It's a little sensational but I feel like she's pretty up front about the limitations. The problem with representativeness is admitted, not really engaged with (I'd expect a more thorough study to compare with older results with better understood biases and extrapolate from there), but it's not hidden either.

I think the main problem is actually that Aella herself is not a faceless researcher but a somewhat controversial online personality who used her own brand to find participants. So for example anyone who blocked her on twitter probably didn't have a chance at taking this, even if they are also likely to be closer to kink communities than the average twitter user who wouldn't even know of her existence.