The Drivers of Scientific Productivity

The Drivers of Scientific Productivity

January 22, 2020


727 E. Tyler St.
Tempe, AZ 85281


Biodesign Institute, Auditorium

Date and Time

January 30, 2020, 11:00 am (Length: 1 hour 0 minutes)

iCal Download

Aaron Clauset, PhD, Associate Professor, University of Colorado Boulder

Although meritocracy is a key organizing principle of academia, the social mechanisms that explain individual and institutional differences in scholarly outputs are poorly understood. What shapes a researcher's productivity over the course of their career? Which is a better predictor of early-career productivity and prominence, where a scientist was trained or where they currently work? Sixty years of research shows that researcher productivity tends to rise rapidly to an early peak then gradually decline. It is common knowledge that productive researchers tend to have doctoral training from top schools. In this talk, I'll show that both of these ideas are false. Using data on the publications, citations, education, and employment of nearly 2500 early-career Computer Science faculty, I'll show that the canonical career productivity pattern holds only for a minority of researchers. Instead, individual career productivity patterns exhibit enormous diversity. Second, through a matched-pairs analysis around initial faculty job placement, I'll show that the prestige of a researcher's working environment---not the prestige of their doctoral training---is a causal driver of their productivity and prominence. Hence, the scholarly output of early-career faculty is driven by where they work, not where they trained, and their current productivity and prominence cannot be separated from their place in the academic system. I'll close with a brief discussion of the environmental correlates of greater productivity, and the implications of these results for academic policy.


Hosted by Biodesign Center for Biocomputing, Security and Society.