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

Changes in College Completion Rates

Another part of the argument against the college movement is that completion rates in 4-year colleges are lower than they used to be and dropout rates, higher. Estimates of college completion and dropout rates vary widely, depending on definitions, sample specifications, reporting methods, and other factors. Some studies define completion as attainment of a bachelor's degree, others define it as completion of 4 years of college, still others, a combination of elements. Studies vary in the length of time allowed for college completion to occur – 4 years, 5 years, 6 years, or more. Some studies include just full-time students, while others include both full-time and part-time students. Some measure completion at the institution first entered, others measure it at any college. In most studies, the unit of analysis is students, but in some, it is institutional reports of retention. In assessing changes in completion and dropout rates, then, it is important to be sure that the phenomena measured and the methods of measurement remain the same over time. The best way to do this is to examine changes reported in single studies or in series, each of which has the same definitions, sample specifications, reporting methods, etc. at different points in time.

Several studies have examined changes in college completion rates, using systematic national data (American College Testing Program 1996, Astin et al.1996, Grubb 1989). One common measure from these studies is the proportion of full-time students entering a 4-year college who graduate from the same institution within a given period of time. This is useful information, but it has some limitations. If we assume that the issue in question is the aggregate 4-year college completion rate, the fact that part-time students are excluded from these estimates probably biases them upward, while the fact that completions are defined as graduation from the college of initial entry (rather than all colleges the student may attend) probably biases them downward. In assessing changes over time, however, such biases are often tolerable, as long as the samples, methods etc. remain constant.

Figure 3-Completion rates over time in college of initial entry

We will first examine completion rates at the college of initial entry. Then we will turn to the few statistics describing changes in rates of dropout from one institution and from all institutions attended. Figure 3 shows that rates of completion at the college of initial entry over given periods of time have declined moderately since 1970.

In surveys by the Cooperative Institutional Research Program (Astin 1972, 1990, 1996, Green et al. 1983), 4-year graduation rates showed a net decline from 1970 to 1989. An estimated 46.7 percent of 1966 college entrants had graduated by 1970, while 39.9 percent of 1985 entrants had graduated by 1989. However, there was a slight apparent increase from 1988 to 1989. In another study, Grubb (1989) using data from the National Longitudinal Survey of 1972 and the High School and Beyond survey, calculated the 4-year graduation rates of students entering college from the high school classes of 1972 and 1980. He found that 30.8 percent of the 1972 entrants had earned bachelor's degrees at their college of initial enrollment within 4 years. The 1980 entrants had a little lower graduation rate, 27.4 percent. Other data come from the American College Testing Program, which surveys colleges each year, asking institutions to report on their retention and completion rates. The American College Testing Program (1996) found that graduation rates over 5-year periods at 4-year colleges declined from 57.5 percent in 1983 to 53.3 percent in 1996.³¹

Grubb (1989) reported that there was no difference in 4-year dropout rates at the college of initial enrollment for those who entered in 1972 and 1980 — 19.4 percent and 19.3 percent respectively. (Completion rates can fall while dropout rates remain the same over 4 years because at the end of this period, some students are still in college or have attained other credentials.³² ) However, across all colleges — not just the college of initial enrollment — Grubb found some increase in the dropout rate: 25.1 percent of the 1972 entrants had dropped out 2 years later, as compared to 28.8 percent of the 1980 entrants.

In other measures of persistence, the American College Testing Program (1996) found that freshman-to-sophomore dropout rates from the college of initial entry increased from 24.5 percent in 1983 to 26.9 percent in 1996. Along the same lines, Postsecondary Education OPPORTUNITY (1996) reported that freshman-to-sophomore continuation rates decreased from 68 percent in 1983 to 66.9 percent in 1995.

Some of the modest decline in college completion rates over a given period of time — whether 4, 5, or 6 years — is due to the fact that the average college student is taking longer to complete a degree. NCES found that between 1977 and 1990, the proportion of bachelor's degree graduates who completed their education within 4 years of high school fell from 45 percent to 31 percent, while the proportion completing degrees more than 6 years after high school increased from 25 percent to 32 percent (Smith et al. 1996).³³

Although noncompletion rates have not changed greatly over the last several decades, they may still be a matter of concern. How many students leave college each year without getting degrees? We can make a rough estimate from the completion rates over time, just discussed, and from single-time completion rates reported in other studies. Table 2 shows these rates.

The rate of graduation from 4-year colleges can be approximated by examining the estimates in bold for later measurement years (1994?96) and longer times from point of entry (5 or more years). These range from 44.9 percent to 57.1 percent. Periods longer than 6 years do not seem to increase completion rates much. Astin et al. (1996) found that completion rates after 9 years were less than a percentage point higher than rates after 6 years.³⁴ Based on the data in bold print, we conclude that somewhere around half of the freshmen entering 4-year colleges eventually graduate.

In 1995, NCES estimated that there were 1,151,000 first-time freshmen in 4-year colleges (Snyder et al. 1998). ³⁵ If half of them eventually graduated, that would leave 575,500 who did not. Noncompletion among transfers from community colleges, who are less likely than first-time freshmen to graduate, would increase this number somewhat (see Dougherty 1992, Pascarella and Terenzini 1991). If we take transfers into account, we would guess that over 600,000 students who enter 4-year colleges and universities each year, whether freshmen or other students, fail to graduate.

There is an extensive literature on factors in persistence in 4-year colleges, most of it reviewed in Pascarella and Terenzini (1991). Among individual factors, one of the most important for our purposes is educational achievement, a very good predictor of college graduation. High school grades and test scores are strongly related to college completion. For example, Astin et al. (1996) crosstabulated college graduation rates by high school grade and SAT score range. Without exception, the graduation rates increased with both grades and SAT scores. At the extremes, only 10.5 percent of college students with high school grade averages of C or less and SAT totals less than 700 graduated within 4 years, while 80.4 percent of students with high school grades of A or A+ and SAT scores of 1300 or more graduated in that period. Similar findings are reported in Astin et al. (1990) and Green et al. (1983).

Among the institutional factors in college graduation, an important one from our perspective is whether a high school graduate begins his or her postsecondary education at a 2-year or 4-year college. Many studies have found that controlling for demographics, personal background characteristics, motivation, aspirations, and a variety of other variables, high school graduates who enter 2-year colleges seeking a bachelor's degree are less likely to attain one than those who enter 4-year colleges. A comprehensive review by Pascarella and Terenzini (1991) cited as typical the study by Velez (1985), which found that after controlling for family socioeconomic status, gender, race, initial degree aspirations, academic ability, secondary school achievement, college grades, and place of residence, 4-year college entrants had an 18.7 percent advantage over 2-year college entrants in completing a bachelor's degree within 7 years after graduating from high school. More recently, Whitaker and Pascarella (1994) and Pascarella et al. (1998) estimated that on average, across many studies, 4-year college entrants seeking a bachelor's degree had about a 15 percent greater chance of attaining one than did similar 2-year college entrants. It is possible that there are other, unmeasured characteristics of 2-year college entrants that make them less likely than 4-year entrants to earn a bachelor's degree, but to date the evidence suggests that institutional characteristics have an impact on the probability of attaining one.³⁶

In a major study, Clark (1960) argued that high dropout rates in 2-year colleges were the result of a process of "cooling out," in which many 2-year college students of modest ability were quietly diverted from 4-year college. Manski (1989), however, saw the process less as a "cooling out" than as an "experiment" in which many people enroll in community colleges knowingly to test their motivation and ability to succeed in postsecondary education. Along the same lines, Grubb (1996b) noted after studying the intentions of community college enrollees, Many students are there because they are unsure of their options, and the community college is the lowest-cost way of discovering the alternatives and testing their abilities.

In sum, there was a modest decline — from 3 to 7 percentage points — in completion rates and a 0 to 4 percentage point increase in dropout rates among 4-year college entrants over fixed periods between the mid-1970s and the mid-1990s. However, some of this change was due to the fact that students tend to stretch out their college experience more than they used to, taking fewer courses per semester or leaving college and returning. While the decline in retention rates has been small, the proportion of college students who fail to complete is large. We estimate that over half of 4-year college entrants, in excess of 600,000 students per year, leave without getting degrees. High school grades and test scores are among the best predictors of college completion status. Whether one initially enters a community college or a 4-year college is also a factor in persistence.

[Changes in Ability Levels]

[Outcomes for Noncompleters]