The "study" is by Eric Kaufman, a DeSantis supporter who left his post at the University of London citing "cancel culture." He believes that trans identity is a social contagion. Here's what he believes his study proves.
He used two sources that lack questions on gender identity (the FIRE survey doesn't ask gender identity, I don't think) and then used student surveys from a private prep school, Brown University. He dismisses (basically) the General Social Survey has not reaching enough college age people. The other survey that would give an actual answer to this question is the CDC's Youth Behavior Survey, which was last administered in 2023 (and so cannot supply data since then). Similarly with the General Social Survey, which was administered a year ago. Across three of the five data sources - Fire, Andover, and Brown - there's about a halving of the number of students that ID as non-binary, according to him.
Eric concentrates on "elite" and "ivy league" institutions throughout - which are NOT representative of the broader US population. He does so because he has a background belief that trans identities are a luxury afforded to the elite classes, not something working people concern themselves with. This is, of course, a major fallacy and one that appears to go unexamined in his work. He also narrows the FIRE data which, overall, seems to show not much change in non-binary identification, solely to Ivy League institutions, which DOES show a moderate shift.
He also takes the same changes in percentages over time for one group as evidence of a major shift, while take the *exact same* amount of change in other groups (ones he finds acceptable!) as evidence of stability. If a 2-3% change in bisexuality is evidence, then it's also that in homosexuality.
He also doesn't contend with the fact that the survey participants *more than tripled* over the five year time period he's examining. More people surveyed will absolutely change the statistics by 2-3% points across the board, which is the change we see.
Similarly, one of the other surveys he relies on saw a decline in survey respondents from 200,000 to TWENTY FIVE THOUSAND. A decrease of nearly 90%.
He dismisses much bigger studies as "not representative" while relying on fucking BROWN as somehow representative of the college age population in the US. It's actually absurd. Just gonna leave this here.
"If we absolutely torture the data and reverse it around to say things it doesn't say, then LGBTQ people ARE mentally ill!"
Just straight up dismisses the idea that the rise in fascism has anything to do with people not willing to put their ID on a survey. He's assuming this because *student political attitudes* on the surveys remain constant. That's not a thing that dismisses a confounding factor?
Like, if I decline to give my identity on a survey that's going to a known conservative action group, that doesn't mean my politics have changed? My politics would show consistent year over year. Just my consideration for my own safety would change. There is a massive trans survey he could look at. This simply isn't supported by the data. He cherrypicks data from "elite" schools, does not do any controlling for representative samples, ignores larger studies for specious reasons, and draws a much bigger conclusion than any data suggests.
Indeed, the "changes" he's noting in the data can basically be accounted for as statistical noise considering the amount of people surveyed varied *drastically* between years, and he tossed out data that didn't agree with his conclusions. Why put out this study at all? Anti-trans policy requires a belief that being trans is a trend. If the numbers of people ID'ing as trans is shown to have major fluctuations, it can be used to argue that trans identity is fake and social contagion. That's the point of this whole project.