We consider three different species (S = 3) in the gender dimension: ”male”, ”female” and ”other”. We compute Shannon evenness by means of the Pielou diversity index. For calculating the Gender Diversity Index, we consider three different communities: keynotes (k), authors (a) and organisers (o). Our final GDI performs a weighted average among the Pielou index in each community with the following weights: 1/3 for keynotes, 1/3 for authors and 1/3 for organizers.
Geographic diversity index
As we have multiple species (S = number of countries present by community), we want to measure the richness together with the evenness, so we apply the weighted average of the Shannon Index community.
We compute the Shannon Index for each of the following communities: keynotes (k), authors (a) and organizers (o). We equally weight each community using the following weights: 1/3 for keynotes, 1/3 for authors and 1/3 for organizers.
We consider three different species (S = 3) in the business dimension: ”academia”, ”industry” and ”research centre”. We compute Shannon evenness by means of the Pielou diversity index. For calculating the Business Diversity Index, we consider three different communities: keynotes (k), authors (a) and organisers (o). Our final BDI performs a weighted average among the Pielou index in each community with the following weights: 1/3 for keynotes, 1/3 for authors and 1/3 for organizers.
Conference diversity index
This index is computed by the combination of gender, geographic and business indexes using the following formula: CDI = 1/3 *(GDI + GeoDI/2 + BDI)
More information in our paper:
Freire, A., Porcaro, L., & Gómez, E. (2020). Measuring diversity of artificial intelligence conferences. arXiv preprint arXiv:2001.07038.