That French study was an online questionnaire sent to a number of French autism groups asking them to send the link to the questionnaire to their members. 225 women completed the questionnaire. So the research participants were self selecting.
In their report the researchers make passing reference to a survey of 300 female college students in the US, 61% of whom reported experience of sexual aggression but they ignore that survey as a comparator. They also mention a US phone survey of 12000 participants (both m and f) which found 44% of female participants had experience of rape and sexual violence. They disregard that survey as a comparator too.
Instead they use a comparator with a 30% rate. The comparator seems to be a WHO study estimating global rates of sexual violence against women. But in that WHO study “lifetime” refers to events since the age of 15. The French study adopts a different methodology and does not exclude events before the age of 15. (The study states that 112 of the 225 participants’ first experience was before the age of 15.)
I’m not in any way disputing that autistic women are likely to be vulnerable to assault, by the way. But I don’t think that a study covering a completely different geographic area, different sample size, focused on a different age range, and using a different methodology is the most relevant comparator for the French online questionnaire study. The study authors seem to have looked for the lowest estimated rate of SVAW (achieved by using a study which excludes girls’ experiences) and used that as their comparator as it makes the 9 out of 10 rate in their survey look even more shocking.
It would still have been worrying if they’d used the (probably more directly comparable, based on methodology) rate of 61% as their comparator. But less headline grabbing perhaps.