We scrolled through the feed every couple of weeks to check what was being served up.

This is a critical flaw in the studies methodology. 'loiter time' is a metric used by algorithms to serve up new content, if the researchers where checking each post for signs of misogyny, then they were probably skipping by totally innocucus stuff whilst paying attention to misogyny. This (being the only feedback given) will have shown the researchers what they wanted to see.

Three months later, The Office, Star Wars, and now The Boys memes continue to punctuate the feed, now interspersed with highly sexist and misogynistic images that have have appeared in the feed without any input from the user.

It'd be good to know the actual ratios, given this was a guardian study there's no reason to withhold data, nor a secondary source I can go and find the data. It's possible that Facebook is simply serving up the entire spectrum of posts proportionately to their activity on the internet, or even favouring anti-sexist posts that are just not noticed/mentioned by the guardian.

Does anyone genuinely believe banning this sorta stuff is going to "end violence against women and children in one generation"?

The John Doe profiles were set up in April as generic 24-year-old males

Does the fact that they only used dummy male accounts and didn't bother to find out what women are subjected to represent the authors' own misogyny? Perhaps The Guardian should look at cleaning it's own house up before finding dirt on others'.

What's this got to do with the UK exactly?