We find that researchers tend to co-publish with same-gendered colleagues, and that this tendency is presently increasing. We found no evidence that senior academics drive this pattern, or that the pattern is stronger in fields where women are in the minority. Interestingly, journals with a high impact factor for their discipline tended to have comparatively many mixed-gender teams.
We recorded the gender of 36 million authors from 100 countries publishing in 6000 journals from most STEMM, and made a [web app](https://lukeholman.github.io/genderGap/) facilitating exploration of the data. Despite recent progress, the gender gap appears likely to persist for generations.
Using text mining to help redress gender inequality in the STEM workforce.