Faculty productivity in supervising doctoral students' dissertations at Cornell University

Most research on faculty productivity focuses on the research ability of faculty members. This paper provides a new means of looking at faculty productivity by analyzing a second responsibility of faculty members: producing new Ph.D.'s. The authors first utilize a Lorenz curve analysis to establish that graduate student supervision is most equally distributed among faculty members in the physical sciences and least equally distributed among social science faculty. The second part of the paper uses a negative binomial regression model to investigate what faculty attributes affect how many graduate students a faculty member will supervise in a six-year period. The results suggest that, on average, a faculty member's prestige and her length of time at the institution are significant factors in predicting productivity.