Back

The W-index: a novel tool to evaluate gender equity in STEM research

Holland, E. P.; LaMontagne, J. M.; Chaudry, V. B.; Charman-Anderson, S.; Gillson, L.; Helgason, T.; Lipscombe, A.; Robinson, E. J. H.; James, A.

2025-11-01 scientific communication and education
10.1101/2025.10.29.685415 bioRxiv
Show abstract

Gender inequities in scientific research persist, from citation indices to collaboration networks and funding success. Current estimates for achieving equal representation remain in timescales of decades to centuries. To help individuals and groups improve their gender representation in a shorter timeframe, we present the W-index, a simple metric to evaluate gender ratios, and specifically the representation of women, in research collaborations and other aspects of scientific life. Here it is applied to three case studies: internal collaborations derived from six years of research outputs across a university, co-author groups from 60 years of publications across a journal collection, and supervisor-student relationships over 60 years. Despite contrasting sources, these datasets show common features: W-index increases as the proportion of women increases W-index also increases with age or career stage, and women have a consistently higher W-index compared to men. We finish with reflections and recommendations for individuals, organisations, and funding bodies to action positive and timely change towards gender equity.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.