Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
The Participation of Young Women in Employment and Training Programs Margaret Simms When the federal government initiated employment and training programs in the 1960s, the focus was on assisting adult males who had been displaced from their jobs by technological change or who were structurally unemployed. Even in youth programs, where displacement and long-term unemployment were less important, the emphasis was on males. It was thought that their employment needs were greater and that unemployment was likely to lead to criminal behavior among young men, but not among young women. Over time, the economic needs of women, especially young women, became an issue. Increased labor force participation by women and the "feminization of poverty" made policymakers and others aware of the importance of providing meaningful employment and training opportunities to young women. This interest has been reinforced by studies that indicate that unemployment among young women can have a deleterious effect on their future employment and earnings (Taggart and Linder, 1980~. In the absence of intervention, however, many young women will not have favorable labor market experiences. This is especially true for black women; the data indicate that labor force and employment conditions for black teenage women have been deteriorating over the past 25 years (Stormsdorfer, 1980; Swinton and Morse, 1983~. This paper reviews the participation of young women in employment and training programs. The first section describes the variety of programs that youths have participated in, the level of participation by young women, and the characteristics of young female participants compared with their male counterparts. The second section reviews the type of service received by participants and examines program outcomes. The final section summarizes the findings and suggests subjects for additional research. Margaret C. Simms is director of the Minorities and Social Policy Program at The Urban Institute. The opinions expressed in this paper are the sole responsibility of the author and do not necessarily represent the views of The Urban Institute and its sponsors. 462
463 YOUTH PARTICIPATION IN EMpLo~qENT AND TRAINING PROGRAMS The federal government's post-World War II involvement in employ- ment and training programs began with the Manpower Development and Training Act (MDTA) of 1962. Government training programs designed especially for youths started with the creation of the Neighborhood Youth Corps (NYC) and the Job Corps through the Economic Opportunity Act of 1964. When the Comprehensive Employment and Training Act (CETA) was passed in 1973, most of the existing youth programs were included in the consolidation of employment and training activities although they remained separate activities. In 1977 the Youth Employment and Demonstration Projects Act (YEDPA) added additional programs targeted on youths. This legislation expired in 1981 and all youth training programs were subsumed under the Job Training Partnership Act (JTPA) in 1982. (Job Corps and the Summer Youth Employment Program (SYEP) are separate programs under JTPA.) Young people have not only enrolled in youth programs but have also participated in the full range of employment and training programs since the mid-1960s. Between 1965 and 1972 youths were between 20 and nearly 50 percent of enrollees in such programs as the PUD] ic Employment Program, the Work Incentive Program (WIN), MDTA, Job Opportunities in the Business Sector (JOBS), and the Concentrated Employment Program (CEP). Youths (under age 22) constituted between 48 and 62 percent of enrollees in CETA Title I in the years 1975 to 1981 and more than 20 percent of enrollees in Titles II and VI during the same years.1 Female Participation Studies of participation in employment and training programs by women indicate that women have not been treated equally over the history of these programs. Between fiscal 1965 and fiscal 1978, women were less than one-half of program participants (Harlan, 1980~. In fiscal 1978 women were 45 percent of enrollees in locally operated programs. By the last quarter of fiscal 1979, women were more than 50 percent of enrollees in all local programs except on-the-job training. Very few studies have focused exclusively on participation by young women, but estimates of their participation can be constructed from data on youth programs and from female and youth participation in adult programs. Table 1 shows female participation in youth programs prior to the 1973 passage of CETA and for selected programs under CETA and YEDPA. Before 1973, the largest youth program was the Neighborhood Youth After 1979, Titles I, II, and VI should be interpreted as Titles II-B and C, II-D, and VI, respectively. No data are yet available on JTPA enrollment. See Burbridge (1983) for a history of youth participation in employment and training programs.
464 TABLE 1 Female Participation in Selected Youth Programs, 1965-1980 Percentage Female 1965 1972 1978 1980 1965-1972 Neighborhood Youth Corps In-school youths Out-of-school youths Job Corps 1978-1980 Job Corps Youth Employment and Training Programs Youth Community and Conservation Improvement Projects Summer Youth Employment 45.2 47.7 27.0 29.0 51.3 24.8 51.4 NA 52.5 24.8 48.1 SOURCES: L.C. Burbridge (1983) and S.E. Berryman, W.K. Chow, and R.M. Bell (1981). Corps. Over 4.5 million youths enrolled in NYC between 1965 and 1972, and between 45 and 48 percent of them were young women. In the much smaller, residential Job Corps program, young women were only 27 percent of enrollees. By 1978, women were 51 percent of enrollees in the summer youth program (SYEP) and in Youth Employment and Training Programs (YETP), but still lagged in Job Corps enrollment. Even so, they may still have been underrepresented in SYEP and YETP, since it ~. ~ is Yea anal cney were be percent or the eligible population (Berryman et al., 1981~. Female enrollment in Youth Community Conservation and Improvement Projects (YCCIP) was only 24.8 percent in both 1978 and 1980, although it is estimated that young women were about 46 percent of the eligible population for YCCIP. This discrepancy is thought to be due to sexual stereotyping, since YCCIP involved intensive manual labor, more so than did other programs (Burbridge, 1983~. Overall, young women have done slightly better than adult women in terms of participation in employment and training programs. However, this is primarily the result of their greater representation in youth programs. In adult CETA programs, women have been a smaller proportion of enrollees under baa 29 then Shiv have he - n of "nrm1 1^~= =~ O~ -~A over (Table 2). ~.z ^ ~ ~ ~ ~ _ _ _ _ ~ ~ ~ ~ ~ ~ ~car ~ ~ ~ ~ _, An examination of the enrollment of youths in CETA and other employ- ment and training programs by race and sex reveals some minor differ- ences by sex, but far less than might be expected. Data from the 1977 Continuous Longitudinal Manpower Survey (CLMS) and Current Population Survey {CPS) reveal that women were just under 47 percent of the
465 TABLE 2 Female Participation in Adult CETA Programs, 1978 Women as a Percentage of Participants Aged 22 and Over Under Age 22 Title I 52.5 48.1 Title II 41.6 34.7 Title VI 37.4 Total 45.5 48.6a aIncludes Summer Youth Employment Program, Youth Community Conservation and Improvement Projects, and Youth Employment and Training Projects. SOURCE: S.E. Berryman, W.K. Chow, and R.M. Bell (1981~. enrollees in CETA programs and 50.3 percent of the youth population (aged 22 and under) (Table 3~. Most of the discrepancy was among white women, whose participation in CETA was well below that of white men. Black and Hispanic women, whose CETA participation exceeded their representation in the population, participated at rates comparable to those of their male counterparts. The National Longitudinal Survey (NLS) of Young Americans (see Borus, 1983:120) produced roughly comparable estimates of participation in all government employment and training programs by youths between the ages of 14 and 21 during 1978 and early 1979. Characteristics of Program Participants In addition to concern over their lower participation in government- sponsored employment and training programs, it has been said that the young women are better qualified than the young men who enter the programs. This may occur because there is more "creaming among women than among men. In creaming, the program sponsor selects those applicants who, in the sponsor's view, are the best qualified of those eligible to participate in the program. The young women who are selected into the program may have to meet higher standards in terms of education or prior experience. However, there could also be differences between young men and young women in terms of who chooses to apply for the employment and training programs. For example, if young women have different perceptions about their likely participation in the labor force as adults, they might also differ in their interest in enrolling in employment and training programs. In a study done for The Rocke- feller Foundation, The Urban Institute examined the characteristics of young women who participated in employment and training programs and compared them with the characteristics of young men who participated
466 TABLE 3 Percentage Distribution in the CLMS and CPS Populations by Race and Sex for Individuals Under Age 23, Fiscal 1977 Percentage of CLMS Percentage of CPS Females Males Females Males White17.8024.1640.4840.45 Black23.1823.936.946.36 Hispanic5.944.962.922.86 Total46.9253.0550.3449.67 NOTE: These calculations exclude other minorities (i.e., nonblack, non-Hispanic minority groups). SOURCE: Data from L.J. Bassi, M.C. Simms, L.C. Burbridge, and C.L. Betsey (1984). and nonparticipant youths. The objective of the study was to identify possible differences and to determine which differences, if any, affected the probability of participation in government-sponsored employment and training programs (Simms and Leitch, 1983~. The Urban Institute study was based on data from the National Longitudinal Survey of Young Americans. The NLS is a good data set for analyzing youth participation in employment and training programs because it allows one to compare participants with nonparticipants. The data set includes more than 12,000 individuals aged 14 to 21 in 1979. The survey oversampled minorities and low-income whites, the groups most likely to participate in employment and training programs. Just over 2,000 respondents had participated in at least one government- sponsored employment and training program prior to the 1980 survey interviews. Although participation rates were higher among minority groups--one-third of the blacks interviewed had participated in a program, compared with 11 percent of whites and 24 percent of Hispanics--there were no substantial differences by sex (Table 4~. In general, young women in the sample had higher levels of educational attainment than young men. Overall, young men in the sample were more likely to have less than a high school education (49.2 percent of the men versus 44.5 percent of the females in the sample) and were less likely to have completed any formal education beyond high school (19.2 percent for males versus 23.8 percent for females) (see Table 5~. This differential was similar for all ethnic groups, except Hispanics. For those who had participated in government programs, the educational gap was much wider. Among participants between January 1978 and spring 1980, there were large differences between men and women; a larger percentage of the female than the male participants had
467 TABLE 4 Percentage of the Population Participating in Employment and Training Programs Prior to 1980 Men Single Multiple Women ~. Single Multiple Program Programs Total Program Programs Total White 7.6 3.3 10.9 7.5 3.8 11.3 Black 21.3 11.8 33.1 19.6 12.3 31.9 Hispanic 13.4 10.7 24.1 15.1 8.3 23.4 Other 12.0 5.1 17.1 7.6 5.2 12.8 NOTE: Based on National Longitudinal Survey sample of youths aged 14 to 21 who answered this question. SOURCE: M.C. Simms and M.L. Leitch (1983). completed high school and acquired some college education. With the exception of Hispanics, the gender differences in educational attainment were greater for government-program participants than for nonpartici- pants. Educational gaps are also apparent when another data set is used. For example, a comparison of 1977 youth enrollees in CETA with youths in the Current Population Survey reveals that while young men in general were more likely to be high school dropouts, the differences between men and women in CETA programs were somewhat greater, at least among whites: white male enrollees had dropout rates that were at least twice as high as those of white women who were enrolled in CETA programs (Bass) et al., 1984~. To investigate the possibility that young women who entered government programs had different qualifications because of differences in willingness to participate in the program, Simms and Leitch (1983) analyzed the attitudes of young women toward work as part of the Rockefeller study. Since young women have two options not available to young men--childbearing and, generally, work in the home--young women as a group might be less interested in employment and training programs. Researchers had hypothesized that women who participate in employment and training programs have less traditional attitudes than nonparticipants and are also more likely to expect to be in the labor market for most of their adult years. This was the case for women who participated in programs prior to 1978. However, since it is possible that program participation and maturation affected their attitudes, it is more useful to concentrate on those who participated after January -~978, since the information available on respondents' attitudes is most likely to precede participation. Here we find that women who were participants in government programs after 1977 were no less likely to
468 TABLE 5 Highest Grade Completed, by Sex, 1979 MalesFemales Total Sample 100.Oa100.Oa Less than high school 49.244.5 High school 31.531 .8 More than high school 19.223.8 Enrollees in Government Programs after 1977 100.0100.0 Less than h igh school 80.469.4 High school 8.927.8 More than high school 10.72.8 NOTE: Based on National Longitudinal Survey sample of youths aged 14 to 21 who answered this question. aTotals do not add to 100.0 due to rounding. SOURCE: M.C. Simms and M.L. Leitch (1983). think that a woman's place is in the home than were women who did not participate in training programs. And they were more likely to think that a woman's place is in the home than those who participated in private or military training programs (Table 6~. However, when asked what they expect to be doing at age 35, government-program participants were more likely to say "working," indicating that they expect economic reality to be a factor in their actions. Another factor that might affect participation by young women is family responsibilities. Young women with children or other family responsibilities may not be able to participate in programs, either because they lack child care or because they cannot take time away from nonmarket work. In 1979 the vast majority of the respondents in the NLS sample lived with their parents. Young women, however, were less likely to live with their parents than young men and were more likely to be married or living on their own. There were large racial differences in living arrangements beginning at age 18; blacks of both sexes were much less likely to live on their own than members of other racial or ethnic groups. The women in the sample, both married and unmarried, were much more likely to have children than the men in the sample; 17 percent of the women in the sample had at least one child, compared with 6.9 percent of the men. There were 496 women in the sample who were heads of households (7.8 percent of all women in the sample). Participants in government training programs were no more likely to be married than participants in the sample as a whole. They were, however, more likely to have children. A significant proportion of
469 TABLE 6 Women's Responses to Statement "A Woman's Place Is in the Home," by Participant Status (in percentages) Time and Strongly Strongly Participant Status Disagree Disagree Agree Agree Pre-1978 Nonparticipantsa 39.5 Participants in other training programs Participants in government programs Post-1978 Nonparticipants 39.0 Participants in other training programs Participants in government programs 44.5 11.6 4.3 41.1 47.3 8.5 3.1 46.7 42.4 7.3 3.6 44.8 11.7 4.5 53.4 39.3 5.9 1.4 39.3 45.5 10.7 4.5 NOTES: Based on National Longitudinal Survey sample of youths aged 14 to 21 who answered this question. aTotals do not add to 100.0 due to rounding. SOURCE: M.C. Simms and M.L. Leitch (1983~. women who were heads of households had participated in some type of employment or training program. The last phase of the study of the determinants of participation in employment and training programs was a multivariate analysis. The main objective of this analysis was to determine which factors are likely to affect participation in employment and training programs, whether those factors differ for men and women, and whether after adjusting for all the relevant factors there is still a sex differential. Included in the regression analysis were independent variables in six broad categories: (1) background and demographic characteristics, (2) education, (3) family responsibilities and attitudes, (4) financial need, (5) work experience (including prior participation in training programs), and (6) local employment conditions. Based on past studies of women's participation in adult programs, we expected to find that young women were less likely to participate in employment and training programs and that young women who did partici- pate in programs were likely to be better qualified than their male counterparts. In our regression analysis, we found some evidence to indicate that young women were less likely to be enrolled in government-sponsored employment and training programs, other things
470 being equal (Table 7~. This is clearly the case for enrollment in the pre-1978 period. After January 1978, the significance of sex is not as clear. In equations that did not include participation prior to 1978 as an independent variable, the coefficient for sex was negative and significant. Once prior participation was entered into the equation, the coefficient for sex, while still negative, became insignificant. The interpretation of this finding is not obvious. On the one hand, it may mean that previous participants in programs were more adept at obtaining a slot in another program, and since women were less likely to have had that prior experience, they fared worse than their male counterparts. On the other hand, it may mean that sex differences in participation continue to exist and are similar in nature to those in existence prior to 1978 and the lagged variable is picking up this connection. To the extent that it exists, differential participation by women does not seem to be related to perceptions about future participation in the labor market. None of the variables that were used to measure work expectations proved to be significant. In contrast to women who entered private or military training, women in government programs were no less likely than nonparticipants to think that a woman's place is in the home. Therefore, there is no evidence to indicate that women who enter government programs have expectations of greater labor force attachment than those who do not or that different attitudes about work might explain why there are differences in participation between young men and young women. The presence of dependents is negatively correlated with participation for both men and women. In general, the variables that are important in explaining participation are the same for men and for women. For the most part, the relationships are consistent with a "scraping" hypothesis rather than a "creaming" hypothesis. The probability of participation is negatively correlated with total family income and socioeconomic status (measured by father's education). It is positively correlated with prior enrollment in remedial education, with the number of periods of no work, with the number of months on welfare (for women), and with being black or Hispanic. We found no evidence to indicate that the women who entered government programs were better qualified than their male counterparts. PROGRAM TREATMENT AND PROGRAM OUTCOMES An analysis of female participation in employment and training programs would be incomplete without an assessment of the treatment young women receive, how it differs from that of young men, and the effects of those treatments on some set of outcomes. This section focuses on analyses that have been done in this area. Although a variety of data sources were used, most analyses relied on information available from the CLMS and from the 1979 NLS because those data sets include enrollees in different types of employment and training programs and also provide information on groups who have not participated in programs.
471 TABLE 7 Summary of Ordinary Least-squares Runs on Participation in Government Training, Post-1978, With Lagged Participation Variables Full Sample Males F Statistic Ba FB Statistic Females F B Statistic AGE79b-.01415.55*-.01713.35*-.0092.93** BLACK.10174.89*.09231.18*.11143.84* HISPANIC.06219.36*.0322.62.09020.47* OTHERACE-.0110.36.0060.05.0230.81 FAMINCOME-.00338.36*-.00328.98*-.0029.52* SEX-.0111.37------- EMPLOYRATE1.0111.55.0141.16.0080.35 EMPLOYRATE2.10110.90*.1137.08*.0964.74*** MARRIED.0471.34-.0630.90-.0380.53 DIVORCED-.0913.46**-.0890.92-.0902.37 DEPENDENTS-.0589.82*-.0433.03**-.0787.66* HIGHGRADE1-.0232.35-.0251.47-.0221.10 HIGHGRADE2-.0170.67-.0301.10-.0100.13 WORKEXPECT-.0010.00-.0180.39.0180.42 NOWORK.04158.17*.03828.30*.04328.14* LIVEWITH1-.0240.82-.0481.55.0060.03 LIVEWITH2-.0130.09-.0410.46.0230.15 LIVEWITH3-.0809.91*-.12211.08*-.0350.97 FATHERGRAD1-.0317.76*-.0334.67***-.0293.39** FATHERGRAD2-.0429.69*-~0425.08***-.0465.49*** MOTHERWORK1-.0110.77.0020.01-.0272.20 MOTHERWORK2-.0010.00.0130.44-.0160.64 FEMALEWK14-.0275.65***-.0232.04-.0303.45** FAMILYATT.0010.01.0020.05-.0020.05 REMEDIAL.0368.25*.0231.97.0537.57* MOSWELFARE.0064.69***.0030.25.0074.69*** GOVTPRE78.219214.62*.216112.87***.22097.52* PRIVPRE78.0250.88.0130.12.0421.21 CONSTANT 0.478 0.596 0.318 NUMBER 6,172 3,208 2,964 R2 .14 .15 .13 . *F significant at .01 level. **F significant at .10 level. ***F significant at . 05 level. aUnstandardized coefficient. bVariable names on following page. SOURCE: M.C. S. imms and M.L. Leitch (1983~.
472 TABLE 7 (continued) Variable Definitions AGE79 BLACK HISPANIC OTHERACE FAMINCOME SEX EMPLOYRATE1 EMPLOYRATE2 MARRIED DIVORCED DEPENDENTS HIGHGRADE1 HIGHGRADE2 WORKEXPECT NOWORK LIVEWITH1 LIVEWITH2 LIVEWITH3 FATHERGRAD1 FATHERGRAD2 MOTHERWORK1 MOTHERWORK2 FEMALEWK14 respondent's age in 1979 respondent's race; white is the omitted variable total family income of respondent's household in 1978 respondent's sex 1979 unemployment rate for labor market of current residence; rates under 6 percent is omitted variable marital status, single is omitted variable number of dependents in 1978 highest grade completed by respondent in 1978; less than high school is omitted variable work expectations in 5 years periods of no work in 1978 who individual lived with at age 14 highest grade completed by respondent's father; less than high school is omitted variable number of hours respondent's mother workedin 1978; zero hours omitted adult female in household worked for pay when respondent was 14
473 TABLE 7 (continued) FAMI LYATT REMEDIAL MOSWELFARE GOVTPRE7 8 PRIVPRE7 8 respondent's attitude toward the statement "a woman's place is in the home" whether respondent took remedial math or English or English as a second language months on welfare participated in a government-sponsored employment and training program before 1978 participated in a private employment and training program before 1978 It should be noted that analyses using these data sets examined treatment of and outcomes for individuals who were enrolled in government-sponsored employment and training programs in the late 1970s. There are two reasons for this. First, there is generally a lag between the time data are collected and the time data sets are made available for use. This lag may be up to two years in some cases. Second, if one wants to examine postprogram outcomes, additional time must elapse so the enrollee can leave the program and follow-up data can be collected on the status of the participant at various points after leaving the program. Consequently, postprogram data on employ- ment, earnings, and other outcomes for a person leaving a training program in 1978 might be collected in 1979 and 1980 and be available for analysis in 1981. These lags make it difficult to measure the effectiveness of current programs if they are different from those offered in the recent past. However, evaluation of past programs can be helpful in identifying patterns of treatment and in determining whether some types of programs are more effective than others. Program Treatment Three aspects of program treatment are of interest: program assignment, services received, and type of training received. The type of activity an individual is assigned to and the training received can affect postprogram outcomes. Support services, such as medical care, transportation, and child care, can affect the ability of an individual to enter or continue a program.
474 Program Assignment Employment and training programs consist of four basic activities: classroom training; on-the-job training (OJT); public service employment (PSE), which was discontinued in 1981; and work-experience programs. There are differences in the assignment of men and women to these activities. Although the differences are greater among adults, they do exist among youths. Young women are more likely than young men to be assigned to work-experience and classroom-training activities and are less likely to be assigned to OJT and (prior to 1981) PSE (Table 81. These differences are more apparent among youths over age 18, since those under age 18 are most likely to be in work-experience programs, regardless of gender. Sex differences are slightly greater among whites than among blacks or Hispanics because minorities of both sexes tend to be more likely to be placed in work-experience programs or classroom training after the age of 18. Differences in program assignment do not necessarily represent discriminatory treatment. They may be based on differences in the type of treatment deemed appropriate, given the individual's background, or they may be the result of differences in preferences between men and women. In their analysis of the type of skills training received, researchers at Ohio State University noted that young women in the NLS sample were more likely to receive classroom training--be it college preparatory, skills training, or basic education--and men were more likely to receive on-the-job training (Bows, 1983:126~. They thought that these differences could stem from preferences on the part of young people; young women do better in school and are therefore more likely to accept such an assignment. However, this would certainly be at odds with a strategy of assigning enrollees on the basis of need, since male enrollees tend to have more educational deficiencies than female enrollees. An alternate explanation may be found in the fact that the type of jobs traditionally held by women are more likely to require skills that are learned in the classroom and those held by young men are more likely to be learned on the job. A Rand Corporation study of sexual equity in CETA (based on the CLMS) used a different approach to evaluating the gender differences in program assignment (Berryman et al., 1981~. This study compared the proportion of males who received the CETA services they requested to the proportion of females who received the services they requested. Although the differences between young men and young women were smaller than the differences between adult men and adult women, some patterns were found. In the categories of job training and jobs, men who requested OJT and PSE were more likely to get those assignments than women who requested them. On the other hand, young women who requested classroom training in basic skills were more likely to get it than young men who requested it (Table 9~.
475 U] t) .,, - U] .- a, U] .,' ,' ~S a) v ~: .,, .,' ~: o P~ ~: E~ C) oo m ~: E~ U] S~ S o a' ., Y ~ ~ V O `: 3 a) ·_I s s~ a.' 3 O X D:1 a a V V .,' .,' o Q ~ Q4 :s U] ~ 1 c: S 4U 1 ~Q ~ O O ~ E~ O O `:: L4 U] U] ~ ~ ~S U E~ s4 S 4~ O a ._, 4 £ a' ~ V 3 .,, s O 4 ~ 8 V V >, ., .,l o Ll Q] a ~U) 1 a) .- s ~ · - Q ~ O O ~ E~ O O ~ · - U] U] ~ U' .- ~s ~: C: E~ f~ In ~ In u~ ~) C~ ~\ oo a:) ~ oo ~ ~ a, · . · ~ oo ~ CD ~ CN ~ C~ c~ ~n oo ~ oo · . · ~ 0 Lr, ~ ~ 0 1- Cn 1-d'~ ~1 ~)U ~1 ~ ~1 ~1 o u~ ~10o t- ~1 1- · . . ... · .· · . u~ ~D ~ ~ C ~c~ In ~ ~ C~ L~ N 1- ~ t- · . · . 00 ~ C~ 00 C~ ~ C~ c~ r~ r~ ~N · · · ~ O kD · · · ~ In c~ 1 ~ ut oo 1 ~ ~ t_ · · ~ a, ~r1 ~ c~ o ~ ~1 ~ a~ . . · · · ~ C~ 00 CO 00 ~r · . · . c~ r~ c~ 0o ~ ~ IO er ~o · . · ~ ~ ~ O ~ u~ I O,- C~ · . · ~ Lr C~ C~ oo ao L~ 1 ~ ~1 0 1 ~ o u~ · · ~ 0o ~ ~ C~ U~ k0 · . · ~ ~ CO C~ ~0 ~ 00 ~ 0D ~ O · . · ~ U) ~ U~ C~ O · . · ~ ~ ~ ~ O 1 l~ 1 k9 U ) o, · · ~ ~r Ln rn In 0D · . · ~ u~ In C~ ~ ~ c~ 0~ ~ r~ 0a~ ~ co 0 a~ ~ 0 cs~c~ U~ ~ cn~ ~ ~ ~° · . · .. . · .. . . . 00 ~ ~ (D~ ~ ~ a, ~ c~ ~ ~o~ 0 ~ 0 u~ ~ c~ 0O~ rn ~ ~ r~ · . · .· . · .· . · . 1- ~ ~CS~ 1 - 0 1 -0 L~ u~ ~ (D ~ - cn~1 ~ ~ U)1 rn 1- CO ~l ~ ~l ~l ~r Lf~ ~r1 t~ u~ · . · .· . · .· · ~ a~c~ ~ ~ ~r oo ~ c ~ 1 ~ d a~C~1 In kD ~1 l- 1- 1 Ln ~ I_a~ O ~1 oo1 a, a . · . ·. · . · · ~ ~ 0 ~ ~ ~n kD ~ r~ u~ ~ ~ ~ ~D · . · ~ ~r Ln ~ ~ ~ c~ ~J 1 1 1 1 n~ vo ~ ° -4 ~4 r~ ~ ~ 3 ~r oo a, - ~ 0 ~ ~r · . · ~ ~r ~ 0 u~ In [_ Y 1 1 1 1 'D o C~ · . · ~ v ·,~ ,~ I_ a~ c~ C 05 1 1 1 1 a, 0 U] _, ~: a) U] m 0~ k4 a). - ~S~ v m S~ ._t~ ~U] V C: · -U) ~C) . ~ a)£ Y OUl 3u' m U] a) v H o U]
476 TABLE 9 Youth Distribution of Obtained CETA Services within Desired Service (in percentages) In-Program Assignment Basic Skills Job Training Job Sex FY76 FY77 FY78 FY76a FY77 FY78 FY76a FY77 FY78a Classroom Male 37.3 55.S 71.0 35.3 42.7 43.1 9.1B.08.0 training Female 81.4 74.1 71.8 34.8 54.6 52.5 8.610.511.6 (CT)b On-the-job Male 4.4 2.3 3.4 8.6 9.7 12.1 7.2 10.6 8.1 training Female 3.5 1.5 2.7 3.7 4.2 5.7 4.5 6.3 6.2 (OJT)C Youth work Male 56.8 41.4 22.3 experience Female 15.1 22.6 23.4 (YWE)d Public service Male employment Female 0.0 1.8 (PSE)e 1.5 0.5 50.339.235.1 63.4 52.5 56.8 51.936.236.6 74.4 63.2 63.6 5.88.49.7 20.3 28.9 27.1 9.55.05.2 12.4 20.0 18.7 NOTE: Includes only participants who expressed their desired CETA services. aTotals do not add to 100.0 due to rounding. bIncludes CT and YETP CT in fiscal 1978. CIncludes OJT and YETP OJT in fiscal 1978. dIncludes YWE, YETP Other, and YCCIP in fiscal 1978. eIncludes PSE sustainment, PSE nonsustainment, and PSE unknown in fiscal 1978. SOURCE: S.E. Berryman, W.K. Chow, and R.M. Bell (1981:67). Occupational Segregation Both the Rand study and the Ohio State analysis found considerable segregation in occupational training under CETA and other government programs. Among youths enrolled in employment and training programs in 1978, Crowley et al. (Bows, 1983:20, 162-163) found that 80 to 85 percent of all enrollees in professional, clerical, and sales training programs were women, while 78 percent of the enrollees in skilled labor and craft training were men. The low percentage of women in the latter programs is not unexpected given that only 25 percent of women in the 1979 NLS survey aspired to atypical jobs. And among those young women who were interested in nontraditional jobs, most were interested in managerial and professional careers; few indicated an interest in blue-collar jobs. Young women from more disadvantaged families, who are the target group for government programs, are less likely to aspire to nontraditional occupations and, without encouragement, may be unlikely to pick employment or training slots in traditionally male fields. Berryman et al. (1981:44-45) also found CETA enrollees to be very traditional in their job preferences, although that seemed to decrease
477 somewhat with time. In 1976, 82 percent of the young women wanted traditionally female jobs and only 4 percent wanted traditionally male jobs. The remaining 14 percent wanted jobs that were categorized as mixed (neither predominately male nor predominately female). By 1978, the proportion wanting traditionally female jobs dropped to 60 percent, those desiring traditionally male jobs rose to 15 percent, and those seeking mixed jobs rose to 24 percent. Over the same period, CETA began to place more young women in traditionally male jobs and fewer in traditionally female jobs, and the shifts were greatest for minority females. Supportive Services The service received most often by youths is job counseling. About one-half of both men and women received counseling. Medical services, transportation, and child-care services were not received by large proportions of the youth population, but the need for such services among the youth population is not known. Female heads of household were more likely to receive health and child-care services than others, but less than one-half of that group received any services in 1978 (Bows, 1983; Simms and Leitch, 1983~. Program Outcomes The expected outcomes of program participation have been and continue to be diverse. In addition to increased postprogram earnings (the primary goal for adults), other objectives include increased educational attainment (lower school dropout rates), a reduction in early childbearing, reduced welfare dependency, and reduced criminal activity. Success in achieving these multiple goals, however, is often difficult to measure. To assess the net impact of program participation, information is needed on the outcomes and variables, other than program participation, that are likely to affect outcomes for both the preprogram and post- program periods. Moreover, comparable information is needed for a group of individuals who have similar preprogram characteristics but who were not enrolled in the program. Such comprehensive information is seldom available. Therefore, the outcomes examined most often are postprogram earnings and employment because more data are available on those outcomes, both for program participants and for individuals who may be part of a comparison group. Earnings and Employment Gains One fairly consistent conclusion, at least in evaluations of outcomes for adults, is that women and the economically disadvantaged receive the greatest gains from participation in employment and training
478 programs. 2 This was true for the early programs established under MDTA and has continued to be true under CETA (Perry et al., 1975; Bassi, 1982; Congressional Budget Office and National Commission for Employment Policy, 19821. There is no consistency, however, in the assessment of which program activities have the greatest effect or the mechanism by which those gains are made. For example, some studies conclude that OJT and skills training have the highest payoffs (Harlan, 1980), but other studies point out that the employment gains of those two programs decay over time (U.S. Department of Labor, 1977~. A recent study using the Continuous Longitudinal Manpower Survey, however, found that the net earnings gains for women do not vary by program activity but are in the range of $800 to $1,300 for all programs. Moreover, the gains do not appear to decrease over time (Congressional Budget Office and National Commission for Employment Policy, 1982~. Findings from a recently completed study by The Urban Institute (Bass) et al., 1984), which used the 1977 CLMS, indicate that many of the programs that work for adults also work for youths. Participation in PSE programs increased earnings for white women by $882 to $990 in the first postprogram year and by $1,035 to $1,144 in the second postprogram year.3 For black women in PSE the gains were $1,126 to $1,196 in the first year and $608 to $678 in the second postprogram year, and for Hispanic women in PSE the significant gain was $1,705 to $1,862 in the first postprogram year. Black women also benefited from participation in OJT; they showed gains of $861 to $877 in the first year and of $1,389 to $1,406 in the second year after leaving the program. The only significant gain for men was for white men in OJT, who experienced an increase in earnings of $452 to $463 in the first postprogram year. Most of the gains for women were the result of increases in time in the labor force, time employed, and hours worked; only 3 to 10 percent of the gains were attributable to increased average hourly wages. A larger proportion of the gains for white men (16 percent) was attributable to increases in hourly wages and less to additional time employed. Even though women benefited more from CETA participation in terms of gains in earnings, mean postprogram Social Security earnings for young women were lower than those for young men. Youths in other programs either showed no gain or a loss in earnings compared with a matched sample drawn from the CPS. However, even individuals in those programs increased their labor force activity between preprogram and postprogram years and showed gains in reported 2Note that some of the studies cited measured gross earnings and/or employment impact, and others measured net earnings and/or employment gains (using a control group). This changes the magnitude but not the sign of the results for women. 3However, these findings are not definitive since the Chow tests indicated a significant difference between young white women who participated in CETA and their comparison group.
479 hourly wages. It would appear that the participants' failure to make gains relative to the comparison group is due to the fact that the increases in labor force activity of CETA participants are not as large as the gains made by nonparticipants. This may or may not be the result of non-labor-market activity, such as time spent in school. It is impossible to confirm these assumptions, since no postprogram infor- mation on the labor force activity or school status of members of the comparison group was available. Using the 1976 CLMS to measure the effect of program activity on employment immediately after leaving the program, Harlan and Hackett (1984) found that programs that enrolled more men than women (such as OJT) provided the greatest possibility for immediate postprogram employment and that those with the largest proportion of women had the lowest possibilities. If population groups were shifted among programs so that minorities and women were distributed like white men, post- program employment for those groups would increase, although it would still lag that for white men. (No separate analysis was done for youths.) Hahn and Lerman (1983) used the NLS to analyze the effect of CETA programs on school enrollment and unsubsidized job experience. They found that while CETA did seem to increase school enrollment among women and nonwhite men, it had very little positive effect on unsub- sidized employment. Youths who had not been enrolled in CETA had higher rates of unsubsidized employment and had higher earnings from unsubsidized employment. This was especially true for women, although young female CETA participants who mixed school and work had higher unsubsidized earnings per week in the first year. By 1980 young female CETA participants who were both in school and working were more likely to have unsubsidized jobs than their non-CETA counterparts. Other Outcome Measures Very little information on outcome measures other than employment and earnings is available on CETA activities as a group. Bassi et al. (1984) did examine the effect of CETA on welfare dependency for CETA enrollees in 1977 who were between the ages of 18 and 65. The results of the analysis show that CETA does decrease the level of welfare dependency, but it does not lead to removal from the welfare rolls (at least not under the regulations in force in 1978 and 1979~. In 1978 the estimated annual welfare savings for women who headed households was $250. This finding is consistent with the fact that women received the highest gains in earnings from CETA participation. No significant welfare savings were found for men. This is consistent with the finding that there was not, in general, a substantial gain in earnings for men. A recently completed analysis of the long-term effects on youths of government-subsidized employment and training programs used the NLS to examine the impact of participation in five program activities (sub- sidized employment, classroom skills training, basic education, job counseling, and other) on employment, earnings, educational attainment, and welfare dependency (Crowley, 1984~. The effect of participation in
480 1978, 1979, or 1980 on 1981 status or outcomes was found to be insig- nificant for earnings and hourly wages for both men and women, although it seemed to lead to subsidized employment at a later date. Participa- tion in a basic educational program was positively related to obtaining a General Equivalency Diploma (GED) for both men and women and to school enrollment for women. Participation in a skills training program had a negative effect on subsequent school enrollment for young women. Young women who received job counseling (in 1979) or participated in sub- sidized employment programs (in 1980) actually were more likely to be on welfare and to receive larger amounts of welfare than nonpartici- pants. This could be related to greater knowledge of the benefit programs to which they might be entitled. Women who were in programs in 1978 had lower levels of dependency on welfare in 1981, which indicates a possible lag between program participation and movement off welfare among younger women. There have been some evaluations of the effect of individual employment and training programs on a variety of outcome measures. One such study is the evaluation of the Job Corps conducted by Mathematica Policy Research Corporation (Mallar et al., 1980~. This study found that in addition to increasing employment and earnings, Job Corps also increased the probability of high school completion and college enrollment and decreased criminal behavior and welfare dependency. For young women, participation in Job Corps also appeared to delay family formation and to reduce the incidence of extramarital childbearing. The impact on employment, earnings, education, and welfare payments was greater for women without children than for those with children. This may be due to the fact that the burden of family responsibilities on those with children limited their labor force participation after leaving the program. Another program for which a variety of outcome measures has been evaluated is the Youth Incentive Entitlement Pilot Projects (YIEPP). This program, initiated under YEDPA, guaranteed jobs to eligible 16- to 19-year-olds (part-time jobs during the school year and full-time jobs during the summer) if they stayed in or returned to school and met specified attendance and performance standards. A comparison of men and women eligible to participate found that the participation rates for young men and young women were quite similar and so were the average number of months in the program (Farkas et al., 1984~. There were substantive gains in earnings for both men and women during program participation, mainly due to increased employment. However, the difference in earning gains between young women and young men in the postprogram period was significant--gains in weekly earnings for men ($13.66) were twice those for women. Since nearly one-half of the young women in YIEPP had had at least one child by the time they reached the age of 19, it was thought that home responsibilities may have had a negative effect on labor force participation among women. However, the rate of childbearing, while quite high, varied sub- stantially from site to site within YIEPP, and on average, the 45 percent rate for YIEPP participants was comparable to the 47 percent rate for the comparison group e The study concluded, therefore, that the program had no effect on the rate of childbearing among this group
481 of young, low-income women and that the high rate of childbearing probably explained at least part of the difference in earnings gains between male and female participants in YIEPP. The latter conclusion seems to be consistent with the fact that the gap between the weekly earnings of men and women in the program increases as they grow older. Even though the gains to young women are smaller than those to young men, however, female program participants still do better than women in the comparison group. While YIEPP did increase labor force participation and lower unemployment, it did not seem to increase school enrollment. This may have been due, in part, to the failure to attract or retain high school dropouts. However, the program did not appear to increase dropout rates either (which had been true of some other programs), since there was a school enrollment requirement. A program that puts less emphasis on job training, per se, and more emphasis on a system of supportive services is Project Redirection, jointly sponsored by the Ford Foundation and the U. S. Department of Labor. This program sponsors or brokers services for pregnant teen- agers and teenage mothers (under age 18) who are without high school diplomas and who are in welfare families. The final report on the impact of Project Redirection has not been released yet but findings from the 12-month follow-up study indicate positive results from program participation in terms of employment and education, with slight decreases in the incidence of pregnancy (Polit et al., 1983~. Since 12 months is a short follow-up period for evaluating program outcomes, especially since many individuals were still enrolled in the program, the findings must be regarded as tentative. What may be of greater interest is that the program provides an effective set of support services, such as child care and housing assistance, that frequently are not available in other programs and are obviously felt to be needed by teenage mothers. SUMMARY AND CONCLUSIONS Government employment and training programs have been utilized by both young men and young women over the past 20 years. While differ- ences in participation based on sex were quite clear during the early years, they are less clear now. A review of young women's participa- tion in government-sponsored employment and training programs reveals that the level of participation has increased in recent years to a level approaching parity in most programs, although a few programs, like Job Corps, still lag the others. Although young women who enroll in government programs have somewhat higher educational levels than young men, this appears not to be a significant factor in program enrollment. The factors that affect young women's enrollment appear to be quite similar to those that affect young men. However, there still appear to be differences in the treatment received by young women and young men. Women are more likely to be involved in classroom training or work-experience programs that are less likely to integrate them into the job market, and they continue to be trained in traditionally female
482 occupations. Few young women receive supportive services other than counseling, and even women with children receive low levels of support in terms of child care. Young women, like their adult counterparts, benefit more from participation in employment and training programs than do young men. They receive higher employment and earnings gains from the activities in which they are least likely to be assigned, such as on-the-job training and public service employment. Most of these gains come from increased time employed, however, and not from higher wage rates. Evaluations of selected programs, like Job Corps, also reveal gains in such areas as educational attainment, reduced welfare dependency and criminality, and delayed family formation. Women with children seem to benefit less from the programs than those without children, perhaps because family responsibilities prevent them from increasing their postprogram labor force participation. This review of studies of young women's participation in employment and training programs has identified several shortcomings in both our knowledge and in the operation of employment and training programs. These shortcomings prompt the following research recommendations: 1. Relatively little is known about the effects of employment and training programs on nonemployment outcomes, such as educational attain- ment, welfare dependency, and childbearing patterns. For youths, especially young women, these outcomes may have greater long-term economic consequences than the impact programs may have on short-term employment and earnings. Therefore, more research should be done on nonemployment outcomes and their link to long-term employment and earnings gains. While the 1979 NLS has advantages as a data set because it includes nonparticipants in the sample, precise information on the programs in which the enrollees participated is scanty. Program-based data sets, such as the CLMS or the new JTLS, will provide better information on programs and services received. However, the CLMS comparison group (drawn from the Current Population Survey) lacks longitudinal information on outcomes other than participants' earnings. It is to be hoped that the Job Training Longitudinal Survey (JTLS) data base, to be developed under JTPA, will not have the same shortcomings. In the interim period before the JTLS is available, it would be possible to conduct research on nonemployment outcomes by drawing a comparison group from the NLS for use with the late CLMS cohorts (1979 and later), since information is available for roughly comparable periods. 2. Another area in which we lack knowledge concerns the low level of support services provided to youths. For example, is child care not provided because it is not requested or because it is not available? To what extent does nonavailability of child-care services or transportation reduce program participation, especially among young women? [An on-going Urban Institute study (Sonenstein and Wolf, 1985) will provide insight into the relationship between the availability of child care and employment or participation in education or training programs.]
change. 483 In terms of program operation, there are also suggestions for 1. The fact that young women are assigned to different programs than young men (and disproportionately to those with lower expected earnings gains) suggests that program operators need to be more sensitive to possible gender differences in program assignment and that more emphasis on preprogram counseling might be needed. 2. Related to the above, while today's young women are more likely to expect to be working during their adult lives than earlier genera- tions of young women, they are not as open to nontraditional careers as they need to be if they are to increase their earning power signifi- cant~y. Employment and training programs (particularly job counseling programs) should include information about job opportunities in non- traditional careers and the skills and education needed to enter those careers. Support and encouragement may also be needed to get more young women into the training "pipeline" for those occupations. 3. Low-income young women have relatively high birth rates during their teenage years. While little is known about the effect of participation in employment and training programs on subsequent childbearing, we do know that young women with children are less likely to participate in programs and may receive lower benefits from participation. More emphasis needs to be placed on outreach and on facilitating the participation of young mothers in programs. Moreover, support services (like child care) need to be available to these women after they leave programs in order to increase their postprogram labor force participation. REFERENCES Bassi, L.J. 1982 CETA--Is It a Cost-Effective Method for Increasing the Earnings of Disadvantaged Workers? Washington, D.C.: The Urban Institute. Bassi, L.J., M.C. Simms, L.C. Burbridge, and C.L. Betsey 1984 Measuring the Effect of CETA on Youth and the Economically - Disadvantaged. Washington, D.C.: The Urban Institute. Berryman, S.E., W.K. Chow, and R.M. Bell 1981 CETA: Is It Equitable for Women? Santa Monica, Calif.: Rand Corporation. Borus, M.E. 1983 Tomorrow's Workers. Lexington, Mass.: Lexington Books. - Burbridge, L.C. 1983 Employment and Training Programs for Youth: An Interpretation . ~ and Synthesis for Measured Outcomes. Washington, D.C.: The Urban Institute. Congressional Budget Office and National Commission for Employment Policy 1982 CETA Training Programs: Do They Work for Adults? Washington' D.C.: U.S. Government Printing Office.
484 Crowley, J.E. 1984 Long-term outcomes of government-subsidized employment and training programs. Pp. 179-212 in P. Baker, S. Carpenter, J.E. Crowley, R. D'Amico, C. Kim, W. Morgan, and J. Melgosz, ems.. Pathwav to the Future. Vol. IV. Center for Human Resources Research. Columbus: Ohio State University. Parkas, G., R. Olsen, E. Stromsdorfer, L.C. Sharpe, F. Skidmore, Smith, and S. Merrill 1984 Post-Program Impacts of the Youth Incentive Entitlement . Projects. New York: Manpower Development Research Corporation. Hahn, A., and R. Lerman 1983 The CETA Youth Employment Record: Representative Findings on the Effectiveness of Federal Strategies for Assisting Disadvantaged Youth. Final report to the U.S. Department of Labor. Center for Employment and Income Studies. Waltham, Mass.: Brandeis University. Harlan, S.L. 1980 Sex Differences in Access to Federal Employment and Training Resources under CETA: An Overview. Working Paper no. 58. Wellesley, Mass.: Wellesley College Center for Research on Women. Harlan, S.L., and E.J. Hackett 1984 Job Training Programs and Employment Outcomes: Effects by Sex and Race of Participants. Working Paper no. 129. Wellesley, Mass.: Wellesley College Center for Research on Women. Mallar, C., S. Kerachsky, C. Thornton, M. Donihue, C. Jones, D. Long, E. Noggoh, and J. Schore 1980 Evaluation of the Economic Impact of the Job Corps Program. Second Follow-up Report. Washington, D.C.: ~ ~ ~ ~ ~~~ D.A. ups. Department of Labor. Perry, C.R., B.E. Anderson, R.L. Rowan, and H.R. Northrup 1975 The Impact of Government Manpower Programs. Philadelphia: University of Pennsylvania. Polit, D.F., M.B. Tannen, and J.R. Kahn 1983 School, Work, and Family Planning. New York: Manpower Demonstration Research Corporation. Simms, M.C., and M.L. Leitch 1983 Determinants of Youth Participation in Employment and Training Programs with a Special Focus on Young Women. Washington, D.C.: The Urban Institute. Sonenstein, F.L., and D.A. Wolf 1985 Child Care Policies and Employment Behavior of Welfare Mothers. Washington, D.C.: The Urban Institute O ' Stromsdorfer, E.W. 1980 The effectiveness of youth programs: an analysis of the historical antecedents of current youth initiatives. Pp. 88-111 in B.E. Anderson and I.V. Sawhill, eds., Youth Employment and Public Policy. Englewood Cliffs, N.J.: Prentice-Hall.
485 Swinton, D.H., and L.C. Morse 1983 The Source of Minority Youth Employment Problems. D.C.: The Urban Institute. Washington Taggart, R., and B. Linder 1980 Youth employment policy background material. In Vice President's Task Force on Youth Employment, A Review of Youth Employment Problems, Programs, and Policies. Vol. 2. Washington, D.C.: U.S. Government Printing Office. U.S. Department of Labor 1977 The Work Incentive (WIN) Program and Related Experience_. Washington, D.C.: U.S. Government Printing Office.
