Great summary: "In general, stereotype accuracy was not easy to predict, even using 24 predictors (k-fold cross-validated R2 = 4%)."https://twitter.com/KirkegaardEmil/status/873613008894865410 …
-
I think we somewhat underestimated the predictive validity due to using cross-validation instead of bootstrapping, but still.
It's surprising that stereotype accuracy is so hard to predict. Perhaps it's not a stable trait at all (no study reports a retest cor).
