Faculty research performance over time and its relationship to sources of grant support.

Abstract

This study examines changes in the correlates of faculty publication performance over time, focusing in particular on the role of grants. It compares the interrelationships between grants received in the past year (institutional, federal, industrial, and other) and publication output (two-year publication rate and lifetime articles). Data are from the 1969, 1975, 1980, and 1988 national surveys of U.S. faculty. The samples consist of faculty in the humanities (English, history), natural sciences (biology, chemistry, mathematics), and social sciences (political science, psychology, sociology). Faculty are represented from three institutional settings: research, doctoral-granting, and larger comprehensive institutions. The study develops weights from an outside data source to minimize data distortions due to sampling discrepancies and response bias. Weighting stratification cells account for changes by discipline, institutional setting, and institutional control. Multiple Classification Analysis is used to measure the strength of correlates while controlling for the effects of other predictors. Also part correlations are computed to isolate the unique variance explained by grants after accounting for the effects of age, rank, and institutional setting. Federal grants and institutional setting are the strongest correlates of productivity. All grants explained about 10% of variance after accounting for institutional setting, age, and rank; federal grants alone accounted for 5% to 7% of the marginal variance. The importance of institutional setting (partial beta coefficients) strengthened over time as a result of slower productivity gains at comprehensive institutions. Meanwhile productivity differences narrowed between research and doctoral university faculty, although research universities retained their initial (1969) productivity advantage. Age and academic rank also were strong predictors of two-year publication rate (betas $<$.20). Productivity of younger faculty (under age 36) rose markedly at each successive time period. Gender productivity differences declined over time as more women worked at research universities and received grant support. Still women remain a minority in academe, especially in the natural sciences. This study contributes one of the few syntheses of national faculty survey data over time. The findings, spanning 20 years, explore the strength and stability of productivity correlates within a multi-disciplinary and cross-institutional context.