this post was submitted on 07 Apr 2025
362 points (97.6% liked)

Funny

8834 readers
2129 users here now

General rules:

Exceptions may be made at the discretion of the mods.

founded 2 years ago
MODERATORS
 
you are viewing a single comment's thread
view the rest of the comments
[–] unexposedhazard@discuss.tchncs.de 6 points 10 hours ago* (last edited 10 hours ago) (3 children)

Man i hate confirming these kinds of biases. I wish this plot werent true, but it makes so much sense.

[–] growsomethinggood@reddthat.com 20 points 10 hours ago (1 children)

The R^2 is trash, this is just a very squished cloud

[–] taiyang@lemmy.world 3 points 6 hours ago

It's Add Health, so it's not even the R2 that's the problem but the fact it's probably got a ton of covariates not included in the model. Lots of contextual variables not in that set drive me nuts, although I'm not even sure if 5th wave exists or has testosterone data lol (haven't used it in years)

[–] nasteva@jlai.lu 17 points 10 hours ago

Well, it's also important to note that the variability is high so the prediction for a single individual won't be of much use

[–] MBM@lemmings.world 3 points 9 hours ago

I wish this plot werent true

That's an easy wish! Last time I saw this plot people were saying that this kind of data doesn't even exist in the mentioned source (that and IQ is kind of bogus, but y'know)