White dots hugging the 0 line in Figure 2 in http://people.virginia.edu/~jra3ee/LaiEtAl2016.pdf …
-
-
Replying to @Selerax
The claim in the studies I cited wasn't that IAT never changes, but that it does not reflect real behavior.
1 reply 0 retweets 0 likes -
Replying to @slatestarcodex
Selerax Retweeted Tom Stafford
OK, look at this:https://twitter.com/tomstafford/status/598938727117914113 …
Selerax added,
1 reply 0 retweets 0 likes -
Replying to @Selerax @slatestarcodex
The individual test-retests are all over the place, but the mean is essentially identical
2 replies 0 retweets 0 likes -
Replying to @Selerax @slatestarcodex
One interpretation is that IAT is a very noisy measurement of a very real, reliable effect
1 reply 0 retweets 0 likes -
Replying to @Selerax @slatestarcodex
But if you use the noisy measurement, obviously you would't be able to predict much.
@tomstafford1 reply 0 retweets 0 likes -
Replying to @Selerax
Shouldn't you be able to do that with high enough n?
1 reply 0 retweets 0 likes -
Replying to @slatestarcodex
If you have multiple measurement for each individual, maybe. Or if you average by (large) quantiles.
1 reply 0 retweets 0 likes -
Replying to @Selerax
Don't these big meta-analyses count as using large quantities?
1 reply 0 retweets 0 likes -
Replying to @slatestarcodex
I may be off, but I'd think they would be swamped by massive noise in individual measurements
1 reply 0 retweets 0 likes
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.