This project collects tweets and compares the prevelence of discussion about trans women vs that of trans men.
Content warning: the tweets in raw_results.csv
are unmoderated and many are transphobic.
The Twitter API was used to query recent Tweets matching the following keywords:
- trans man/men
- transgender man/men
- trans woman/women
- transgender woman/women
A Tweet was classified (case-insensitively) as...
- ...M if it contained any of the first two queries, as well as "transman"* or "transmen"*.
- ...F if it contained any of the final two queries, as well as "transwoman"* or "transwomen"*.
- ...Both if it could be classified as both M and F.
- ...None if it could be classified as neither M or F.
This led to the following summary of results from the 34478 Tweets pulled:
Gender | Mentions in Tweets | Average Likes per Tweet | Average Retweets per Tweet |
---|---|---|---|
Men | 19% | 20 | 4.4 |
Women | 70% | 11 | 1.5 |
The raw data raw_results.csv
is created with main-scraper.py
and the full results overview.txt
are created with main-analyser.py
.
*These terms should not be used. 'Trans' is an adjective which can modify a gender - using 'trans(wo)man' and not 'trans (wo)man' can imply that trans (wo)men are not (wo)men.