The way to do this, however, is exactly not how you've approached this, which is to attack the character and associations of people, do a good bit of mindreading (albeit with some pretty suggestive circumstantial evidence), and point out zero detailed flaws in their work.

There's some value in asking people to do as you say and not as you do. But at least you ought to very clearly tell them: DON'T DO IT THE WAY I DID.

There is immense danger in trying to bolster a scientifically-valid perspective through character assassination and shunning because character assassination and shunning are completely unrelated to the truth of the matter.

This causes two disastrous consequences: (1) people who are naturally contrarian assume that you're adopting methods that are truth-independent because the truth is not on your side: it is evidence to them that you're actually a manipulative liar in service to some possibly hidden agenda (because otherwise why not just appeal to evidence?!?!); and (2) if you make a mistake, you never figure it out because anyone who would is terrified of being shunned and having their character assassinated (your psychological safety is zero around the entire area of inquiry).

You can see both of these unfortunate trends happening with climate science and Covid vaccines: rampant disbelief because too much of the rhetoric feels like propaganda not scientific evidence, and important questions addressed too slowly if at all because if the answer comes out "wrong" it's

Just keep debunking the arguments. Find better and better ways to debunk the arguments (more compact, more compelling). Develop a collection of resources you can link to so you don't have to re-invent the argument each time (like talk.origins did for all the pro-creationist arguments). If there's some new argument whose implication rubs you the wrong way, check it out and see if there's anything to it--occasionally there might actually be something to it because unlike conservation of energy, conservation of distribution of traits across races is not a fundamental physical law.

The presentation of a bad, discredited argument is a great teaching moment for a general audience. "Hey look! Here's someone who says that race causes low or high IQ. How do we know? We got a million people of different races and raised each one in identical conditions...hahaha, no of course not! But so-and-so is pretending that we did, ignoring all the other differences like where and how people live, how much money their parents and grandparents had, all sorts of stuff like that. That's really foolish! We know lots other stuff matters! What happens if we measure all the different factors we know about that might be important as well and ask: okay, which thing predicts IQ, if we consider it all together? Is it it actually race, or is it something else? And (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7506587/) we've done that in some cases, and we see that it's either entirely or almost entirely something else."