Always Highly Productive, Always Low Productive? A Longitudinal and Class-Based Approach to Research Output of Full Professors

Abstract

Lifetime biographical and publication histories of 2,326 full professors were examined. A combination of administrative, biographical, and bibliometric data was used. Retrospectively constructed productivity, promotion age and speed classes were examined. About 50% of current top productive professors have been top productive throughout their academic careers, over 30–40 years. Top-to-bottom and bottom-to-top transitions in productivity classes over academic careers are very rare. We used prestige-normalized productivity in which more weight is given to articles in high-impact than in low-impact journals, recognizing the highly stratified nature of academic science. The combination of biographical and demographic data with raw Scopus publication data from the past 50 years (N = 935,167 articles) made it possible to assign all full professors retrospectively to different productivity, promotion age, and promotion speed classes. In logistic regression models, there were two powerful predictors of belonging to the Top productivity class for full professors: being highly productive as associate professor and as assistant professor (increasing the odds by 180% and 360%). Neither gender nor age (biological or academic) emerged as statistically significant. Our findings have important implications for hiring policies as scientists stay in Polish academia usually for several decades.