-I freely admit that this art of looking at how people behave in the world and trying to describe what you see, is not a Science, because it relies on your brain's magical (as the term of art for "capabilities we don't know how to program") pattern-matching abilities that other people might doubt, whereas if you have a pre-registered survey of fixed questions, and the target demographic comes back and says 3.4, other people can't seriously doubt that they did, in fact, say 3.4.
-
-Nevertheless, looking at how people behave in the world and trying to describe what you see is a form of _empiricism_, even if it's less third-party-legible than Science. In trying to upgrade our naïve empiricism to a Science, we should hope to design careful surveys to give us quantitative measurements of the qualitative patterns we see "in the wild", but getting this right is a surprisingly tricky endeavor. We're not obligated to throw away all our qualitative observations, in favor of single (!) survey question, just because the latter is quantifiable.
-
-You wrote about this in ["My IRB Nightmare"](https://slatestarcodex.com/2017/08/29/my-irb-nightmare/), expressing skepticism about a screening test for bipolar disorder:
-
-> You ask patients a bunch of things like "Do you ever feel really happy, then really sad?". If they say 'yes' to enough of these questions, you start to worry.
+The thing is, I don't see my theory as making particularly strong advance predictions one way or the other on how cis women or gay men will respond to the "How sexually arousing would you find it to imagine _being_ him/her?" questions asked on the survey.