Cal pndnp — I f a convenient absenteeism record i s m aintained in the p la n t fo r each employee, record absenteeism fo r the year 1956 up to the time o f p la n t v i s i t , i f a v a ila b le . I f absenteeism has to be recon structed from tim ecards, record absenteeism fo r a period o f 8 to 12 weeks, covering the 8-week f u l l production period used fo r the measurement o f p ro d u c tiv ity . In d ica te by "A " on a m aster card the f i r s t and l a s t d ates o f the absen teeism survey p e rio d . A lso in d ica te on t h is card, by marking M- B in the appropriate c e l l , the h olid ays and weekdays which the e n tir e p la n t was shut down. On th e calen d ar, fo r each employee mark by " 0 " each workday m issed by the em ployee. Do not mark an employee as m issing ( " 0 " ) on days on which you know th a t the e n tir e p la n t was shut down. Clock Mo. — Repeat clock number from other sid e o f card . Days Scheduled. Days A bsent. A ttend. Rate — Do not f i l l in a t p la n t. - 52 Dates Hired — Record o r ig in a l date hired and subsequent d ates rehired from beginning o f absentee period to time o f p la n t v i s i t . Include day, month, and year i f a v a ila b le . Dates Separated — Record a l l separation d ates (by day, month, and year) from beginning o f absenteeism period to tim e o f p la n t v i s i t . Reason — R efers to reason the worker l e f t . Try to ob tain reason fo r each in cid en t and in d ica te by the fo llo w in g coding: L ayoff Quit Discharge -1 -2 -3 R etired Other - U - 5 Date F ir s t Hired and Separated Code — Do not f i l l in a t the p la n t. - 53 - APPENDIX I I . 1. D erivation o f Formulas F ir s t Stage o f A ggretation The b a sic index fo r each in d ica to r takes the form : is i f bi Where Xc^ i s the average performance ra te o f workers o f a s p e c ific 1 0 -y e s r age group (c ) w ithin the b a sic comparison group ( i ) , and X ^i i s the average performance ra te o f workers in the base age group (3 5 -4 4 ) w ithin the same b a sic comparison group. The performance ra te may be average hourly earn in gs, attendance r a te , or c o n tin u ity -o f-s e r v ic e r a te , and, as in d icated elsew here, the b a sic com p arison group i s d iffe r e n t fo r each o f the three m easures. In aggregating th ese o r ig in a l indexes so th a t they w ill represent la r g e r groupings, i t i s d e sira b le th a t the aggregate indexes should have the minimum p o s s ib le v a ria n ce. T herefore, each component index i s w eighted according to i t s r e lia b ilit y , i . e . , error according to the recip ro ca l o f i t s squared standard I f the numerator and denominator samples are u n correlated , then the r e l-v a r ia n c e o f each age group index fo r a d ir e c t comparison group i s _ V 2 V2 ( I o i > = 5s- w. 2 + Is»bi ci where Vc^ and and Nci and 2 / are the population r e l-v a r ia n c e s o f the in d ivid u al scores are number o f in d iv id u a ls in the age group and base group, r e s p e c tiv e ly . Another form o f V * (IC^) i s V2 ( l c i )= » b iNc i S e ttin g W± = Nb lNo l V » c i + N,b i vc2 ^ d e i) then = ! but v2 ^ ) = <r2 ( I °| ) wc i I .2 where I c i s the population index fo r age group c and <5“ * ( l ci ) i s the variance o f the sample in d ex. 8 / In the cases o f attendance and o f co n tin u ity o f s e r v ic e , the r e l variance o f the p la n t indexes i s dependent not on ly on the number o f in d i vid u a ls in the sample but a lso on the len gth o f the observation p e rio d . T his i s because the u n it o f observation fo r attendance i s a man-day and fo r co n tin u ity o f s e r v ic e , a man-month. The on ly m od ifica tio n in the d eriv a tio n which i s en ta ile d by th is i s th a t V2( I ci ) - V N ,S, + V ^ _J2__ Nb is k where Sfc = the number o f days scheduled in the attendance period in the K^b establish m en t, or the number o f months in the observation period fo r co n tin u ity o f s e r v ic e . 55 U sing the re cip ro c a l o f the variance o f each d ir e c t comparison group sample index as the w eight, then the aggregate index i s E I c l * c * <r 2 d « i ) l Since h ? * * 1 Vc2 and I c^ c I 2 c are constant w ith resp ect to the summation, V c l Vb2 E Hi 7 2 + I f the assumption i s made th a t the r e l-v a r ia n c e o f the age groups are the same, i . e . th en , Vb2 Wi = Vc2 2/ ------Nb l + Nc i 2 / There i s l i t t l e th a t Vc 2 = Vb2 # = 12/ evidence in the fin d in g s to ch allen ge th e assumption The d iffe r e n c e s between the age group c o e ffic ie n t o f v a r ia tio n were not c o n siste n t n o r, fo r the most p a r t, s iz a b le . I t should a lso be noted th a t any d iffe r e n c e s between Vb2 and Vc2 would have l i t t l e e ffe c t on the r e la tiv e w eights used fo r averaging the indexes fo r the d ir e c t comparison groups, 1 2 / For attendance and employment c o n tin u ity » b i» c A + Nc i - 5 6 - becomes 2. Variance o f an Age Group Index fo r an Occupational Group The occupational group index Ic Tor age group C i s T 6 ^ l I Wi and the variance o f Xc ■ KA-) is « - 2( i .) • li/ V 2j 2 As shown above, J S L l ft- = o -^ ( I ci ) , % th e r e fo r e , <r*^(Ic) i = V (I« i)2 L » A 2 V 2 "1 2 2 2 c 1 1 / In th is d e riv a tio n , use i s made o f the p r in c ip le s th a t (a) the variance o f a sum o f uncorrelated v a ria b le s i s the sum o f the varian ces o f the v a r ia b le s , and (b) the variance o f a v a ria b le tim es a constant i s the square o f the constant tim es the variance o f the v a r ia b le . 57 - 3. Combining ftfle s e a .fg r .F jn al .Aggregates The indexes fo r the i n i t i a l aggregations (occupational groups, higher and low er paid jo b s and len g th o f serv ic e c la s s e s ) were in turn combined to d erive indexes fo r fin a l aggregations (men and women). These indexes were combined w ith w eights equal to the to ta l number o f workers o f a l l ages in a p a r tic u la r group, i . e . , I c l "A I Hi where I i s equal to a major grouping index (hand, machine, e t c .) fo r age group c , M i s equal to the number o f workers o f a l l ages in the i n i t i a l aggregate group. The variance o f t h is in d ex, fo llo w in g the same procedure ou tlin ed in "2 " is i M j 2 0- 2 ( l e ) d M i)2 ( jr - 58 - A* C o e ffic ie n t o f V ariatio n fo r Output per Man-Hour Occupational Group The age group c o e ffic ie n t o f v a ria tio n fo r an occupation group was derived by combining the re l-v a r ia n c e s o f s p e c ific occupation samples and tak in g the square root o f th is average r e l-v a r ia n c e . This sectio n exp lain s how the w eights were applied to obtain the average r e l-v a r ia n c e . For an age group in a s p e c ific occupation l e t % = the number o f in d iv id u a ls in occupation i (must be > 2) X i = mean o f the sample X£ = mean o f the population in the s p e c ific occupation (t h is value i s estim ated by the sample mean X^) £ — JL % -l ( X i j -X i ) 2 which i s an estim ate o f the population variance o - j 2 V i2 s S j2 / X i2 , the r e l-v a r ia n c e o f the sample, an estim ate o f the population r e l-v a r ia n c e V2 V2 = ° i 2 / *1 * * Assuming a normal d is tr ib u tio n in the s p e c ific occu pation, the re l-v a ria n o e o f S ,2 i s 1 V2 ( ? i 2 ) ~ 2 N i-1 and the r e l-v a r ia n c e o f AV2 , -S I 1 2 / X<2 * *i / S2 • *i according to a general r u le fo r the r e l-v a r ia n c e o f a Ni square• Xi i s - 59 - w I f SA2 and X i O A are u n co rrela ted , the r e l-v a r ia n c e o f ^ ( S la) + v 2 ( P = ) ^ i s approxim ately - 2 & + z £ l 272 1 % H ^ l , (1 * 2 V2T sin ce the second term o f the bracketed H i-1 exp ression w ill be le s s than 0 .0 8 provided V < 0 .3 and n £ 2 and l e s s than 0 .1 2 provided V < 0 . A and n ^ 2 . SVMl + 2V2 So the variance o f V i2 i s approxim ately ^ where Wi = N ^ -l. wi The estim ate o f the re l-v a r ia n c e fo r an occupational group i s th erefo re I W jV i2 and i t s estim ated variance i s Z*± 27^(1 + 2V2 ) Z wi - 60 * U. S. GOVERNMENT PRINTING OFFICE : 1957 0 — 449348
@FedFRASER