Evaluating Ph.D. sociology programs: Theoretical, methodological and policy implications

Abstract

Reputational surveys frequently are used in evaluating doctoral programs. A sample of faculty-rated graduate programs and the resulting data are used to rank the programs in the academic discipline being considered. As an alternative, we present a new and auspicious assessment methodology by analyzing sixty doctoral programs in sociology for 1975/80 on six measures of research productivity and graduate education. We calculate two proximity matrices and reduce each with ALSCAL. Two solutions are generated: (1) per capita and (2) gross productivity of departments. These analyses show that conventional ranks contain a number of deficiencies and more comprehensive measures of departmental performance add important dimensions. We empirically document a tendency for conflict between faculty productivity and graduate education. Finally, our analyses suggest that omission of several assessment variables may have penalized departments emphasizing qualitative research. Recommendations for improving the next national assessment are advanced.