Process of Undergraduate Education
The number of students enrolled as undergraduates in all institutions of higher education has increased substantially over the last decade, from 10.6 million in the early 1980's to 12.6 million in 1991.
(See appendix table 5-2.) The numbers rose for both men and women and for all racial/ethnic groups. These trends in enrollment  portray growing diversity within the
student population, and provide a context for considering the outcomes by discipline areas.
- Courses Taken
More than half (56 percent) of the higher education students in 1991 were enrolled in 4-year institutions. An even higher proportion of total full-time enrollment-72 percent-was at these institutions.
Full-time enrollment is the predominant pattern of enrollment at 4-year institutions: More than three-quarters of the students were registered as full-time undergraduates. Full-time enrollment constitutes an important factor in participation in
higher education because of its obvious relationship to progress toward an earned degree and continuity of academic study. Thus, monitoring the full-time study of population groups of interest and the composition of the population of full-time
students provides an important statistical basis for considering access to higher education. Women comprised 52 percent of the students enrolled on a full-time basis at 4-year schools, having been in the majority since the mid-1980's at both 2-year
and 4-year institutions. (See appendix table 5-3.) Minorities comprised 21 percent of full-time enrollment at 4-year schools; underrepresented minorities, 16 percent. 
Within 4-year institutions, the increase in full-time enrollment was 11 percent overall. Over the past decade, the increase was greater for women than for men: The number of women enrolled as full-time
students increased from 2.4 million in 1980 to almost 2.8 million in 1991; among men, the increase was from 2.4 million to 2.5 million. Underrepresented minorities enrolled full time increased nearly 30 percent.
Similar percentages of women and minorities were enrolled full time-nearly three-fourths. (See appendix table 5-3.) The percentage of full-time enrollment for
underrepresented minorities and for whites was the same, 75 percent of their total enrollment, a generally stable figure during the decade. Almost 81 percent of the Asian students were enrolled full time.
Since 1980, underrepresented minority women have been more likely than their male counterparts to be enrolled as full-time students. By 1991, Asians were the only racial/ethnic group for which women
remained under half (49 percent) of the enrolled full-time students at 4-year institutions.
Trends in enrollment of first-year and first-time first-year students are important indicators for future enrollment in higher education. They not only reflect to some extent the size of the population
traditionally entering college, but they also provide indicators of changing higher education enrollment choices by students. The number of full-time first-year students at all institutions has declined since 1980. (See appendix table 5-4.) Although numbers of both men and women have decreased, the loss has been less for women; the drop in enrollment for men has been 7 percent, compared with a drop of less than 4 percent for women. At
4-year institutions, the number of men enrolled declined 11 percent, the number of women, 5 percent.
The decline in first-year enrollment was not shared by all racial/ethnic groups, but was exclusively due to a decrease in enrollment of whites, both men and women. Enrollments of black and American
Indian students fluctuated, but 1991 enrollment levels exceeded those of 1980. (See appendix table 5-4.) The number of Asian and Hispanic students of both sexes increased. The largest increase over the decade was
among Asians, and among Asian women in particular.
An even more focused view of changes in college entrance patterns is provided by examining the trends for first-time, full-time first-year students at 4-year institutions. (See appendix table 5-5.) By 1991 the numbers of men and women enrolled as full-time, first-time first-year students had decreased by 10 percent and 3 percent, respectively. The decline, however, was due entirely to
decreases in enrollment for whites. Enrollment for all minorities, both Asians and underrepresented minorities, increased. (See figures 5-1a and 5-1b.) The number of Hispanic full-time,
first-time first-year students at 4-year institutions, both male and female, rose by almost 40 percent over the decade. The numbers of black men and women enrolled were higher in 1991 than in 1980 following some intermediate decreases: 5 percent
higher for men, 6 percent higher for women. Although the numbers remain small, the increase for American Indians was 23 percent overall.
Persistence is obviously essential for completion of undergraduate education. Some students drop out. Others change fields of study, in many instances switching from highly structured and rigorous fields
to fields requiring less extensive prerequisite knowledge. Rates of attrition, from higher education altogether and from study in science and engineering, are believed to be uneven among groups underrepresented in science and engineering
occupations. Thus, an examination of attrition rates within higher education is an important supplement to an examination of enrollment trends.
While enrollment of minorities in higher education is up overall, there is some evidence that they leave higher education without completing degrees in higher proportions than do white students. Many
students leave higher education between the lower division and the upper division.  Comparisons of enrollment profiles in each division for gender and racial/ethnic groups, while not actually showing the
results of longitudinal analyses, nonetheless indicate changes in the composition of student groups that would not be encountered if all groups progressed at identical rates. In 1988, 1.5 million women were enrolled as full-time lower division
students; in 1990, 1.2 millon women were enrolled as full-time upper division students, 17 percent fewer. (See appendix tables 5-6 and 5-7.) A similar decline of 18 percent occurred
in the enrollment for men between the lower and upper divisions.
The drop in recorded enrollments for racial/ethnic groups was uneven. (See figures 5-2a and 5-2b.) The percentage change was
greater for each of the underrepresented minorities than it was for Asians. Approximately 40 percent fewer black students enrolled in 1990 as upper-level full-time students than enrolled in 1988 as lower-level full-time students, suggesting a high
attrition rate. Declines of approximately one-quarter of the number of Hispanic students and of more than one-fifth of the American Indians also occurred. This is in contrast to a 15 percent drop in white enrollment and a 3 percent decrease in
Asian enrollment. (See appendix tables 5-6 and 5-7.)
Figure 5-2a Figure 5-2b
Undergraduate attendance patterns are important in examining participation of underrepresented groups in higher education for several reasons. Attendance affects the time necessary for the completion of a
degree. Time spent on study also imposes costs, both the direct costs of education and costs of possible foregone earnings. In addition, the disciplines of science and engineering are particularly hierarchical, building on prerequisites; a loss in
continuity can be particularly detrimental.
College attendance is commonly perceived as following a "traditional" pattern. According to this pattern, students graduate from the school where they began their studies, having attended school full
time and without interruption for 4 years. This model has provided the basis for much of the current college curriculum, especially in many science and engineering disciplines (with some notable exceptions, such as formal "co-op" programs as part of
an engineering curriculum). It appears that students who follow the traditional pattern generally have a higher probability of success, defined as satisfactory completion of a bachelor's degree.
Variations from traditional attendance patterns occur in several ways: (1) part-time enrollment (fewer courses per term than would be needed to complete a bachelor's degrees in 4 years); (2)
intermittent enrollment (taking no courses for at least one term between formal entrance into an institution and graduation); and (3) transferring from one institution to another. These variations each have differential effects on college careers.
In reality, college attendance patterns are changing. Only slightly more than half of all enrolled students now follow the traditional pattern  and there is
very little difference in the proportions of men and women following traditional attendance patterns, 52 percent and 51 percent, respectively. (See appendix table 5-8.) Students from minority groups are slightly
less likely to have followed the traditional enrollment pattern, between 47 and 50 percent of each racial/ethnic group, while 53 percent of whites follow the traditional pattern.
Students who majored in most science or engineering fields are more likely to have followed traditional attendance patterns than students generally, with percentages ranging from over 50 percent to
almost 75 percent. Only in mathematics and computer science do students follow a variety of attendance patterns, with fewer than half (48 percent) completing traditional patterns. (See figure 5-3.) In these fields there
was also very little difference between men and women as traditional students, 47 percent and 48 percent. Among engineering majors, however, a substantially higher percentage of women than men completed degrees with traditional enrollment patterns.
(See figure 5-3.) Students not following the traditional pattern often had enrollment characterized by more than one of the variations.
Grade distributions of college graduates show that a B average is now the median grade. About one-third of all students who graduated in 1991 had a grade point average (GPA) between 3.0 and 3.49, or
between B and B+. (See appendix table 5-9.) Somewhat more students were in the adjacent lower range, 2.5 to 2.99 (31 percent) than were in the highest range, 3.5 and above (20 percent). An analysis of GPA
distributions between all students and those with science and engineering majors shows little difference. Students in science and engineering graduated with high proportions registering GPA's of 3.0 or better: mathematics and computer sciences, 54
percent; engineering, 52 percent; and physical and life sciences, 62 percent.
There were more striking differences among racial/ethnic groups completing science and engineering degrees than between men and women. In each of the possible attendance variations, proportions ranged
from below to considerably above the proportions for all students, suggesting that the undergraduate experience is highly varied and that generalizations about groups should be tentative at most.
The greatest differences among racial/ethnic groups in science and engineering fields appeared in engineering, where profiles of attendance patterns varied considerably. Between 55 percent and 60
percent of each racial/ethnic group followed traditional enrollment patterns, but the distribution of nontraditional patterns was distinct for each group. White engineering majors were more likely to attend school intermittently than were Asians (20
percent to 10 percent), but less likely to do so than black engineering majors, 30 percent of whom attended intermittently.
Variations in grades occur between men and women. Although women tend to score lower on the Scholastic Aptitude Test as they enter college, they tend to graduate with higher grades than men, regardless
of major. Fifty-nine percent of the women receiving bachelor's degrees in 1991 had a GPA of B or better, compared with approximately 47 percent of the men. (See figure 5-4.)
The amount of coursework in a particular discipline is generally considered an indication of exposure to, and familiarity with, the discipline. Information on the number of courses taken in science and
engineering disciplines is therefore of interest in considering both majors and non-majors in science and engineering. In the case of majors, this information may indicate depth of knowledge and preparation for further study. In the case of
non-majors, it could signal the feasibility of transferring into a field, as well as a general knowledge of the discipline. Other factors, including the level and content of the coursework, also offer critical dimensions of evidence that a student
possesses knowledge of a field, so a measure of coursework alone must be taken as only one dimension of preparation. 
The pattern of higher grades for women prevails in science and engineering fields as well as overall. For example, nearly two-thirds of female mathematics or computer science majors achieved a GPA of B
or higher, compared with fewer than half of the men who majored in those fields. In engineering, a higher percentage of women (63 percent) than men (49 percent) had GPA's of B or better.
By field, and by race/ethnicity, the distribution of college grades varied considerably. Grades for white and Asian students were generally higher than grades for students from other racial/ethnic
groups. Life and physical sciences majors generally had higher grades than students in other science and engineering fields, regardless of race/ethnicity.
Course-taking by Majors
Women and men showed only small differences in the number of courses completed in their major. In mathematics and computer science, a slightly higher percentage of women than men took 13 or more courses in
their major. They took about the same amount of coursework in the social sciences as their male counterparts, as was the case in the life and physical sciences. On the other hand, female engineering majors took fewer engineering courses than men.
In life and physical sciences, Asians took the most courses, with 91 percent taking 13 or more. Students with this major generally took many courses, regardless of their sex or racial/ethnic group.
(See figure 5-5.) Among mathematics and computer science majors, differences once again were small among racial/ethnic groups. Among the groups studied, engineering majors exhibited the greatest variation.
Course-taking by Non-majors
The differences in course-taking patterns in particular science and engineering fields by non-majors far exceed the differences across gender or race/ethnicity, even though the latter are in some instances
quite marked. Most striking is that so few non-majors take engineering courses; at least 95 percent of women, Hispanic students, and black students did not take a single engineering course. (See figure 5-6.) Reasons for
this phenomenon could be the availability of engineering courses, or requirements for formal entry into the study of engineering, including prerequisites. In any case, engineering does not appear to offer the opportunities for "sampling" that occur
in other science and engineering fields. In comparison, for example, courses in the social sciences are completed by all but 5 to 7 percent of most student groups. Well over half of each gender and racial/ethnic group took five or more courses in
social sciences. Physical and life sciences and mathematics and computer sciences fall between engineering and the social sciences in the numbers of courses taken by non-majors. (See figure 5-6.)
1. The enrollment data for the complete population of higher education students are from the U.S. Department of Education, National Center for Education Statistics, Integrated Postsecondary
Education Data System (IPEDS) Fall Enrollment Survey. This annual data collection obtains information from all accredited institutions of higher education in the country; imputation is made for nonresponding units. Survey data are available only by
level and type of enrollment, not by major field of study.
2. Underrepresented minority groups in science, mathematics, and engineering are blacks,
Hispanics, and American Indians.
3. The notion of lower division and upper division is used here to classify enrolled students on the basis of their
progress towards a degree. Lower division includes those students formally matriculated who have earned fewer than half the number of credits needed to graduate, e.g., 60 hours in a 120-hour degree program. Upper-division students are those who have
earned more than half of the necessary credits, but have not yet graduated. The data discussed here compare cohorts enrolled at 4-year institutions and can only suggest changes in the status of particular students.
4. The data on attendance, grades, and course-taking presented in this chapter are drawn from the Curriculum Assessment Service (CAS), a data base of the transcripts of 1991 spring
graduates of American colleges and universities. The project has been conducted at the Institute for Research on Higher Education, University of Pennsylvania, under the direction of the American Association of Colleges (University of
Pennsylvania/Institute for Research on Higher Education 1994a), with funding by the Pew Charitable Trust, the National Science Foundation, and the National Endowment for the Humanities.
5. Results discussed here are incomplete for technical reasons related to the data base: course credits granted to transfer students are not included in the course counts. Also, courses for which students "tested
out" by demonstrating competence in an area are not included.