![[Table of Contents]](/icons/up.gif){kind=icon}
There is no question that GEDs are more likely to participate in postsecondary education and vocational training than are other dropouts. The number and proportion of dropouts enrolled in degree-granting college programs is minuscule often reported as zero in the studies reviewed. Some dropouts do participate in vocational-technical and other training programs, but the rates of participation are very low. For example, NCES estimated that 0.5 percent of dropouts among the NELS 1988 eighth-graders had enrolled in postsecondary degree programs by 1994, 3.8 percent had enrolled in or completed certificate programs, and 2.2 percent had enrolled in other postsecondary programs (U.S. Department of Education 1996d).
Moreover, multivariate studies have shown that GED certification predicts participation in college and in training programs, controlling for other factors. Murnane et al. (1997) found that attaining a GED within 4 years of leaving high school was associated with an initial 2 percent increase in the probability of attending college for both males and female dropouts. The magnitude of the increases grew over the next several years, at a greater rate for women than for men.[35] The authors controlled for race/ethnicity, mother's education, age and grade at which the student dropped out, and age at GED attainment.
Garet, Jing, and Kutner (1996) found that receipt of a GED was a strong predictor of years of vocational training acquired, especially for women. This analysis controlled for race/ethnicity, family status, work experience, and local unemployment. Further, Kroll and Qi (1995) showed that GED recipients were three to four times as likely as those who failed the tests to enroll in 2-year or 4-year colleges or vocational schools.
Given that attainment of a GED increases a dropout's chances of getting additional education and training, what proportion of GEDs and high school graduates actually attend college? Several studies based on the NLSY have estimated the enrollment of GED recipients and high school graduates in college, whether in 2-year or 4-year institutions (see table 11). Somewhat over 60 percent of high school graduates, both male and female, had attended college by their late twenties or age 30. NCES estimates of college attendance by 1980 high school seniors are similar (U.S. Department of Education 1997). GED females in table 11 were about half as likely as high school graduates to attend college: 30-33 percent did so. GED males were also less likely than their high school graduate counterparts to attend, though there is less consistency in the estimates--28 percent and 42 percent. The data in table 11 all come from the NLSY. However, newer estimates based on the 1994 follow-up to the 1988 National Education Longitudinal Study[36] are roughly similar--32.8 percent of GEDs and 73.3 percent of high school graduates in the sample were enrolled in postsecondary institutions at some time during this period.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||
Greater detail on postsecondary education and on occupational training outcomes of the GED is provided by five national-level studies. Cameron and Heckman (1993) and Cameron (1994) provided detailed breakdowns of postsecondary participation by type of institution. Cervero and Peterson (1982), Kroll and Qi (1995), and Behal (1984) also analyzed postsecondary participation by type of institution. Their work was based on follow-up surveys of GED examinees conducted one and one-half to two years after GED attainment. The response rates were typically low (24 percent, 22 percent, and 35 percent, respectively). The Cervero and Kroll studies had significant nonresponse biases. For example, Cervero and Peterson found that respondents were older, more likely to be female, and more likely to have passed the tests than the population which they represented. The Cervero study and the Kroll study both urged caution in generalizing from their survey results.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Table 12 shows the GED enrollments in 2- and 4-year colleges, vocational/technical institutes, and on-the-job training and apprenticeship programs reported in these studies.[37]
Several observations are in order before we try to interpret these results. First, the college and training estimates in Cameron's study are not mutually exclusive, as they are in the other studies. If a woman attended a community college and a vocational/technical institute, her attendance at both types of institution was counted. Cameron's estimates of education and training experiences are higher than they would have been had he used mutually exclusive categories.[38] Second, Behal's relatively low enrollment rates in 2-year and 4-year colleges are partly explained by the fact that she reported enrollments for GED test takers, while other authors examined enrollments for GED recipients. Those who passed the tests and received GED certification would be more likely to attend college than would all test takers. Third, the Cameron/Heckman and Cameron studies had much longer follow-up periods (8 and 10 years respectively) than the GED studies (one and one-half to two years), a fact that probably helps account for their higher rates of participation in postsecondary education, especially in 4-year colleges.[39]
Despite differences in survey methods, response rates, analyses, and estimates, the broad picture emerging from these data is fairly clear. First, over half of the GED recipients--between 50 and 63 percent--got some additional civilian education or training after they passed the tests, whether in degree-granting colleges, vocational schools, apprenticeship programs, or on-the-job training. In addition, according to Cameron and Heckman, about 10 percent of GEDs entered the military during the period they studied, and enlistees receive occupational training as a matter of course. (Military enlistment is classified as one of the "other outcomes" in table 12. Only Cameron and Heckman broke it out separately.)
Second, most of the education and training that the GEDs received after passing the tests was acquired at the subbaccalaureate level in community colleges and vocational/technical schools. This finding is consistent with our estimate, based on NCES data, that 55 percent of GEDs in postsecondary institutions in 199293 were enrolled in 2-year schools and 23 percent were enrolled in less-than-2-year schools.[40]
Third, in choosing among postsecondary education and training options, GED graduates are strongly oriented toward acquiring occupational skills. The majority of community college students are enrolled in occupational programs (Boesel and McFarland 1994), and GED enrollments in vocational/technical schools (including trade schools) and participation in on-the-job training and apprenticeship programs are substantial. GED participation in the military, with its well developed occupational training programs, further strengthens this emphasis. The focus of GED recipients on acquiring occupational skills is understandable, given that most are adults, that many have family responsibilities, and that labor market returns to high school dropouts are declining rapidly (see figure 2).
Did GED attainment help these students get additional formal education? It almost certainly did. As we have seen, most postsecondary institutions require some kind of secondary certification for admission, and both bivariate and multivariate studies show that GEDs are much more likely than dropouts without the credential to participate in college or vocational education.
Although the GED clearly increases opportunities for postsecondary education, the share of postsecondary enrollments accounted for by GED recipients declined from 7 percent in 1986 to 4 percent in 1992 (U.S. Department of Education 1996a). The reasons for the decline are not clear, but it was evident in less-than-2-year, 2-year, and 4-year institutions.
How well do GED recipients perform in postsecondary institutions? One good measure of performance is grade-point average (GPA). Many studies, including a large number of doctoral dissertations, have compared the grade-point averages of GED graduates and high school graduates at specific institutions. The studies have been conducted at different times and locations for a variety of reasons.[41] Thus, it is not advisable to treat them as samples of a larger population. In examining the results of these studies, it is important to consider the ability and motivational mixes of students at different points in the postsecondary educational process. In general, we expect incoming cohorts, such as freshmen, to include a smaller proportion of able and motivated individuals than would be true of graduating seniors, because the less able and motivated students would tend to discontinue their education over time.
To take the effects of this selective attrition into account, we sort studies that compare the grade-point averages of students into three groups: (1) those that examine the performance of beginners--students in the first year of enrollment only, whether the period covered is a quarter, semester, or year; (2) those that examine the performance of postsecondary graduates; and (3) those that examine student performance in any year of college. The third group includes studies of all enrollees (e.g., freshmen, sophomores, juniors, and seniors) in a given time period as well as studies of incoming student cohorts over time. Some studies in the third group include beginners, but this overlap with the first category is unavoidable, because the beginners cannot be sorted out.
Table 13 presents the mean grade-point averages of GED completers and high school graduates, and the ratios of the means, derived from 7 studies of GED performance in postsecondary vocational programs, 15 studies of their performance in community and junior colleges, and 19 studies of performance in 4-year colleges and universities. The data in table 13 are summary statistics for the results of these studies.[42] The full tables containing the relevant data are found in appendix B-4.
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
As expected, the mean grade-point averages of both GEDs and high school graduates increased with student longevity in these institutions, except between the first year and "all years" in the 4-year colleges. It may be that more rigorous selection procedures in the 4-year colleges identified and excluded weak candidates before admission rather than afterward, in contrast to procedures in 2-year colleges and vocational schools.
The grades of GED students in these institutions were close to those of high school graduates even in the first year. For example, among beginners in 4-year institutions, the difference was roughly that between a C and a low C+. The grades of GEDs and high school graduates (HSG) tended to converge over time, as selective attrition equalized the groups. In vocational programs, which often last only 1 year, this convergence effect was slight; the grade-point averages of GEDs were approximately equal to those of high school graduates both during and at completion of the program. In community colleges, the GED/HSG ratio of grade-point averages increased from .82 to 1.06 upon completion, and in 4-year colleges, it increased from .85 to 1.00. In all three types of schools, GEDs who graduated earned about the same grades as those with regular high school credentials.[43]
Early studies of the postsecondary performance of GED recipients focused on World War II veterans in 4-year colleges. The GED had been designed for veterans, and they were the primary users of the tests for more than a decade after the war. Over time, however, the number of veterans who were test takers declined, and the number of other test takers increased. By 1959, nonveterans outnumbered veterans as GED examinees.
Veteran and nonveteran GED examinees constituted substantially different populations. The veterans were mostly male and white, and by definition, all had military experience. The nonveteran examinees were second-chance civilians, who were more likely to be female and members of minority groups. The differences in these populations led to speculation that the earlier veteran GEDs performed better in college than the later civilian GED recipients. Our data on the grade-point averages of early and later examinees shed some light on this issue. Table 14 shows the mean GPAs of GED recipients and high school graduates in studies of performance in 4-year colleges before and after 1959, along with the ratios of the GED and HSG means. (There were few community colleges before 1959.)
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
These data do not support the hypothesis that early test takers, mostly veterans, earned higher grade point averages than the later, mostly civilian, examinees. In fact, the data suggest the opposite: the mean GPAs in studies done after 1959 were uniformly higher than those in studies before 1959 for both GED students and high school graduates. The GED/HSG grade ratios are similar over time: the performance of GED students relative to high school graduates in each category (first year, all years, graduates) remained fairly constant in these two periods. The higher GPAs after 1959 may reflect grade inflation at these institutions, rather than better performance, but there is no evidence here that the change in test-taking populations affected the performance of GED students in 4-year colleges.
Typically, GEDs enrolled in postsecondary institutions differ from high school graduates in ways that are related to persistence. One NCES study found that GEDs had an average of 4.1 "nontraditional" characteristics that predicted postsecondary attrition, as compared to 1.2 for high school graduates (U.S. Department of Education 1996a). The characteristics were delayed enrollment, part-time enrollment, full-time employment, financial independence (e.g., lacking parental support), single parent status, having dependent children, and lacking a regular high school diploma. Larger proportions of GED recipients than of high school graduates had all of these characteristics except one (full-time employment).
Because the GED selects primarily on cognitive skills, it is not surprising that postsecondary students with GED certification earn about the same grade-point averages as high school graduates. However, because the test battery does not select on "non-traditional" characteristics or on other characteristics related to persistence, we would expect to see more postsecondary attrition among GED recipients than among high school graduates. The fact that GEDs initially dropped out of high school reinforces this expectation. What does the research tell us about the postsecondary persistence patterns of GED students and high school diploma holders?
Persistence in postsecondary education can be seen as reflecting the extent to which individuals achieve educational goals they set for themselves, such as the attainment of a degree or certificate. It can also be seen as reflecting patterns of continuation within institutions, consonant with institutional goals and regardless of individual goals.[44]
Table 15 shows the individual and institutional degree attainment of GED recipients and high school graduates. The lines in bold print represent individual attainment--students in surveys and studies who met the goals they set for themselves. The lines with asterisks represent individuals who exceeded their goals.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
The NCES (1996) data show the institutional attainment of respondents in the Beginning Postsecondary Students Longitudinal Survey (BPS) from 1989-90 through the second follow-up in 1994. The NCES (1997) data show the 1994 attainment of BPS students in 2-year and less-than-2-year institutions who specified the kind of credential they were seeking in 1989-90. The data in appendix B-5 are mean graduation rates from local studies of postsecondary students enrolled in credential-granting programs. The Cameron and Heckman (1993) data are for males in the National Longitudinal Survey of Youth, 1979-87; the authors make the reasonable assumption that students who begin degree-granting programs in 4-year institutions intend to graduate.
Examining the lines in bold, we see that GEDs who sought vocational certificates were almost as likely as high school graduates to attain them; those who sought associate degrees were about half as likely to attain them; and male GEDs who sought bachelor's degrees were very unlikely to attain them. (The .88 GED/HSG ratio from appendix B-5 does not carry much weight, because it is based on 4 local studies conducted before 1959. Other studies based on national surveys and examined in this section show that no more than 5 or 6 percent of GEDs in postsecondary institutions earn bachelor's degrees or complete four years of college.)
As might be expected, the data suggest that attainment rates for individuals specifically seeking credentials (individual rates) are higher than for all students in a given type of institution (institutional rates). This is especially evident among beginners at 2-year colleges.[45] Among associate's degree seekers (in bold), 12.2-14.9 percent of GEDs and 24.8-28.5 percent of high school graduates earned the degree. Among all community college beginners (NCES 1996), 4.3 percent of GEDs and 18.5 percent of high school graduates earned an associate's degree.
In general, table 15 shows that the longer it takes to graduate from a program or institution, the smaller the proportion of GEDs who graduate, relative to high school diploma holders. An interesting exception is the roughly one-half percent of GEDs and high school diploma holders who start vocational programs intending to earn a certificate and actually earn bachelor's degrees.
Table 16 presents another measure of institutional persistence-years of college completed by beginners--at all colleges, 2-year colleges, and 4-year colleges. The data do not indicate what the students' educational goals were or whether they earned degrees.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
On average, GEDs complete fewer years of postsecondary education than high school graduates. According to Garet's (1996) analysis, almost three-fourths of GEDs at age 28 who began at postsecondary institutions completed 1 year or less, as compared to somewhat over one-fourth of high school graduates. Less than 5 percent of GEDs (males and females combined) completed 4 years or more, as compared to just under one-half of high school graduates. Cameron and Heckman's (1993) data for men and Cameron's (1994) data women at age 30 are consistent with this pattern. (The relatively high GED/HSG ratios for men completing 2 years of college (1.36, 1.19) do not mean that GEDs are more likely than high school graduates to finish 2-year programs. It means that they are a little more likely to complete just 2 years of college, not more or less. Cameron and Heckman's data show that 55.7 of high school graduates who began at 2-year colleges completed 2 years or more, as compared to 27 percent of GEDs.)
In all, the data support the hypothesis that GED recipients are less likely than high school graduates to persist in postsecondary education, whether persistence is measured by individual attainment rates, institutional attainment rates, or years of college completed. (Also see U.S. Department of Education 1996b and Murnane et al. 1997.)
-###-
[Labor Market Performance of GED Recipients]