Connections in Academia: Can Networking Opportunities Help Women?

Abstract

Female researchers have been show to publish fewer articles and in journals of lower impact across most fields in academia. A potential contributing factor may be that, on average, women have fewer connections and are less central in their collaboration networks, potentially due to the lack of networking opportunities. This paper analyses whether women have a different propensity to form new connections, exploiting a large scale natural experiment across all disciplines in Italian academia. I exploit exogenous variation in the networking opportunities available to professors, stemming from some pairs of researchers randomly being drawn to sit on the same committee, while other pairs are not. I find that researchers who sit on the same committee are more likely to collaborate on future projects in STEM fields, with new joint publications getting published as soon as 3 years after the initial shock. While I find no gender differences in the propensity to form new co-authorships, I observe that women in STEM change the gender composition of their networks. Female researchers substitute away from new female collaborators in favour of men. Point estimates suggest that female researchers, optimise by substituting a new female collaborator for 0.75 males. Through these new connections, women become more central within their coauthor networks. My results show that new collaborators do not change the number of papers published but they increase the quality of future publications. The evidence suggests that providing women with more networking opportunities may help in decreasing the gender productivity gap in academia.