Gender differences in predicting faculty publication output in the natural sciences.

Research has shown that the academic sector offers a "chilly climate" for women faculty, especially for the relatively few women scholars in the natural sciences. Since faculty members serve as role models and creators/disseminators of scientific knowledge, the status of faculty women scientists is an important issue. The theoretical framework for this study offers an interdisciplinary approach to motivation incorporating sociological (including socio-demographic and career), psychological, and environmental variables that can be used to explore gender differences in faculty productivity. The purpose of this study is to investigate similarities and differences with regard to sociological, psychological, environmental, and behavioral variables that predict research productivity among female and male full-time faculty in the natural sciences at three different time periods (1988, 1975, and 1969). The data originate from three national data surveys. Several questions are repeated across survey years, thereby enabling some comparisons over time. The data analyses proceeded through two stages: (1) descriptive techniques (frequencies, plots, ANOVAS with post hoc Scheffe tests, and chi-squares), and (2) separate multiple regressions by gender for predicting publication rate. It appears women natural scientists have made some progress in their academic status. Furthermore, the proposed framework for understanding faculty publication differences for each gender is a useful one. Although the gender groups differed in the sociological, psychological, and environmental variables that predict publication performance, the variables explained a high percent of the within group variance (ranging from 41-62) in two-year publication rates (more for females). In all three time points, the variables predicting performance for the female scientists were fewer in number than for the male scientists. These results show that over the past two decades, women at these surveyed institutions have less within group variance in research effort, background and career characteristics, determinedness to do research, support from funding sources and colleagues, research interest, and perceived success. The strongest gender difference predicting publication rate, using the most current data (1988), was perceived research self-competence. If women believe they are competent, they publish; if they do not have this valuation of themselves, they do not.PhDEducationUniversity of Michigan, Horace H. Rackham School of Graduate Studieshttp://deepblue.lib.umich.edu/bitstream/2027.42/105619/1/9135711.pdfDescription of 9135711.pdf : Restricted to UM users only