Gender differences in productivity and collaboration networks of top-ranked academics

This week we were pleased to be joined by Ana Jaramillo and Mariana Macedo to discuss this topic of gender differences in academia. Both are PhD candidates at the University of Exeter in Computer Sciences.

First we heard from Mariana about `Differences in the spatial landscape of urban mobility: gender and socio-economic perspectives’. This study is based on the definition of human mobility as the displacement between locations in urban areas e.g: how far someone travels for work. This has historically been very different with women either staying at home or working closer to home for reasons such as childcare and safety. This work only considers the social constructs affecting urban mobility, and not possible biological reasons. The aim of the work is to address a gap in the current literature and shows that urban mobility can be impacted by a variety of factors including cultural constructs, modes of transport, city development and spatial opportunities.

The work done by Mariana considers mobility in three different locations (Medellin, Bogotá and São Paulo), and quantifies the mobility diversity as a quantity H. It is no surprise that men consistently had a higher level of mobility in these regions. Interestingly, the upper class had lower urban mobility and a larger gender difference compared to middle class where the gender gap was smaller. Overall, socio-economic factors gives a higher diversity in mobility then gender difference, but gender difference can become ten times higher when we also consider socio economic factors.

We then heard from Ana on the topic of `Gender difference in productivity and collaboration networks of top-rated academics’. It is well known that women are underrepresented in academia, particularly in STEM subjects. The good news is that representation is growing, but very slowly. From this work, it is hoped that it may be possible to diagnose the problem of this slow growth. Ana’s work focuses on the productivity and collaborations of the top 10% most productive computer scientists as quantified by their number of publications.

From the number of publications over a person’s career, it is interesting that women are seen to have much shorter careers at 35 years vs 47 years for men on average. Once productivity is understood, collaboration and networks can be studied by looking at co-authorship on papers. It was interesting to hear that women work with more women than men do, while men work with men more than they do with women. Women also have networks which are more clustered meaning that they tend to work more in research groups. The conclusion of this is that men work with women less than expected. The good news is that this is slowly changing as representation improves and working patterns become more collaborative.