> college admissions shouldn't be about past performance but about future potential These two variables are highly correlated. Do you think adding a fudge fact

> college admissions shouldn't be about past performance but about future potential These two variables are highly correlated. Do you think adding a fudge factor for students from disadvantaged groups will give you a more accurate predictor of future potential than just looking at past performance? Is there any data to support that claim? > Why should employers take into account how someone got into Harvard, instead of judging them solely on their performance at Harvard? Judging them solely on their performance at Harvard should still lead you to the conclusion that AA students are worse performers on average. That's true even in the hypothetical world where AA students don't come from a disadvantaged group in the first place. If blue weasels and red weasels have exactly the same distribution of skill, and you add affirmative action in favor of blue weasels, then blue weasel students post-admission will underperform their red weasel peers on average, for reasons that seem mathematically obvious.