What kind of accuracy would you expect if you sort just by race?
-
-
-
That depends on which race and which data distribution. In the simple case of half White half Black with 1 d gap, one would get about r = .45. This value is related to the 1 d gap, which converts to r = .45 for equal groups. https://www.polyu.edu.hk/mm/effectsizefaqs/calculator/calculator.html …pic.twitter.com/t9vE75kMp5
-
Thanks for the detailed response! So in your 2 race 50/50 proportion, predicting race from Facebook profiles would get the algorithm in the ball park for rating intelligence?
-
Yes. But reality is not just 2 race, and many differences are smaller than 1 sd, so relying on just racial stereotypes cannot be as accurate as utilizing individuating information too.
End of conversation
New conversation -
-
-
modern day phrenology
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.