The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.
Comparative Job Performance by Age: Office Workers Bulletin No. 1273 UNITED STATES DEPARTMENT OF LABOR James P. Mitchell, Secretary BUREAU OF LABOR STATISTICS Ewan Clague, Commissioner Comparative Job Performance by Age: Office Workers Bulletin No. 1273 February 1960 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, W ashington 25, D.C. - Price 30 cents Preface The U.S. Department o f Labor has been deeply concerned with the problem o f the employment of older-workers and has inaugurated several programs dealing with various facets o f this problem. Among these are studies designed primarily to compare the relative work performance o f older workers with that o f other workers in sim ilar occupations. Other studies conducted by the Department have considered the older worker in relation to pension costs, insurance plans, counseling and placement service, and other employment problems. 1./ This study was conducted in the Bureau o f Labor S ta tis tics' Division of Productivity and Technological Developments under the direction o f James F. Walker, assisted by Stanley F. M iller and Ronald E. Kutscher. The s ta tis tic a l procedures used were those developed by Jerome A. Mark in an e a rlier study o f factory workers. The Bureau wishes to express it s appreciation to the organizations which cooperated in the study. The O ffice Executives Association o f New York was also helpful in locatin g companies having work measurement systems. 1/ For a l i s t o f the studies made by the Department see the back cover. CONTENTS Page In trod u ction ...................... General Findings.................................................................................................... Scope o f Survey ......................................................... Concepts and D e fin itio n s .................................................................................... Limitations o f the D a t a .................................................................................... Comparisons Within Groups ................................................................................ Differences by Sex ............................................................................................ Private and Government Employees ................................................................. Effects o f Incentive Payments ..................................................................... Occupational and S k ill Differences ............................................................. Individual Worker Variation ............................................................................. Consistency and Accuracy of P erform ance...................................................... O ffice Jobs Versus Plant J o b s ........................................................................ 1 2 5 6 7 8 8 8 9 10 13 15 16 TABLES 1. 2. 3. 1*. 5. 6. 7. 8. 9. 10. 11. 12. 13. ill. Indexes of output per man-hour of o f fic e workers, by age group and experience on job ........................................................................................ Indexes o f output per man-hour o f women o ffic e workers, by age group and experience on j o b ..................................................................... Indexes of output per man-hour o f men o ffic e workers, by age group and experience on j o b ................................................................................ Indexes o f output per man-hour o f women o ffic e workers in Govern ment agencies, by age group and experience on j o b ........................... Indexes o f output per man-hour o f men o ffic e workers in Government agencies, by age group and experience on j o b ...................................... Indexes o f output per man-hour o f women o ffic e workers in private industry, by age group and experience on j o b ...................................... Indexes o f output per man-hour o f men o ffic e workers in private industry, by age group and experience on j o b ...................................... Indexes of output per man-hour of women o ffic e workers paid on a time-rate basis, by age group and experience on j o b ............... ... . Indexes o f output per man-hour o f women o ffic e workers paid on an incentive basis, by age group and experience on j o b ....................... Indexes o f output per man-hour o f men o ffic e workers paid on a time-rate basis, by age group and experience on j o b ....................... Indexes o f output per man-hour o f women o ffic e workers, by age group, occupational group, and experience on j o b ............................... Indexes o f output per man-hour o f women o ffic e workers, by age group and s k ill l e v e l ................................................................................ Indexes o f output per man-hour of women o ffic e workers with 18 or more months' service with the com pany.................................................. Indexes o f output per man-hour o f men o ffic e workers with 18 or more months' service with the com pany.................................................. - iii - 18 19 20 21 22 23 21+ 25 26 27 28 29 30 30 CONTENTS—Continued Page CHARTS 1. 2. 3. Relative output per man-hour of women o ffic e workers, by age group and experience on j o b .................................................................................... 1+ E ffect o f experience on o u t p u t ................................................................. 12 Percent of women o ffic e workers having indexes o f output above average a fter 9 months or more experience on j o b ...............................llj. APPENDIX Scope and Method o f S u r v e y ........................................................ Concepts and Methods .......................................................................................... 31 Derivation o f Formula............................................................................................. 33 - iv - - 1 - COMPARATIVE JOB PERFORMANCE BY AGE: OFFICE WORKERS Currently, almost 1|0 percent of the workers in the labor force are U5 years old or over. When these workers become unemployed, for any reason, they often meet resistance when seeking a new job, as evidenced by the fa ct that long term unemployment is proportionately greater among those 1*5 years of age or older. Many firms, either through p olicy or practice, re s trict the hiring of older workers, frequently setting age lim its below U5. This practice is esp ecially prevalent in the hiring of o ffic e workers. 2 / One o f the most frequently cited reasons fo r refusal to hire workers over Individual output per man-hour, therefore, is one of the most important factors affecting a worker's a b ility to find and hold a job. Studies on the work performance o f o ffic e workers by age groups have generally been limited to opinion surveys. This study, based on actual records o f work performed on the job, attempts to show how the output o f o ffic e workers aged h$ or over compares with the output o f other age groups, k/ The workers covered in the study were o ffic e cle rica l workers for whom production records were maintained. The great majority of them were in routine jobs such as typing, filin g , posting, sorting, and card punching. h$ is that they are less productive than younger workers. 3 / Earlier studies made by both government agencies and ty private organi zations have shown that older workers have records o f attendance, safety, and turnover equal or superior to those o f younger age groups. 5/ This study did not include such records. 2 / See Counseling and Placement Services for Older Workers (BES No. E 1^2, 1956). 3 / The reasons most frequently given by employers for not hiring older workers were "in a b ility to meet production standards," "in a b ility to meet physical requirements," "lack o f fle x ib ilit y to meet changing conditions," "pension and insurance costs," and "too close to retirement age." See a rticle by Abraham Stahler, The Older Worker, Job Problems and Their Solution (in Monthly Labor Review, January 1957, pp. 22-26); also see the a rticle The Older Worker (in Factory Management and Maintenance, New York, March 1958, pp. 8596). h/ Previous studies by the BLS covered factory workers. For example, see Comparative Job Performance by Age (BLS Bull. 1223, 1957). 5 / For example, see Absenteeism and Injury Experience o f Older Workers (in Monthly Labor Review, July 191:8, pp. 16-19) and James H. Mullen, Proceed ings o f Second Conference on the Problems o f Making a Living While Growing Old, Commonwealth o f Pennsylvania and Temple University, September 1953, PP. 183-20U. - 2 - General Findings Three important findings emerge from this study. First, the differences in the output per man-hour among age groups of the o ffic e workers within the scope o f the survey were for the most part insignificant. Second, there was considerable variation among workers within age groups, so that large propor tions of workers in the older age groups exceeded the average performance o f younger groups. Third, workers in the older age groups had a steadier rate of output, with considerably less variation from week to week, than workers in the younger age groups. Thus, arbitrary barriers to the hiring of older workers which are related to output seem unwarranted. The findings substanti ate the need fo r individual evaluation o f workers. The average output per man-hour o f workers in the group aged 35 to UU was used as a base and assigned the value o f 100 fo r purposes of comparison. The average fo r each o f the other age groups was within 5 percent o f the base group, except for the youngest group, which was greatly affected by lack of experience. Workers in the higher age groups not only maintained output equal on the average to younger groups but also maintained an equal degree of accuracy. Experience appeared to be the major factor where differences in average performance occurred among the various age groups. "When only the more experi enced workers on the j ob were considered, a ll age groups studied had similar average rates o f output. 6/ The e ffe c t o f experience on average output was also indicated when workers with less than 18 months' service in the company were excluded, and the variation between groups was reduced. Similarly, when workers were divided into smaller comparison groups, such as time and incentive workers, higher and lower sk illed jobs. Government and private employment, and into different occupations, no major differences in work performance that can be attributed to age appeared. Although there were p ra ctica lly no differences in averages between age groups among experienced workers, there was considerable variation in output among workers within the groups. Frequently, an individual worker produced at least twice as much as some o f the other workers in the same group. Generally, however, the great majority o f the workers within a group produced within 20 percent of the average output fo r a ll workers in that group. About I4.5 percent o f workers aged U5 or over produced at a higher rate than the average o f workers in the 35 to UU age group (tables 1, 2, and 3). 6 / Information was obtained fo r trainees, but they were omitted from a ll tabulations because their work standards were not considered comparable to experienced workers. - 3 Workers 65 or over generally averaged as high as any other group-and many o f the individual workers had very high rates of output. This relatively high rate o f output may have been the result o f retirement plans which were in e ffe ct in the establishments studied. 7/ With adequate retirement plans, the less e ffic ie n t older workers could be encouraged to retire, with the result that those who remained would tend to be the more able and productive. 7 / Supplementing socia l security in the private companies. In the case of Government workers, most are covered by a retirement plan that generally provides greater benefits than social security. Chart 1. RELATIVE OUTPUT PER M A N -H O U R OF W O M E N OFFICE W O RKERS By A ge Group and Experience on Job ( A G E G R O U P 3 5 -4 4 = 1 0 0 ) INDEX TOTAL, WOMEN UNITED STATES D E P A R T M E N T OF B U R E A U OF L A B O R S T A T I S T I C S LABOR WOMEN OR OFFICE MORE OFFICE WORKERS WORKERS WITH M O N TH S’ EXPERIENCE. 9 Scope o f Survey 8 / Data for the survey were collected in the winter of 1958-59 for about 6,000 workers whose employment was almost equally divided between 5 Federal Government agencies and 21 companies in private industry. The companies included mail-order houses, insurance companies, a ircra ft manufacturers, machinery manufacturers, refin eries, public u t ilit ie s , banks, publishing companies, and re ta il stores. A ll o f the establishments studied were located in the eastern half o f the country and were large employers in their indus t r ia l categories. Almost half o f the workers studied were employed under some type o f incentive pay system. Many additional establishments were asked to cooperate in the survey but they had inadequate data; e. g ., their records were on a group basis or output could not be associated with hours worked. The occupations covered included the usual cle rica l functions, such as typing, secretarial work, and filin g , maintaining, sorting, or classifying of records. To those occupations were added keypunch and business- and dupli cating-machine operations. 8 / A detailed report on the procedures employed in the survey is given in the appendix. Concepts and D e fin iti ons In evaluating a w orker's performance, many fa c to rs are important. Among these are h is output per man-hour, the accuracy o f h is work, the con sisten cy o f h is performance, h is absenteeism, h is con tin u ity o f s e r v ic e , h is adapta b i l i t y , and h is a b i l i t y to get along with others. Some fa c to rs cannot rea d ily be measured. As mentioned e a r lie r , many have alreatfy been stu d ied and the o ld e r w orker's r e la tiv e p o sitio n has i_een assessed. This study was designed p rim arily to compare work performance o f in d iv id u a l workers by age group. Information on two other fa c to r s, accuracy and consistency o f performance, was obtained as a byproduct o f the study. The o f f ic e workers stud ied in each establishm ent were c l a s s i f i e d by sex and occupation. Within these c la s s if ic a t io n s , they were divided in to s ix age groups. An index o f output was obtained fo r each worker by d ivid in g h is output per man-hour by the average output per man-hour o f workers aged 35 to IjJU years in h is same c la s s if ic a t io n . The indexes fo r in d ivid u al workers from the various establishm ents were combined in to age-group indexes with r e l i a b i l i t y weights based on the number o f workers in the base group. In order to hold constant, or to keep to a minimum, the in flu en ce o f fa c to r s other than age which might a f f e c t worker performance, comparisons were r e s tr ic te d to workers in d iffe r e n t age groups w ithin the same major c la s s if ic a t io n . Some fa c to rs th at could a f f e c t output, such as education, m a rita l sta tu s, and previous experience, were not studied because they were considered too d i f f i c u l t to measure, or were o f r e la t iv e ly minor importance, o r randomly d istrib u te d . The v a ria tio n o f a w orker's output per man-hour from week to week i s a lso important in evaluating h is o v e r a ll performance. Output data were c o lle c te d fo r periods covering from U to 13 weeks fo r each employee in the study, and fo r those workers fo r whom information was obtained on a weekly b a s is , i t was p o s s ib le to measure consistency o f performance, through measuring the v ariatio n o f a worker's weekly average output per man-hour from h is own average output. Accuracy o f work, another important c r i t e r i a o f performance, was a lso measured whenever the employer kept records o f in d ivid u al errors. Data were c o lle c te d on the length o f time each worker had been employed in the s p e c ific job on which he was being measured, and on the t o t a l length o f time that he had been employed in the establishm ent studied. Experience obtained in sim ila r work, in the same or d iffe r e n t establishm ent, was not recorded. - 7 - Lim itations o f the Data In considering the re su lts o f the survey, some q u a lific a tio n s and lim i ta tio n s o f the data should be noted. Information was c o lle c te d only from o f f ic e s which kept production records fo r in d ivid u al workers. Therefore, the data cover only a small proportion o f t o t a l o ffic e workers in the se le c te d establishm ents and do not represent a cross section o f various o f f i c e s k i l l s . For example, production measures generally were lim ite d to the more routine types o f work and included only a few h igh ly s k ille d occupations. Moreover, a high proportion o f the workers studied were on in c e n tiv e work, and those on time work were aware th at th e ir work was being measured by the company. I t i s not known whether these employees would work at the same or d iffe r e n t rates i f they were not receivin g a pay in cen tive or i f they knew th e ir production was not being measured. Consequently, the workers covered were p o s s ib ly operating under d iffe r e n t m otivations from most o f f i c e workers. I t i s p o ssib le th at the older workers studied were not representative o f a l l older workers since only employed persons were included in the study and they mifcht. th erefore, have been a s e le c t group. On the other hand, i t may also be assumed that superior workers have gone on to b e tte r paying job s, where production i s rarely measured. Thus the older w orker's average output rate was undoubtedly influenced by the omission o f these two extremes. 'Porkers Ii5 and over represented about 27 percent o f a l l those included in the study. Information on the age d istr ib u tio n o f a l l o f f i c e workers i s not a v a ila b le . Since current population reports in d icate workers or over repre sent about I4O percent o f t o t a l ncnagricultural employment, i t would appear that the age d istr ib u tio n o f the workers studied might have been somewhat d iffe r e n t from the age d is tr ib u tio n o f a l l o f f ic e workers. Comparisons Within Groups In order to evaluate the e f f e c t o f some o f these other fa c to rs on the re la tio n sh ip o f ape to work performance, tabu lation s were prepared comparing indexes o f output per man-hour w ithin lim ited groups. Although the various d iv isio n s o f the data revealed some v ariation s in the average indexes o f output none o f the d ifferen ces could be a ttrib u te d to age. Differences by Sex Among women, l i t t l e v a ria tio n was found in the average output per man-hour, among the d iffe r e n t age groups except that output was s ig n ific a n t ly below average fo r the youngest, where lack o f experience was apparently an important fa c to r. Wien workers with le s s than 9 months' experience on the job measured were excluded from each o f the age groups, average indexes generally rose and the average fo r the youngest group rose su b sta n tia lly , 9 / There was somewhat greater variation in the average performance, by age group, among experienced men, although the differences were still not large. In addition, both the younger and older age groups (except for the youngest group of all men) showed a higher average than the base group aged 35 to I4U, although this was probably a reflection of the higher sampling error for the relatively small sample. Since men were seldom found in measured o f f i c e a c t i v i t i e s they represented only 15 percent of the workers included in the study. Therefore, i t was necessary to r e s t r i c t to women most o f the d e ta ile d observations and a n a ly sis o f the e f f e c t o f various fa c to r s. Private and Government Employees A few small d ifferen ces appeared in the rela tio n sh ip o f output per man hour to age when output o f workers in Government agencies was compared with that o f workers in p rivate industry. (See ta b le s U, 5, 6, and 7. ) The average performance o f women in Government showed a s lig h t drop in the 55 to 6ig group, but n early h a lf o f the drop was wiped out when only those workers with 9 or more months' experience on the job were considered. Women 65 and over in p rivate industry had a much lower average perform ance index ( 9 3 .7 ) than other age groups, but the sample fo r th is group was very sm all. 9 / In considering experience, only the time spent on the s p e c ific job measured was counted. Experience on s im ila r jobs in the same establishm ent or in other establishm ents was not studied. - 9 - In the other age groups, from under 29 through age 9U, the relation sh ip o f the indexes fo r women in Government and in p riv a te employment was s tr ik in g ly sim ila r. Another d ifferen ce between the two groups was in the v a ria tio n in output between in d iv id u a ls, which was greater among the Government workers in each age group. A higher proportion o f women Government workers had output indexes e ith e r exceedingly high (over 130) or low (under 7 0 ). The d ifferen ce in age d istr ib u tio n i s a ls o o f some in te r e s t. The Govern ment agencies studied employed a much higher proportion o f workers 99 years or over in the jobs studied than did the p rivate companies. 1 0 / In contrast, in the p rivate companies a much higher proportion o f workers under 29 were found. E ffe c ts o f I ncentive Payments I t has been estimated th at no more than 3 percent o f a l l o f f i c e workers are employed under incentive plans. A much higher proportion o f the workers in th is study were employed under an in cen tive system and records kept under such plans were the b est source o f inform ation on in d ivid u al output. 1 1 / However, about h a l f o f the workers studied were not employed under an incentive plan, so that comparison o f r e la tiv e performance by age among workers having d ir e c t wage in cen tives and those workers p aid on a time b a sis could be made. The study revealed groups, whether workers and 1 0 ). Women and men output index o f a l l age over averaged 9 percent only minor d ifferen ces in the rela tio n sh ip between age were paid on a time or an incentive b a sis (ta b le 8, 9, aged 69 or over, paid on a time b a s is , had the highest groups. In cen tive-p aid , experienced women aged 69 or le s s output than comparable workers in the base group. In p riv ate industry establishm ents, where in cen tives are s u b sta n tia lly higher, there was some evidence th at younger workers were r e la t iv e ly more productive when d ir e c t pay in cen tives were o ffe r e d . Workers who were U9 or over in both the in c e n tiv e and nonincentive groups, on the other hand, had about the same r e la tiv e output index. The r e la tiv e production e ffic ie n c y o f experienced women o f f ic e workers in p riv a te establishm ents, c l a s s i f i e d by method o f wage payment, i s shown in the tabulation on the fo llo w in g page. 1 0 / Some o f the establishm ents studied had mandatory retirement p o lic ie s . 1 1 / In contrast to fa c to r y worker in cen tive p lan s, which are most frequently based on a s tr a ig h t piecework system, most o f f i c e in cen tive plans are bonus-type payments given in addition to regular weekly salary. - 10 - Workers 1 / paid on a— Age group Under 25-3U 35-iiU U5-5U 55-6U Time b a sis Incentive basis 9 3 .8 9 8 .5 1 0 0 .0 9 9 .1 9 8 .6 9 7 .6 1 0 1 .2 1 0 0 .0 9 9 .0 9 9 .7 25 y e a r s ............ y e a r s ................... y e a r s ................... y e a r s ................... years ......... 1 / Excludes workers with le s s than 9 months' experience on the jo b studied. I n s u ffic ie n t data were a v a ila b le fo r workers 65 and over to warrant presentation. Incentive and time workers in Government agencies had approximately the same age-performance re la tio n sh ip s. The in cen tive systems in the Government agencies are lim ite d and provide fo r very sm all production bonuses, amounting g en erally to no more than 5 percent above b asic rates p aid to timeworkers and they may be in s u ffic ie n t to a lt e r the age-performance re la tio n sh ip s. In examining the time and in c en tiv e tab u lation s, as in the case o f other comparisons, i t should be kept in mind that the indexes do not r e f l e c t actu al le v e ls o f output, enabling d ir e c t comparisons between the groups. Rather, they r e f l e c t only the r e la tiv e output per man-hour w ithin a group. Incentive plans ev id e n tly reduce the proportion o f workers doing substandard work in a l l age groups. The proportion o f women workers with 9 or more months' experience and with output indexes o f le s s than 70 was a t l e a s t twice as large fo r tim eworkers as i t was fo r in cen tive workers. (See ta b le s 8 and 9 . ) Furthermore, the proportion o f workers with r e la t iv e ly high production rates was greater among the group o f in cen tive workers studied. Since a l l tim e-rated workers were on work measurement programs, th e ir performance may not be in d ic a tiv e o f a l l tim e-rated workers in the occupations studied. I t i s generally agreed that the presence o f a work measurement pro gram has an e f f e c t on the workers, a t l e a s t tem porarily, very sim ila r to an in cen tive plan. Occupational a nd S k i l l D iffe ren c es Workers studied were divided in to 10 occupational categories fo r purposes o f comparison, but some o f the groups were too small to warrant separate a n a ly sis. The la r g e s t group represented those workers c l a s s i f i e d as general c le rk s, who did p ostin g, checking, and general maintenance o f records. Other c la s s if ic a t io n s for which data could be-published were ty p is t s , keypunch operators, sorters and c l a s s i f i e r s , and f i l i n g clerks (ta b le 1 1 ). - 11 - None o f the occupational categories studied showed s u b sta n tia lly d iffe r e n t age-performance re la tio n sh ip s, except that keypunch operators under 35 years o f age had much lower average indexes than other age groups. Older workers, p a r tic u la r ly those 55 or over, had higher than average indexes o f output in routine jobs, such as typing, so rtin g , and f i l i n g . The ro le o f experience was fu rth er illu s t r a t e d when occupational com parisons were lim ite d to workers having 9 months' or more experience on the job. In th e ir case, n early a l l o f the d iffe re n c e s in the keypunch operator c la s s if ic a t io n disappeared and other d ifferen c es between age groups generally narrowed. Almost a l l workers studied, as mentioned e a r lie r , were in routine types o f jobs where performance i s more r e a d ily measurable (ta b le 1 2 ). One out o f 12 o f the women studied was in a job that required higher s k i l l s . 1 2 / When workers in these jobs were compared by age group, the average performance o f workers in the age groups between 25 and 6U showed very sm all d iffe re n c e s. The average f o r those under 25 was much higher (1 1 1 .5 ) but v e r y few workers were in th is category and they were probably very s e le c t. 1 2 / Workers c l a s s i f i e d in the "h ig h e r " jobs were expected to exercise some independent judgment. These jobs could not be considered "h ig h e r" jobs in the o f f i c e as a whole, but on ly the higher among the jobs studied. Typical o f the jobs c l a s s i f i e d as higher was th at o f correspondence cle rk s. Chart 2. EFFECT OF EXPERIENCE ON OUTPUT Percentage Distribution of Women Under Age 25, by Relative Output Under 70 7 0 -7 9 .9 8 0 -8 9 .9 9 0 -9 9 .9 100-109.9 U N I T E D S T A T E S D E P A R T M E N T OF LAB O R 120-120.9 INDEXES OF OUTPUT I' B U R E A U OF LABOR STATISTICS II 0-119.9 | ALL WOMEN fcXXl WOMEN WITH 9 OR MORE IX/Vi MONTHS’ EXPERIENCE. 130 and Over - 13 - Individual Worker Variation The great m ajority o f workers in each age group, both men and women, had production indexes in the range between 80 and 120. This concentration around the average was p a r tic u la r ly n oticeable in the o ld e r groups. In each age group, there were small proportions o f workers w ith average output considera bly above or below the average fo r the group. Only a veiy few workers 55 or over had r a te -o f-o u tp u t indexes below 60 and younger workers with indexes below 60 were generally those with l e s s than 9 months' experience on the job. The proportion o f women o f f i c e workers with indexes o f output over 100 was g rea test fo r the 25 to 3U age group and decreased gradually in each higher ace group. (See chart 3. ) Among experienced workers, the proportion of -workers in the youngest age group with indexes o f output over 100 was the same as fo r those 65 and over. A su b sta n tia l proportion o f workers aged and over had higher than aver age output per man-hour. About U7 percent o f those U5 to 5U, Ui percent o f those 55 to 6h} and Ii2 percent o f those 65 and over had scores greater than 100. The l a t t e r group had about the same proportion as found in the youngest group, under 25. Workers U5 or over ty p ic a lly had a lower modal group than did groups under U5. even though as a group they averaged as high or higher than the younger groups. The percent o f workers 1+5 and over with production indexes o f le s s than 80 was about the same as fo r most of the younger groups, but the lack of extremely low producers among the workers k$ years and over balanced the higher modal groups’ among the younger workers. - 13 - Consistency and Accuracy o f Performance T,,here four or more in d ivid u al weekly production fig u r e s were a v a ila b le , i t was p o s s ib le to examine the week-to-week variation in production among women workers. An "in d ex o f con sistency" 1 3 / was derived fo r each age group by comparing the in d ivid u al weekly output per man-hour o f each woman worker with her average output over the period o f study. The fo llo w in g indexes o f consistency fo r women were found: Age group 1 / Under 23 years 23-3U y e a r s ................. 35-UU y e a r s ................. h5-5h y e a r s ................. 33-6ij y e a r s ................. Number o f workers U5l U68 3U6 2lj.l 61 Index o f consistency 8 3 .8 9 8 .9 1 0 0 ,0 1 0 7 .0 1 2 3 .8 1/ I n s u ffic ie n t data were a v a ila b le fo r workers aged 63 and over to warrant p resentation. Thus, the index o f 1 2 3 .8 fo r the 33-6U year age group showed that they were 23. 8 percent more c o n sisten t in th e ir week-to-week performance than the base group. The tendency toward con sisten t performance in th is group was noted in nearly a l l o f the establishm ents fo r which data were a v a ila b le . In the younger age groups, output per man-hour varied from week to week, re s u ltin g in lower indexes o f consistency. The low rate fo r the youngest group may be p a r tly a ttrib u ta b le to the r e la t iv e ly la r g e r number o f le s s experienced workers in th a t group. 1 3 / A con sistency index fo r each ind ivid u al was found by comparing the average percent d eviation about h is own average over the t o t a l period with the average o f t h is d eviation fo r the base group (3 3 -lUt y e a r s). The in d iv id u a l indexes were combined to form average indexes fo r each age group by using the same method as fo r the output indexes. The recip ro ca ls o f these indexes were used to measure consistency. Comparisons were lim ite d to women because o f the sm all number o f reports obtained fo r men. - 16 - Although the major o b je c tiv e o f t h is study was to compare the r e la tio n ship o f output per man-hour to age, some comparison o f the q u a lity (accuracy) o f work among the age groups was a lso p o ssib le in some of the cooperating establishm ents where a record was maintained o f the errors made by each in d i vidual on measured work. 1 h/ The r e s u lts , presented in the follow in g tabu lation , while not conclusive because o f the sm all number o f workers covered, show no appreciable d ifferen c e s in accuracy o f output between age groups: Age group Under 25 y e a r s ................... 25-3^ years .......................... 35-U i years ......................... 1*5-51* years .......................... 55-61* years ........... 65 years and o v e r ........... Accuracy index 1 0 0 .2 99. 7 1 0 0 .0 99. 7 100.0 9 8 .0 l l * / This comparison was made by c a lc u la tin g an index o f r e la tiv e accuracy. Each worker’ s accuracy percentage was obtained by subtracting h is error percentage from 100. The average accuracy ra tin g fo r the base group was then determined and an accuracy index f o r each in d iv id u a l ca lcu la ted r e la tiv e to the base group average. These data were then combined in the same manner as the output data to get an average accuracy ratin g for each age group. - 17 - O ffic e Jobs Versus Plant Jobs Two e a r lie r stu d ies o f the Bureau o f Labor S t a t i s t i c s on job performance by age were confined to fa c to ry workers in the men's footwear, men's c lo th in g , and household furniture in d u strie s. 1 5 / A l l workers included in those stu d ies were on jobs with in cen tiv e systems o f payment. The r e s u lts fo r o f f i c e workers and fa c to r y workers were very sim ila r. For both groups there was r e la t iv e ly l i t t l e v a ria tio n in average performance among age groups, but considerable v a ria tio n among in d ivid u als w ithin age groups; a large proportion o f workers in the higher age groups exceeded the performance o f the base group average. There were, however, a few d iffe re n c e s. The study o f o f f i c e workers showed l i t t l e or no v a r ia tio n among age groups; in most cases, the youngest group (under 25) had a lower average, but t h is was prima r i l y because o f l e s s experience. Among fa c to ry workers, performance tended to be hig h est fo r the 25 to 3k year group. The old er workers in the fa c to ry study had somewhat lower average performance ra tes than the base group, 35 to UU, while the averages fo r o ld e r o f f i c e workers were about the same as the base group. A ll o f the stu d ies showed th a t nearly h a lf o f the workers aged 1;5 to 3k had output per man-hour indexes greater than the average in the worker age group 35 to I4J4.. On the other hand, only a sm all number o f o f f i c e workers k3 years o ld and over had very low scores,* whereas the proportion was higher among the fa c to ry workers k3 and over. Another d iffe re n c e in the findin gs o f the stu d ies was rela ted to the in d iv id u a l performance o f workers 55 and over. The proportion o f o f f ic e workers 55 and over who.had output indexes greater than 100 was almost the same as the proportion fo r the 1+5 to 3k age group. In the case o f factory workers, the proportion was lower. This d ifferen c e might be a r e fle c tio n o f the greater p h ysical demands which fa c to r y jobs make on persons in the old er age groups. 1 5 / Comparative Job Performance by Age, op. and Age (BLS B u ll. 1203, 1956). c i t . ; and Job Performance - Table 1. 18 - Indexes o f output per man-hour o f o f f i c e workers by age group and experience on job ^Age group 35-1*1**1007 Pge group Number Average of index workers Percentage d istrib u tio n o f workers with indexes o f Under 70 70 to 7 9 .9 90 80 to to 8 9 .9 9 9 .9 100 to 1 0 9.9 110 to 1 1 9.9 120 to 1 2 9 .9 130 and over A l l workers Under 25 . . 25-31* .......... 35-1*1*.......... 1*5-51*......... 55-61* .......... 65 and over 1,081* 1 ,5 0 6 1,1*66 1 ,0 2 3 1*29 86 92.1* 99.1* 1 0 0 .0 1 0 0 .1 9 8 .6 1 0 1 .2 15 7 6 6 7 7 13 9 6 8 6 1* 17 15 11* 15 18 21* 20 21 21* 23 21* 19 18 21 25 19 18 17 8 13 11* 13 12 11* Workers with 9 months' or more experience on job Under 25 . . 25-31* .......... 3 5 -1 4 * .......... 1*5-51*......... 5 5 -6 1 * .......... 65 and over 582 1,071* 1 ,1 8 9 877 371 71* 9 8 .7 1 0 1 .9 1 0 0 .8 1 0 0 .8 9 9 .5 1 0 1 .1 7 5 5 5 6 7 9 7 5 7 6 3 17 11* 13 17 18 25 23 22 25 21* 25 21 22 23 26 21 19 16 9 H* 11* 13 13 12 5 7 6 6 8 1* 1* 7 5 8 6 10 6 7 6 5 8 5 6 8 5 9 5 ll 1/ 1 / Does not include data for about 1*50 workers for whom data on length o f experience were not a v a ila b le . Previous experience in sim ila r or id e n t i c a l work was not considered. NOTE; Because o f rounding, the percentages may not equal 100. - 19 - Table 2. Indexes o f output per man-hour o f women o f f i c e workers, by age group and experience on job [fge group 35>-Ui=1007 Age group Number Average of index workers Percentage d istrib u tio n o f workers with indexes of. Under 70 70 to 7 9 .9 80 90 to to 8 9 .9 9 9 .9 100 to 1 0 9 .9 110 to 1 1 9 .9 120 to 1 2 9.9 130 and oyer A ll workers Under 25 . . 25-3U .......... 3 5 -iU t.......... U 5-51i.......... 55-6U ......... 65 and over 1 ,0 5 5 1 ,2 7 1 1 ,198 832 351 67 9 2 .1 9 9 .2 1 0 0 .0 9 9 .2 9 8 .1 1 0 0 .9 15 7 5 6 7 8 1U 8 6 8 6 5 17 15 13 16 18 2U 21 21 25 2U 27 17 17 22 26 20 18 111 8 13 111 13 11 17 ii 7 6 6 7 h li 6 5 7 6 10 6 7 6 5 7 5 6 7 5 7 6 10 Workers with 9 months’ or more experience on job 1 / Under 25 . . 2 5 - 3 l i ......... 3 5 - U i .......... U5-5U .......... 55-6U .......... 65 and over 568 871 955 700 301 60 9 7 .8 1 0 1 .5 1 0 0 .9 1 0 0 .0 9 9 .2 9 9 .9 8 5 k h 5 9 9 6 5 7 5 U 17 111 13 17 18 25 23 21 26 2li 28 19 21 25 28 22 19 12 9 111 1U 13 12 15 1 / Does not include data fo r about liOO workers fo r whom data on length o f experience were not a v a ila b le . Previous experience in sim ila r or id e n t i c a l work was not considered. NOTE: Because of rounding, the percentages may not equal 100. - Table 3. 20 - Indexes o f output per man-hour o f men o f f i c e workers, by age group and experience on job /Age group 35-kkm100/ ■Age group Number Average of index workers Percentage distribution of workers with indexes of- 80 70 Under to 70 79.9 90 to to 89.9 99.9 100 to 109.9 110 to 119.9 120 to 129.9 130 and over 12 5 8 5 11 (1 /) 12 6 16 6 A ll workers Under 25 . . 25-3U ........ 3 5 -U i........ b$-$h ........ 55-61:........ 65 and over 29 235 268 191 78 19 98.6 103.0 100.0 103-5 100.3 101.9 16 5 7 9 7 10 5 8 9 <3/> <3/> 15 5 2h 13 21 17 23 13 20 1U (1 /) (1 /) 29 17 21 16 18 (1 /) 20 Hi 15 10 15 <3/) 2 (3/) Workers with 9 months’ or more experience on job 2 / Under 25 .. 25-3U ........ 35-lUi........ U5-5U ........ 55-6U ........ 65 and over 1U 203 231: 177 70 1U 107.9 103.7 100.2 103.7 100. h 105.9 (1 /) ~h 7 9 6 (1 /) (V) (V) (3/) Tl 23 15 16 21 5 7 25 iu 10 20 13 <3/> (1/) (3/) (1 /) T5 21 15 18 (1 /) (i/) IU 15 10 17 (1 /) "5 8 5 ll (1/) (3/) (1 /) 13 6 15 5 (1 /) 1 / Data were considered in su fficien t to present distribution. 5 / Does not include data fo r about $0 workers fo r whom data on length o f experience were not available. Previous experience in sim ilar or id e n ti ca l work was not considered. NOTE: Because o f rounding, the percentages may not equal 100. - Table ii. 21 - Indexes o f output per man-hour o f women o f f i c e workers in Government agencies, by age group and experience on job /Age group 35-UU=100^ Age group Number Average of index workers Percentage distribution of workers with indexes of 120 100 110 90 80 70 Under to to to to to to 70 79.9 89.9 99.9 109.9 119.9 129.9 130 and over A ll workers Under 25 .. 25-3U ........ 35-UU ........ U5-51i . . . . . 55-6H........ 65 and over 197 U80 593 368 210 57 91.9 99.6 100.0 99.0 96. U 102.6 17 10 6 8 9 7 1U 7 6 8 6 h lit 13 13 17 21 21 22 21 25 22 23 18 Hi 20 26 16 19 15 9 lli Hi 13 7 19 ii 7 ii 8 10 5 5 8 6 8 5 11 Workers with 9 months' or more experience on job 1 / Under 25 .. 25-3U ........ 3 5 -U i........ U5-5U ........ 5 5 -6 ii........ 65 and over 61 311 h66 289 173 50 97.8 101.6 100.6 99.7 98.5 101.6 12 6 5 6 7 8 10 5 5 8 5 3 12 13 13 19 22 22 25 2k 26 20 22 22 16 21 28 19 20 13 10 15 5 6 Hi ii 12 7 17 6 10 6 9 9 6 9 6 10 1 / Does not include data fo r about UOO workers fo r whom data on length o f experience were not available. Previous experience in sim ilar or id e n ti ca l work was not considered. NOTE: Because o f rounding, the percentages may not equal 100. - Table 5. 22 Indexes o f output per man-hour o f men o f f ic e workers in Government agencies, by age group and experience on job /Age group 35“ kk’B100f Age group Number Average of index workers Percentage distribution : o f workers with indexes of 120 110 100 90 80 70 Under to to to to to to 70 79.9 89.9 99.9 109.9 119.9 129.9 130 and over A ll workers Under 25 .. 25-3h ........ 35-14*........ 1*5-51*........ 55-61*........ 65 and over 11* 172 229 153 57 11 96.1 101*. 3 100.0 10l*.l* 100.5 101.5 (1 /) (1 /) 6 To 6 5 8 8 8 7 (1 /) (1 /) (1 /) (1 /) ?2 12 18 20 22 15 18 15 (1 /) ( ] / ) (V) 17 22 11* 17 (2 /) (1 /) “5 15 11 18 <2/> (1 /) “6 8 5 ll (l/) (1 /) 13 6 16 5 (2 /) Workers with 9 months' or more experience on job 2 / Under 25 .. 25-31*........ 35-14*........ 1*5-51*........ 55-61*........ 65 and over 5 11*6 195 139 1*9 6 (V) 105.2 100.2 101*. 8 100.8 (3 /) (1 /) (1 /) Tl "5 6 5 6 9 8 8 (1 /) (1 /) a (1 /) (1 /) 11* 21 17 16 23 17 11* (1 /) (1 /) (1 /) T5 22 11* 18 (2 /) (1 /) T5 15 11 21 <2/> (1 /) “6 8 6 10 (2 /) Tl*6 (i/) 15 1* (!/) 1 / Data were considered in su fficien t to present distribution. 2 / Does not include data fo r about 50 workers fo r whom data on length o f experience were not available. Previous experience in sim ilar or id e n ti c a l work was not considered. 3 / An age group containing fewer than 10 workers was considered too small fo r presentation o f an index. NOTE: Because o f rounding, the percentages may not equal 100, - 23 - Table 6. Indexes o f output per man-hour o f women o f f i c e workers in private industry, by age group and experience on job /Age group 3$-Uhs100/ Number Average of index workers Age group Percentage distribution o f workers with indexes o f Under 70 to 70 79.9 80 90 to to 89.9 99.9 100 to 109.9 110 to 119.9 120 to 129.9 130 and over All workers Under 25 . . 2$-3k ........ 35-UU........ U5-5U ........ 55-6U ........ 65 and over 858 791 605 U6U 1U1 10 92.1 98.8 100.0 99.5 100. U 93.7 1U 5 $ 13 10 6 k 9 6 $ (1 /) (1 /) 18 17 13 15 12 (1 /) 20 20 25 25 33 (3/) 25 8 12 13 18 (1 /) 17 <2/) 18 2k 2k Ik k k 7 7 5 5 6 6 k u (3/) (3/) Workers with 9 months' or more experience on job Under 25 .. 25-3it........ 3 $ -W ........ U5-5U........ $5-6h ........ 65 and over 1/ 507 560 U89 Ull 128 10 97.8 101. h 101.2 100.2 100.2 93.7 7 9 8 3 3 7 3 U (1 /) (1 /) $ $ 18 23 19 15 26 13 26 15 13 35 (1 /) (1 /) 23 28 28 25 18 (1 /) 9 13 13 Ik 17 (l/> Bata were considered in s u ffic ie n t — to present distribu tion . NOTE: Because o f rounding, the percentages may not equal 100. 6 8 7 3 k (1 /) 5 6 U 5 5 (l/) - Table 7. 2h - Indexes o f output per man-hour o f men o ffic e workers in private industry, by age group and experience on job /Age group 35-UU=1007 Age group Number Average of index workers Percentage d istribu tion o f workers with indexes o f Under 70 70 to 79.9 80 90 to to 89.9 99.9 100 to 109.9 110 to 119.9 (1 /) 18 18 21 18 (1 /) (1 /) 120 130 to and over 129.9 A ll workers Under 25 .. 25-3U ........ 35-UU ........ U5-5U ........ 55-6U ........ 65 and over 15 63 39 38 21 8 102.7 98.3 100.0 99.1* 99.h (2 /) (1 /) (1 /) 9 3 10 5 10 10 0 16 (1 /) (1 /) (1 /) (1 /) T8 31 13 23 30 5 28 10 (1 /) (1 /) 9 15 5 6 (3 /) (1 /) 2 8 5 15 (1 /) ( 1 /) 9 8 13 6 (1 /) Workers with 9 months' or more experience on job Under 25 .. 25-3U ........ 35-UU ........ U5-5U ........ 55-6U ........ 65 and over 9 57 39 38 21 8 (2 /) 9H. u 100.0 99. k 99.h (2 /) (1 /) (1 /) El ~k 10 5 10 10 16 0 (1 ( a /) /) (1 /) (1 /) 18 29 23 13 5 30 28 10 ( 1 /) (1 /) (1 /) 16 18 21 18 (1 /) (1 /) 11 15 (1 /) 2 8 6 15 (1/ ) 5 ( a /) 5 1 / Data were considered in s u ffic ie n t to present d istrib u tion . ? / An age group containing fewer than 10 workers was considered too small fo r presentation o f an average index. NOTE: Because o f rounding, the percentages may not equal 100. ca/> 9 8 13 6 ( a /) - 25 Table 8. Indexes o f output per man-hour o f women o f fic e workers paid on a time-rate basis, by age group and experience on job /Age group 35-kka100/ Age group Number Average of index workers Percentage distribution o f workers with indexes o f— 70 Under to 70 79.9 80 90 to to 89.9 99.9 100 to 109.9 110 to 119.9 120 to 129.9 130 and over A ll workers Under 25 .. 25-3U ........ 35-14: ........ 45-5U ........ 55-6U ........ 65 and over 302 572 712 423 147 24 92.8 99.0 100.0 99.6 97.5 103.7 13 7 5 7 9 18 15 7 6 7 6 0 16 14 12 16 17 8 21 21 25 22 21: 19 17 2k 29 22 22 16 8 ia 5 5 13 13 k 17 k n 5 5 5 6 5 8 6 17 Workers with 9 months' or more experience on job Under 25 .. 25-34 ........ 35-14:........ 45-54 ........ 55-64 ........ 65 and over NOTE: 157 1:67 636 383 134 2k 95.7 100.6 100.5 99.8 97.7 103.7 9 6 5 6 9 18 11 6 6 7 6 0 17 14 11 16 17 8 2k 22 25 23 23 19 19 25 30 22 23 16 7 15 lU 13 11 17 Because o f rounding, the percentages may not equal 100. 6 6 U 5 6 k 6 7 5 8 5 17 - Table 9. 26 - Indexes o f output per man-hour o f women o f f i c e workers paid on an incentive basis, try age group and experience on job /Age group 3$-ljl*“ 1007 Age group Number Average of index workers Percentage distribution o f workers with indexes o f — 70 Under to 70 79.9 80 90 to to 89.9 99.9 100 to 109.9 110 to 119.9 120 to 129.9 18 21 21 19 11* 13 8 12 11* li* 11 17 1* 8 8 7 9 1* 1* 7 5 5 6 6 6 9 8 5 9 5 6 8 5 5 6 3 130 aid over A ll workers Under 25 .. 2$-3l* ........ 35 - 1 * ........ i*$-$i*........ $5-61*........ 65 and over 753 699 1*86 1*09 20l* 1*3 91.6 99.1* 100.0 98.8 98.8 99.0 15 7 5 5 6 0 13 10 6 10 5 9 18 16 15 16 18 36 21 20 21* 25 30 15 Workers with 9 months' or more experience on job 1 / Under 25 .. 25-31* ........ 35-10*........ 1*5-51*........ 55-61*........ 65 and over 1*11 1*01* 319 317 167 36 99.2 102.9 101.6 100.1* 101.1 96.5 6 3 2 2 2 0 j_____ 8 7 3 8 3 8 17 15 15 18 19 1*2 23 21 28 25 31* 19 23 21* 21* 23 11* 9 10 13 13 11* 13 11* 1 / Does not include data fo r about 1*00 workers fo r idiom data on length o f experience were not available. Previous experience in sim ilar or id e n ti ca l work was not considered. NOTE: Because o f rounding, the percentages may not equal 100. - 27 - Table 10. Indexes o f output per man-hour o f men o f f ic e woricers paid on a tim e-rate basis, by age group and experience on job /Kge group 35-l*H=1007 Age group Percentage distribution of workers with indexes o f — Number Average 120 no 70 100 80 90 of index Under to to to to to to workers 70 79.9 89.9 99.9 109.9 119.9 129.9 130 and over A ll workers Under 25 .. 25-3*4........ 35-1*1*........ U5-514........ 55-614 . . . . . 65 and over 16 203 21*0 181* 6*4 15 91.1 103.5 100.0 103.5 100.9 106.6 (V ) (1/) 5 10 7 5 7 9 6 10 (1 /) (V ) <3/) (1 /) 23 15 20 17 1*4 23 17 lU <3/> <2/> (V ) 16 21 16 18 (l/> <Y> 11* 15 10 18 (1 /) (1 /) 5 8 5 12 (1 /) (1 /) 13 6 15 5 <3/> (1 /) "5 8 6 12 <2/> (l/) 11* 6 15 5 (l/) Workers with 9 months' or more experience on job Under 25 . . 25-3*4........ 35-*4*4........ *45-5*4........ 55-61*....... 65 and over 9 19U 231 176 63 lU (2 /) 103.9 100.2 103.8 101.0 105.9 (1 /) (1 /) Tl 14 7 5 7 9 6 10 (V ) (1 /) (V ) (1 /) T5 ?2 16 21 114 21* 17 13 (1 /) (V ) (1 /) T5 21 15 18 (1 /) (1 /) 11* 16 10 18 (3 /) 1 / Data were considered in s u ffic ie n t to present distribu tion. ? / An ape group containing fewer than 10 workers was considered too small fo r presentation o f an average index. NOTE: Because o f rounding, the percentages may not equal 100. Table 11. Indexes o f output per man-hour o f women o f f ic e workers, by age group, occupational group, and experience on job /J-ge group 35-iiU=1007 Age group Under 25 . . . . 25-31* ............ 35-1*1*........... U5-5U ............ 55-61* ............ 65 and over . General Typists clerks Number Number Average Average of of index workers index workers 3U7 573 575 387 165 1*2 93.0 100.3 100.0 98.6 98.1 98.5 168 119 91 135 38 3 9U.0 102.2 100.0 100.1 100.7 (1 /) Keypunch Sorters, operators cla s sifie rs Number Number Average Average of of index workers index workers A ll workers 80 78 61 39 21 5 85. k 86.6 100.0 97.5 97.0 (1 /) 90 138 120 61 18 2 99.1* 100.8 100.0 103.1 103.6 < 1 0 File clerks Number Average of workers index 171* 128 107 58 38 6 88.1* 96.0 100.0 91*. 9 102.3 (1 /) 96 71 81* 1*5 32 5 96.1* 99.5 102.1 96.5 102.1* Workers with 9 months' or more experience on job 2 / Under 25 . . . . 25-3U ............ 35-1*1*............ 1*5-51*........... 55-61* ............ 65 and over . 199 1*60 510 3U8 15U 38 95.5 101.3 100.1 98.7 97.9 96.6 92 81 71* 120 35 3 99. U 105. U 100. k 101.7 102.1* <*/> 28 25 30 19 16 5 100.0 99.6 98.1* 100.6 96.3 < 2 /> 1*5 91* 88 60 16 2 10i*.i* 103.0 103.9 103.1* 101*. 1 < !/) ( V ) 1 / An age group containing fewer than 10 workers was considered too small fo r presentation o f an average index. 21 Does not include data fo r some workers f o r whom data on length o f experience were not available. Previous experience in sim ilar or id e n tica l work was not considered. - 29 - Table 12. Indexes o f output per man-hour o f women o f f i c e workers, by age group and s k i ll le v e l 1 / /Age group 3%-kha100/ Age group Under 25 .. 25-3U ....... 35 - U i........ U5-5U........ 55-6U ........ 65 and over Number Average of index workers 16 111.5 99.7 177 125 U2 7 1 0 0 .0 108 98.2 96.0 (3 /) Percentage d istrib u tion o f workers with indexes o f Under 70 70 80 90 100 to to to to 79.9 89.9 99.9 109.9 Higher le v e l jobs (2 /) (2 /) 12 ~6 9 9 16 9 18 8 (2 /) (2 /) (2 /) (2 /) 9 12 2ii 12 19 12 17 (2 /) (2 /) 110 to 119.9 120 to 129.9 130 and over ( 2 /) T8 21 15 22 (2 /) (2 /) 12 11 9 8 (2 /) (2 /) 9 5 7 5 (2 /) (2 /) 9 9 13 10 ( £ /) 18 23 27 21 18 16 8 13 U 6 6 6 8 5 U 6 U 6 5 12 Lower le v e l jobs Under 25 .. 25-3U ........ 35-W i........ U5-5U ........ 55-6U ........ 65 and over 1,036 1 ,15U 1,006 705 309 60 91.6 99.0 100.0 99.9 98. U 101.9 15 6 h h 6 li 111 9 5 8 5 6 18 20 20 25 2h 30 27 16 17 16 1U 17 111 1U 11 13 1 / Higher le v e l jobs were those in which workers were expected to exercise some independent judgment. 2 / Data were considered in s u ffic ie n t to present 'U stribution. J / An age group containing fewer than 10 workers was considered too small fo r presentation o f an average index. NOTE: Because o f rounding, the percentages may not equal 100. 30 Table 13. - Indexes o f output per man-hour o f women o f f i c e workers with 18 or more months' serv ice with the company /Age group 35-itU=100/ Number Average of workers index Age group Under 25 .. 25-3U ........ 35-JUJU........ U5-5U ........ 5 5 -6 ii........ 65 and over NOTE: Table lU. U09 928 i,ohk 786 33k 67 99. li 101.7 100.9 99.5 98.7 100.9 Percentage distribution of workers with indexes o f — 70 Under to 70 79.9 6 5 li 6 7 8 6 6 6 8 6 5 80 90 to to 89.9 99.9 17 lit 13 16 17 2h 25 21 25 23 27 17 100 to 109.9 22 25 26 20 19 lli 110 to 119.9 120 to 129.9 12 lli lli 13 11 17 6 8 6 6 7 h 130 and over 6 7 6 7 6 10 Because o f rounding, the percentages may not equal 100. Indexes o f output per man-hour o f men o f f i c e workers with 18 or more months' service with the company /Age group 35-lili®1007 Number Average of index workers Age group Under 25 .. 25-3U ........ 35-lili....... Ii5-51i........ 55-6U ........ 65 and over 1/ 15 217 260 187 75 17 9U.5 102.7 99.8 103.8 100.5 10U.2 Percentage distribu tion o f workers with indexes o f 120 100 110 90 80 Under 70 to to to to to to 70 79.9 89.9 99.9 109.9 119.9 129.9 lo (1 /) ( 1 /) "5 7 5 8 9 7 9 (1 /) (1 /) Si (1 /) (1 /) 11; 21 17 lli 23 19 13 (1 /) (1 /) (1 /) 17 22 15 18 (1 /) (V) 12 15 10 16 <3/> (3/) Data were considered in s u ffic ie n t to present d istrib u tio n . NOTE: (1 /) 5 8 5 12 Because o f rounding, the percentages may not equal 100. 130 and over (1 /) 12 5 16 6 (1 /) - 31 APPENDIX. Scope and Method o f Survey Only companies or governmental agencies which had a work measurement system for individual c le r ic a l employees and employed both younger and older workers on comparable work were included in the survey. Twenty-one companies were found that met these crite ria and data were collected covering 3,OU3 o f their c le r ic a l employees. The fiv e cooperating governmental agencies supplied data on 2, 891 workers. Only a few o ffic e s from any one company or agency having a number o f d is tr ic t o ffic e s were included, in order to lim it the e ffe c t any one establishment would have on the fin a l results. Typing and secretarial work, keypunch operation, tabulating, computing and bookkeeping machine operation, operation o f duplicating machines, and filin g , sorting and routing and assembling records were considered to be o ffic e or c le r ic a l functions. The geographic scope o f the study was nationwide; however, no companies from the fa r western region met the necessary criteria . Many companies in i t ia lly contacted could not be included in the study because they met only a portion o f the necessary criteria , such as having group rather than individual measurement programs or employing only younger workers on measured a ctiv itie s during the period covered by the survey. Concepts and Methods For the purpose o f this study produ ctivity was defined as an in d iv id u a l's physical volume o f production per hour worked—his output per man-hour. To derive th is measure i t was necessary, therefore, to obtain data on an employee's output in measurable units and to relate that output to the corresponding hours spent in i t s production. For the most part, companies maintained records o f the ph ysical production which could be related to time on the job. In some cases, however, an in d i v id u a l's output per man-hour was measured in terms o f the standard performance fo r his job. In such establishments, the company maintained a record o f the ra tio o f each in d iv id u a l's actual output per man-hour to the standard hourly output fo r his job. These standards were usually obtained by measuring the production o f a l l individuals over a period o f time to determine the average or expected performance. In eith er case, a single output per man-hour figure was obtained, whether output was recorded d ir e c tly or whether the ra tio o f actual to standard performance was taken. The observation period fo r the output per man-hour data ranged from 1| to 12 weeks in the various companies. This was a compromise between a very long period which would tend to even out atypical influences of a temporary nature, and a very short period which would permit the in clu sion o f a larger number o f individuals. - 32 In order to is o la te the influence o f age from the many other factors which a ffe c t output per man-hour and at the same time to combine measures drawn from, small groups o f persons into larger aggregates, certain s t a t is t ic a l procedures were applied to the orig in a l data. F irst, each individual employee was c la s s ifie d by age, into 1 o f 6 age groups; namely, under 25, 25 to 3U, 35 to Ui, 1*5 to 5U, 55 to 6U, and 65 years and over. The employees were then further c la s s ifie d into groups by selected ch a ra cteristics which might a ffe c t work performance, such as sex, s p e c ific occupation, length o f serv ice, and method o f payment. The purpose o f th is c la s s ific a t io n was to insure that age-performance observations were made only among individuals having in common those ch a ra cteristics which migjit have an important bearing on produ ctivity. This basic comparison group varied somewhat according to the method used fo r deriving an in d iv id u a l's output figu re. Where records were maintained as an actual count o f production, the comparison group was lim ited to individuals o f the same sex and occupation within a company. In companies maintaining records o f p rod u ctivity as a percent o f standard performance, the basic com parison group was broadened to include a l l measured workers o f the same sex. The purpose was to enlarge the sample o f individuals to secure greater r e lia b i l i t y in the basic measures. This procedure was adopted only where company o f f i c i a l s were s a t is fie d that th e ir standards had the same chance o f achieve ment among the d iffe re n t jobs. In cases where the standards were not consid ered en tirely comparable between occupations, the basic comparison group was lim ited to the same occupation within a company. Within each basic comparison group, the average produ ctivity fo r the age group 35 to Ui was determined. This group was designated as the base group and the r a tio o f output o f each individual within the basic comparison group to the average fo r the base group was computed. Through the use o f these indexes and the c la s s ific a tio n system, the influence o f other than age fa ctors was la rg e ly eliminated, since each in d i vidual was being compared only with other individuals having most factors in common. The individual indexes were then comparable from one comparison group to another and combined to derive average indexes fo r each age group. The - 33 - combinations were made using weights based on the r e lia b ility o f the individual indexes. This weight was a reflection of the number o f people in the age group and the number in the base group within each comparison group, 16/ and was applied to the individual output indexes, which were combined to derive average indexes o f output fo r each o f the six age groups. 17/ Derivation o f Formula The basic age group index fo r each comparison group takes the form: X ci V Where X . i s the average performance rate o f individuals within a s p e c ific ci age group (c ) within the basic comparison group ( i ) , and Xb i i s the average performance rate o f workers in the base group ( 35>-UU) within the same basic comparison group. The performance rate represents output per man-hour and, as indicated elsewhere, the basic comparison group varies with type o f records used. In aggregating these o rig in a l indexes so that they w ill represent la rger groupings, i t i s desirable that the aggregate indexes should have the minimum p ossib le variance. Therefore, each component index i s weighted according to it s r e lia b i l i t y ; i. e . , according to the recip roca l o f i t s squared standard error. 16/ For formula used, see derivation o f formula below. IT / In the e a r lie r study o f fa ctory workers' job performance by age, the combinations were made by s ig n ifica n t groups, using constant weights in order to minimize the e ffe c t s o f other factors on the data. This method was not considered necessary in th is study and in order to obtain the d istrib u tion o f indexes the data were combined as described above. A comparison o f both methods showed the resu lts to be sim ilar. - 3k I f the numerator and denom inator samples are u n co rre la te d , then th e r e l v a ria n ce o f each group in d ex f o r a d i r e c t com parison group i s V2d c i) = vc2 + vb2 N. bi Nc i 2 where VQ and Vb 2 are the p o p u la tio n r e l-v a r ia n c e s o f the in d iv id u a l s c o r e s and Nc j_ and Nbj_ are number o f in d iv id u a ls in the age group and base group, r e s p e c t iv e ly . Another form o f V ^ I i ) IS v2d Ci ) Nb i Nc i » b i» ,c x S e t t in g Wi N c i + Nbi V, then V2 ( l c i ) = _ !_ j bu t V2 ( I c i ) = <r 2 ( I c l ) w.c i Ic2 where I c i s th e p o p u la tio n in d ex f o r age group c and <r 2 ( I c i ) i s the v a ria n ce o f the sample index. Using th e r e c ip r o c a l o f each d ir e c t com parison group sample in d ex as the w eigh t, then th e a g gregate in d ex i s w. x Cl LC 1 2( Ic i) V =^= 2 _ <3- 2 ( I c i ) z V vc 2 I c 2 wi I vA. C - S ince Vc 2 and I c 2 35 - are con sta n t w ith r e s p e c t t o th e summation, NbiVci X ^ i^ c i Nc i / V ~ N^ i + Nb i / y Ife - Nc i + Nb i 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. then, Nb i Nc i WjL Nbi + Nc i I f th ese w eigh ts were to be a p p lie d to in d iv id u a ls w ith in each group, th e w eigh t f o r any in d iv id u a l in age group c and com parison group i would be found by. d iv id in g th e above form ula by the number o f in d iv id u a ls in group c to g e t W4 Jk___ H, . + W . wb i ”ci - 36 - The b a s ic form used in c o l l e c t i n g the data i s reproduced below . * U.S. GOVERNMENT PRINTING OFFICE : 1960 0 — 540764
@FedFRASER