National surveys of college and university undergraduates indicated that student interest in majoring in scientific fields and in pursuing careers in science is at or near its all-time low (Astin et al, 1989; Green, 1989). Despite these declines, a number of earlier studies suggest that undergraduate institutions and programs may differ substantially in their capacity to develop students' scientific interests and talents. The principal purpose of the study described in this proposal is to examine a wide range of factors in the college environment that may affect students' interest in science, choice of a scientific career, and level of mathematical-quantitative competence. The study represents a major elaboration and expansion of a national study of the outcomes of general education programs funded recently by a $366,800 grant from the Exxon Educational Foundation to the Higher Education Research Institute at UCLA. The Exxon project involves a four-year longitudinal follow-up of the 1985 entering freshmen classes at a sample of 53 colleges and universities and a Fall 1989 survey of all faculty of these same institutions. The proposed expansion would add 100 institutions to the Exxon sample in order to achieve greater representativeness, to increase the reliability of results using the institution as the unit of analysis, and to generate larger samples of minorities, students in scientific subfields, highly able students, and other subpopulations of interest. Various multivariate techniques will be used in applying an "input-environment-output" model to the analyses of how various environmental variables affect the students' interest in science majors and careers, retention, attitudes toward science, aspirations for graduate study and mathematical- quantitative competence. In addition to institutional characteristics, the independent variables will include measures of curricular characteristics, peer climate, student involvement, faculty climate (teaching techniques, commitment to undergraduate education, involvement in research etc.), and student-faculty interaction.
