Thanks!

The quote (from Slate)above works out to around 0.01%. But it's wrong--the supporting data is given, and it's nothing like 0.01%. Maybe the authors just messed up their math?

It's a little surprising how far apart the two reports are. If you chase the Slate report through CNBC (https://www.nbcnews.com/feature/nbc-out/media-s-detransition-narrative-fueling-misconceptions-trans-advocates-say-n1102686), you end up at a large document detailing all sorts of aspects of the trans experience (https://transequality.org/sites/default/files/docs/usts/USTS-Full-Report-Dec17.pdf). They report 8% detransition of any sort (11% trans women, 4% trans men).

But the LGBT health paper is the exact same data. There was only one study! And their number from the same data is, as you said, 13% (16% trans women, 6% trans men). Why the difference?

Well, because the first one averaged in people who said they hadn't transitioned at all. Of course people who haven't transitioned don't detransition.

Anyway, depending on the case that was covered in the documentary--you didn't mention what it was, so I don't know--it seems like it could either be not too far from the truth (13%), or pretty far from representative (1%--the estimated number whose detransition was for internal reasons, given the 10% rate of reporting that; most of the reasons are lack of support).

But there's an enormous caveat to using this data at all: every person in the study, by design, identifies as transgender. At least enough to be included in the study. As the paper authors say, "Because the USTS only surveyed currently TGD-identified people, our study does not offer insights into reasons for detransition in previously TGD-identified people who currently identify as cisgender."

That's pretty useless if we're trying to look at overall rates, isn't it?

If you go into a coffee bar and you ask how many people tried coffee for a while and then decided they didn't like it, you're not going to get representative results.

Do we really actually genuinely not have a better handle on the numbers than this?

In 2018--three years after the study--there's a paper (https://psycnet.apa.org/record/2018-33024-004) that advocates paying more attention, medically, to supporting de-transitioning, while noting that "In the realm of prepubescent children, we are acting without strong data to tell us how many will continue to identify as transgender adolescents and adults.".

If you go all-out the other way--include only people who have detransitioned and identify as such (i.e. not trans any more), their reasons are almost wholly different (https://www.tandfonline.com/doi/full/10.1080/00918369.2021.1919479) than the ones who ended up selected in the trans survey: internal factors greatly outweigh external ones.

Of course, this doesn't give us a complete picture either; it just shows that when you don't try to get a complete picture, the local part of the picture might look very different than the overall view.

So--honestly, do we really not know the answer here?

And if we actually don't know the answer, can we confidently say that the documentary was wrong to give an example? (There are lots of ways to do things wrong, foster hostility and misunderstanding, and so on; but when it comes to the mere inclusion of an example, that it's wildly misleading not obvious to me.)

(Addendum: please don't misread me as saying that I think gender-affirming therapy is necessarily harmful to adolescents. We have pretty decent evidence that it's helpful on average, e.g. https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2789423. I'm only questioning whether we're adequately counting and taking into account the experiences of people for whom this has not been working out well.)