seems to me you're not really talking about inferences at all, are you? you want an excuse to say other people's concerns are silly. cool!
-
The predicted difference from a "basic stats" perspective is 0.15-0.3 sds. See a 1 sd difference in reality. Kurt's tweet was condescending
Yes, it can be helpful to see illustration of distributions: asians and blacks barely even overlappic.twitter.com/zylJ371QWk
-
-
May I know the source of this graph?
-
oh sorry missed this query before. it's my own plot (unpublished blog post), but derived with data from NLSF http://nlsf.princeton.edu/
-
No worries, and thanks!
-
I didn't know about this survey, but I have a feeling I'm going to spend some time looking at those data...
-
yeah it's interesting (been widely used for papers & books on AA and related topics)
-
at least some of the data are self report and not very granular (i wish it had granular transcript HSGPA), but still worth a gander
-
also they don't report which institution is which, but one could probably triangulate which is which fairly accurately :P
-
speaking of which, some familiar names jumped out at me when I was just reviewing the publications list (cc
@KirkegaardEmil and@jfuerst0)pic.twitter.com/YyHH4sfwDq
- 1 more reply
New conversation -
-
-
The bigger the gap is, the more it supports
@KiraboJackson point. I just did a quick calculation: if group X is 1 sigma > group Y, 1/ -
these are the gaps in scores of the actual students (offered & accepted)
-
Sorry did I misunderstand? Are those the actual distributions of students who have been admitted? 1/
-
yes, reported scores of enrolled students at start of freshman year (or something close to that.... been awhile since I looked at this data)
-
My apologies. Although
@KiraboJackson methodological point is right, it looks like the broader point is wrong. Adequate evidence of 1/ -
bias appears to exist when looking at marginal students, which is evidence of bias. 2/2
-
so you're saying colleges don't go to SCOTUS to defend racial preferences as a theoretical matter, but because they actually use them? wow!
-
I find statistics more interesting than identity politics.
- 4 more replies
New conversation -
-
-
i wonder why
@KiraboJackson didn't post a graphic like this. probably really frustrated about statistical reasoning, not muddying the issue -
Tweet unavailable
-
i posted about that too - i agree that's probably #1 in the absurdly false claims about affirmative action
End of conversation
New conversation -
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.