> By the common definition of ever having erotic arousal to the thought or image of oneself as a woman, 93% of the [cis women] respondents would be classified as autogynephilic. Using a more rigorous definition of “frequent” arousal to multiple items, 28% would be classified as autogynephilic.
You are making my point far better than I could, but please… please stop. HN isn't for this.
> Blanchard proposed that autogynephilia is a natal male's paraphilic sexual arousal in response to the thought or fantasy of being a woman. Furthermore, based on evidence collected from natal males with gender dysphoria, Blanchard argued that autogynephilia is the fundamental motivation among nonhomosexual males (i.e., those not exclusively attracted to men) who pursue sex reassignment surgery or live as transgender women. These ideas have been challenged by several writers who have asserted, or offered evidence, that autogynephilia is common among women.
> However, their evidence was weakened by problematic measures and limited comparison groups.
> We compared four samples of autogynephilic natal males (N = 1549), four samples of non-autogynephilic natal males (N = 1339), and two samples of natal females (N = 500), using Blanchard's original measure: the Core Autogynephilia Scale.
> The autogynephilic samples had much higher mean scores compared with non-autogynephilic natal males and natal females, who were similar.
> Our findings refute the contention that autogynephilia is common among natal females.
This fails to address the limitations of Ray Blanchard's questionnaire, which are described in the (open-access) study I linked, and invalidate this interpretation of the measurements. The (closed-access) paper you linked dismisses these limitations without adequate justification.
Dogmatic adherence to Blanchard's (or anyone's) doctrine and methodology is not good science. Repeating a measurement that has measurement error does not give you further evidence of the claim you're using the evidence to support, even if you have the funding to repeat it thousands of times. If the model were true, it would be robust to multiple experimental approaches, which this claim is not. Pseudoscience does not cease to be pseudoscience if people repeat it a lot.
Actually, they cite someone else's critique which, as I said, is inadequate. For instance, it claims that making a question more specific (and therefore less likely to apply) for the cis-adapted questionnaire makes the results incomparable. Given the one-tailed test for the main result in Moser (2009) is not in the direction that would be invalidated by an underreport in cis women, that's just a totally nonsense criticism.
If we assume it was reported correctly, then the research is not pseudoscience. The claims are pseudoscience: they are not supported by the evidence produced by the research. Like most such quackery, the category boundaries are ever-nebulous, turning to smoke whenever anyone "misinterprets" the "theory", and turning to steel whenever an "expert" makes an "incontrovertible" proclamation.
You might have too strong of an opinion on this topic to be able to see what I'm talking about, so consider reading Karl Popper's writings on falsifiability in the context of Freudian psychoanalysis. Perhaps also look into the "research" on Simon Baron-Cohen's "Extreme Male Brain" theory of autism published after the 2019 foetal steroids study (coauthored by Simon Baron-Cohen!) that pulled the rug out from under its remaining foundational assumption: you'll see authors citing and "critiquing" the mainstream view (i.e., that the whole thing was an honest mistake resulting from incorrect questionnaires designed to validate, not attempted-falsify, an assumption based on a now-discredited stereotype based on a patriarchal notion of gender roles, and we know better now) and yet carrying on with their "research" that presumes Extreme Male Brain regardless, because their academic careers and their egos are too invested in it. Some of their studies have very large sample sizes, too! and yet they don't show what their authors claim they show: they can't possibly, because those claims aren't true.
Note that Simon Baron-Cohen is among the people reluctant to give this up: you'll see that all throughout the 2019 paper, it establishes correlation and then immediately asserts an arrow of causation, and speculates about various other theories without adequate justification. But while Simon Baron-Cohen is bad at science, and has his pet theories, I wouldn't say he goes as far as pseudoscience: I haven't known him to write disingenuous "critiques" as in the paper you cited, and he does seem to acknowledge when his concrete theories are disproven – for example, his "mind blindness" theory (a component of his "extreme male brain" theory), which he's changed his mind about in the wake of Damian Milton's work. (Work on mind-blindness after double-empathy would be a better example of the "people still publishing discredited theories which never had actual experimental support" phenomenon, except that the shape of the methodological flaws in those works are much closer to Ray Blanchard's, so – depending on what exactly is stopping you from seeing them here – you might not be able to see them in those either.)
Given that I've brought him up in this context, I feel the need to clarify: from what I can tell, Simon Baron-Cohen does care, a lot, about the welfare of his research subjects. He does a lot of work outside his immediate scientific contributions. I do not think he should be condemned for what are understandable mistakes, albeit mistakes he probably wouldn't have made had he adhered to "nothing about us without us" in his early research: the whole point of science is to err decreasingly, and for all his faults (most of which would hardly matter, if he were a less influential academic), he has done that.
