My partner is new to mastodon. She said it seemed very male-dominated. I wanted to check this out, because I don't want to be one of those "I only follow dudes in my dudespace" people. In other words, this was me checking my own experience.
What I did:
1. Counted 200 (actually 208) posts in the 'federated' feed for c.im as they came through; tried to identify gender of all individuals posting, from usernames/pics (I know!) or chosen pronouns if that didn't seem glaringly obvious. This is in 'federated_feed'.
2. Did the same for 100 posts in my home feed (really just 99). For this and the previous item, individuals might be counted 2 or more times.
3. Tried to tally apparent gender for the accounts I follow (N=160).
'x' means a bot (e.g. The Guardian or Hourly Fops) or someone whose gender I wasn't confident I could guess at. Mostly it was bots.
Takeaways (right now):
1. There are a LOT of bots/newsfeeds/etc. in the big overall federated feed. My home feed has a much lower percentage of those.
2. Across the board there appears* to be a significant male bias in posts in the federated feed: 2.5 times more apparently-male than apparently-female posts (i.e., 150% bias? Am I doing this right?).
3. Both in the people I follow and the posts I saw (this specific day; IDK how representative that is), there is still a male bias, but it's much less extreme. Posts in my feed: 31% male-female bias; people I'm following: 27% bias.
There is probably a much easier and more representative way to do this, like by IDK using an API or scraping people's pronouns or something, but it was honestly faster for me to do it this way than learn the "simple" way.
* This is unlikely to be a truly representative sample of federated posts/accounts, but it's also probably not ridiculously far off; I suspect the trends would hold for larger/more representative samples.
#mastodon #gender #statistics #dataviz #ggplot #rstats #dataentry #whyAmILikeThis #GenderBias