A r c h i v e d  I n f o r m a t i o n

Outcomes for Noncompleters

Critics of the college movement argue that many college noncompleters do poorly in the labor market and would have been better off entering community colleges and occupational programs. This section examines economic outcomes for the more than 600,000 college entrants who do not graduate, comparing their performance with that of bachelor's degree graduates and that of noncompleters from 2-year colleges. It also compares other outcomes for noncompleters from 2- and 4-year colleges.

In assessing college outcomes, it is important to be aware of the ongoing debate about the extent to which college adds value by developing skills that the marketplace rewards and the extent to which it just "picks the winners." Human capital theory holds that college (and education in general) adds value, e.g., that the skills learned in college lead to greater productivity and greater compensation in jobs later on (for example, see Becker 1975). On the other hand, screening theory holds that colleges just select and certify more able individuals who would do as well in the labor market if they had no additional education (see Spence 1973). This process saves employers screening costs in hiring and adds to the compensation of graduates and attendees. There is an extensive literature on this debate, including controlled comparisons of college graduates and nongraduates who are similar in many respects. College graduates are generally found to earn more than similar nongraduates. However, even with statistical controls, it is possible that unobserved characteristics of college graduates, rather than their educational experience, explain the additional benefits they receive. Appendix D provides a brief review of some major research findings on the issue. We believe that both processes are at work in postsecondary education but that the human capital explanation probably accounts for more of the variation in labor market outcomes than does screening.

Economic Outcomes

In this section we compare the wages and other earnings of noncompleters and degree-holders with the same amount of education, and we also compare the earnings of noncompleters from 4-year and 2-year colleges with similar levels of education.

Noncompleters and graduates. It comes as no surprise that college graduates earn more than those who attend college but do not complete their programs (see Appendix A Table A-1). After all, the graduates have more education. Similarly, bachelor's degree holders earn more than associate degree holders, for much the same reason (e.g., Surette 1997, Grubb 1995e, Kane and Rouse 1995b). A better question is whether college graduates earn more than attenders who have the same amount of education but no degree. Signaling theory holds that a degree's ability to certify skills and other desirable characteristics, such as perseverance, saves employers screening costs and increases the graduate's compensation. Table A-2 presents data from 6 studies that compare the wages and longer-term earnings of degree holders with those of noncompleters who have as much education but no degree. The studies are multivariate, comparing the earnings of individuals who are similar in many ways other than their college attendance and completion patterns. (Since wages are hourly earnings, we simplify the discussion by using the term "earnings" for both kinds of compensation, unless there is a need to distinguish between them.)

Across the studies examined, male bachelor's degree graduates earned more than similar noncompleters with 4 years of college credits, except in the case of vocational majors. Grubb (1995b) found that nongraduates from 4-year vocational programs earned at least as much as graduates from those programs.

For females, evidence of a bachelor's degree effect was mixed. Half of the studies reviewed showed women with bachelor's degrees earning more than women with just 4 years of college. The other half showed degree holders earning the same amount or less. As with males, Grubb (1995b) found that noncompleters from 4-year vocational programs earned as much as completers.

In the case of associate's degrees, most studies showed female graduates earning more than noncompleters with a similar number of credits. Indeed, Grubb's (1995b) SIPP analysis found female associate degree holders earning as much as or more than comparable females with 3 years of college, regardless of where the 3 years were obtained. There were exceptions to this general pattern, however. One was Kane and Rouse's (1995a, 1995b) finding that the earnings (but not the wages) of female associate degree holders were not significantly different from those of female noncompleters with two years of college. Another exception was Grubb's (1995b) finding that female graduates from academic (as distinct from vocational) programs earned less than females with just 2 years of college.

Evidence regarding the additional value of associate degrees for men was mixed. Half of the studies showed male associate degree graduates earning more than noncompleters with 2 years of college. The other half showed them earning the same amount or less.

In sum, the research shows that a bachelor's degree adds to earnings for males, and an associate's degree, for females. Other findings are complex and inconsistent across studies. However, there is some evidence that noncompleters with vocational majors do well, compared to graduates, except for females in 2-year colleges.

Noncompleters in 4-year and 2-year institutions. We turn from a comparison of graduates and noncompleters to a comparison of noncompleters from different kinds of institutions and programs. There have been a number of controlled studies that enable us to compare the economic value of a year spent at a 4-year college with the value of a year spent at a 2-year or less-than-2-year institution, for students who leave without graduating. Kane and Rouse (1993, 1995a, 1995b), Grubb (1995b), and Whitaker and Pascarella (1994) used data from the National Longitudinal Survey of 1972 (NLS72), collected 14 years after the baseline was conducted. Breneman and Nelson (1981), Anderson (1984), and Monk-Turner (1988) also used NLS72 data collected 4, 7, and 10 years after the baseline, respectively. Surette (1997) analyzed data from the National Longitudinal Survey of Youth (NLSY) collected 10 years after the 1979 baseline. In addition, Schochet (1991) examined earnings outcomes for young males collected 7 years after their initial 1979 interviews in the NLSY. He compared returns to one year and two years of "college" (whether 2-year or 4-year) with those to government training programs, military training, and off-the-job vocational training. Table A-1 shows the pertinent results from all these studies except Schochet (1991), whose data are not comparable.

In general, 4-year college students who left after acquiring a year of credits could expect to receive earnings and wages lower than or equal to those of similar community college students with the same number of credits. There are 2 exceptions to this generalization. First, Monk-Turner found that entrants to 4-year colleges earned significantly more than entrants to 2-year colleges. However, her study did not compare noncompleters at 2-year colleges with noncompleters at 4-year colleges. Rather, her sample included both completers and noncompleters from both types of schools. The inclusion of bachelor degree holders in the sample may have been responsible for the significant earnings advantage of the 4-year college entrants. The second exception to the generalization is that 4-year college students who completed a year of vocational credits tended to do better than community college students and 4-year students with an equivalent number of academic credits.

A year at a 4-year college was also associated with earnings that were lower than or equal to those from a year at a private 2-year college. Further, Grubb's (1995b) data show that, for males, a year's attendance at a 4-year college was less valuable than a year in an academic program at a technical institute. His other estimates of earnings associated with a year at a technical institute or a proprietary school were not significantly different from the earnings associated with high school graduation.

Schochet (1991) used a choice model for young males to analyze the returns to college attendance and to participation in several forms of training. The study carefully addressed methodological questions such as possible selection biases, the role of unobserved variables, the use of Armed Forces Qualifying Test (AFQT) scores as a measure of ability, and the possible endogeneity of the training variables. Using logit and probit analyses, he estimated the returns to college attendance, off-the-job vocational training, training in government programs, and military training. With controls for ability and many other characteristics, he predicted hourly wages, salaries, and weeks worked per year separately for high graduates and dropouts, both white and black, who participated in these education and training programs.

Schochet found that across all these different estimates, participants in one-half year of training received hourly wage gains equivalent to those of similar individuals with one year of college (either 2-year or 4-year). He further found that one-half year of training yielded increases in salaries and weeks worked equivalent to those from 1?2 years of college. While these estimates seem remarkably high, Schochet emphasized that they represent short-term returns and that longer-term outcomes might be different.

The data in these studies make it clear that critics of the college movement have a point in questioning the economic value of attendance at a 4-year college for noncompleters. Students who take vocational courses in these institutions seem to do reasonably well, but others might be better off choosing a two-year institution. Even occupational training programs lasting a year or less may pay bigger dividends in the short run.

Cognitive Skill Gains

Four-year college noncompleters might have done well to consider attending 2-year colleges for another reason. Repeated tests conducted by the National Center on Postsecondary Teaching, Learning, and Assessment have shown that, when ability, time in college, and other confounding variables are controlled, 2-year college students make about the same gains in reading comprehension, mathematics, critical thinking, writing skills, and science reasoning as 4-year college students (Pascarella et al. 1998, Pascarella et al. 1995?96, Bohr et al. 1994, and Terenzini et al. 1994). In short, a year at a community college produces about the same cognitive gains as a year at a 4-year college.

Critics of the college movement also argue that many college dropouts (and some graduates) are burdened with debt accumulated from student loans that could have been avoided by choosing other education and training options. Their relatively low-level jobs and earnings make it difficult to repay the debts, in this view.

Debt Burden

Choy et al. (1997) conducted a study of postsecondary debt burden using data from the Baccalaureate and Beyond Study (B & B: 93/94) and the Beginning Postsecondary Student Longitudinal Study (BPS: 90/94). Table 3 provides data from the latter on debt burden accumulated by students who first enrolled in postsecondary institutions in 1989?90 and had either graduated or left without graduating by 1994.

Table 3 Debt burden from postsecondary education, by educational attainment

Among noncompleters, students from 4-year colleges were much more likely than those from 2-year colleges to have some debt (19 percent vs. 4 percent). However, students who left for-profit schools were the most likely to be repaying debts (30 percent).

Among completers, between 24 and 30 percent of those who received bachelor's degrees, associate's degrees, or occupational certificates had some debt. Graduates with associate's degrees were somewhat less likely than those with bachelor's degrees or certificates to have student loan debt.

Noncompleters with debts from 4-year colleges and for-profit institutions had relatively low payments each month–$75 and $72. The sample of those leaving 2-year institutions with debt was apparently too small to permit estimates of their monthly repayment amounts. Among graduates, monthly debt repayments were highest for bachelor's degree holders–$136–and lowest for certificate holders–$71. For most groups, repayment obligations were equal to about 9 percent of salary. Certificate holders were the exception, obliged to pay only 6.8 percent of salary.

Based on these data, we conclude that noncompleters from 4-year institutions are much more likely to leave school with some debt than are their counterparts from 2-year institutions. While the repayments are not large, this difference nevertheless adds weight to the critics' argument that 4-year college noncompleters might have been better off attending community colleges.

[Changes in College Completion Rates]

[Outcomes for College Graduates]