Students Leaving College in General and Science, Mathematics, and Engineering in Particular: Some Causes-And Some Remedies
Persistence is obviously an essential component in successful completion of undergraduate education. Comparisons of 1991 and 1993 enrollment profiles in lower and upper divisions respectively  by sex and race/ethnicity, indicate changes in the composition of student groups, changes that would not happen if all groups progressed at identical rates. Although enrollment of all minorities in higher education is up overall, comparison of enrollment by level suggests that underrepresented minorities quit without completing degrees in higher proportions than do white and Asian students. (See appendix tables 3-19 and 3-20.) These figures indicate only general trends, however, and fail to show the important effects of in-and-out or part-time attendance and transfer students. Minority students dropped out between divisions in uneven numbers. Blacks had the highest rate of attrition. Enrollment percentages of white, Asian, and nonresident students rose slightly from lower division (1991) to upper division (1993).
Longitudinal data on science and engineering dropout rates are unavailable, but studies by Seymour and Hewitt, 1994 (see SIDEBAR: Students with Disabilities Studying Science, Engineering, and Mathematics: The Time Disadvantage), Seymour and Hunter (in press, and see SIDEBAR: Students with Disabilities Studying Science, Engineering, and Mathematics: The Time Disadvantage), and Steele (1995; see SIDEBAR: A Burden of Suspicion: How Stereotypes Shape the Intellectual Identities and Performance of Women and Blacks) offer some insights on attrition in these fields. Many students who enter college planning to study science, mathematics, and engineering change their plans. An analysis of information from undergraduates on seven college campuses who switched out of such majors-and others who persisted-identified 23 factors influencing such decisions (Seymour and Hewitt 1994). Despite many concerns shared by both men and women, substantial differences by gender suggest that they approach college with different goals and experience their undergraduate education differently.
The students who switched agreed on their top five overall concerns, but men and women differed on the rank of their importance. Nine out of 10 of those who left science, mathematics, and engineering were concerned about pedagogy; however, men and women defined good teaching differently. Even women with good academic records often felt their academic performances were not "good enough," unless they had a satisfying personal relationship with one or more of their teachers. Unfortunately, such relationships were reported to be rare. (See NSF 1994, p. 46, and text table 3-1 for details.)
Striking differences appear among reasons why students from particular ethnic/racial groups left science, mathematics, and engineering. Minority and majority students differed about their reasons for switching. Students of color tended to blame themselves for switching, whereas white students more often pointed to institutional failures. For example, white students complained of poor teaching and curriculum overload more than twice as often as did minority students. Many minority students reported that they had been " over-encouraged to enter technical fields for which they were underprepared." These findings suggest a need for better presentation of what science, mathematics, and engineering majors and careers require. (See NSF 1994, p. 48 and text table 3-2.)
 Lower division students (sometimes called freshmen and sophomores), formally matriculated, have earned fewer than half the number of credits needed to graduate, usually under 60 hours in a 120-hour degree program. Upper division students (sometimes called juniors and seniors) have earned more than half of the necessary credits but have not yet graduated. These categories apply only to baccalaureate students in general and can only suggest changes in the status of particular students.