News for hackers. Stuff that matters.
Re:Recruiters and investors think talent counts for more tha... by camgunz on Friday December 11, 2015 @11:31PM (#10721795)
You do it so you can compare what they say they would do to what they actually do. Think Bradley effect. You can survey people and ask them if they'd be less likely to vote for a minority candidate, and then you can look at the voting results. Afterwards, when you do your regression analysis, you can say that statistically speaking, the only explanation for the gap (between people who say race doesn't affect their vote and the votes a minority candidate actually received) is that voters factored in race.

But yeah I mean, you can't take what people self-report as evidence. There are tons of problems with that, social desirability bias, etc. etc.

There are no comments on this comment yet.