Research Quality and Gender Gap in Research Assessment

The literature on the gender gap in science reveals differences in wages, productivity, access to funding and impact on the scientific community that disadvantage women. This paper contributes to work on the gender gap in science by investigating issues such as the presence of differences in research quality between genders, the effect of family responsibilities on research quality, differences in collaborations and international co-authorships, the effect of evaluation methodology, i.e. whether bibliometric evaluation disadvantages women, and the presence of discrimination defined by referees’ gender. We use the data from the National Research Assessment (VQR 2004-2010) conducted by the Italian Agency for the Evaluation of Universities and Research Institutes. These rich data allow us to control for individual variables, research output characteristics and university and scientific sector fixed effects. We find that gender differences in research quality are reduced if we control for researchers' observable characteristics, evaluation method, and referees. In particular, we find that maternity and the intensity of research collaborations and international co-authorships play no role in explaining research quality differences. Further analysis of a random sample of papers evaluated using bibliometric indicators and peer review reveals that bibliometric evaluation does not penalize women with respect to men.