> "Diversity" programs are by nature unfair- they provide explicit preferential treatments to specific groups in order to counteract presumed other factors that make those groups less likely to be present in the first place.
Often, but not necessarily. Since this is a forum for people who like to be technically correct, I'll throw my two cents in here.
To give an example, let's assume that there's a split between group A and group B between high cost and low cost universities. Group A are 90% of the high cost students and group B are 90% of the low cost students. Google send representatives to mostly the high cost universities at the moment, because they're assumed to have the best students. This is resulting in a remarkably large proportion of As at Google. To counteract this, you could go and explicitly target Bs, which would fit with your description. Or, you could lower your threshold and start sending reps to lower cost unis as well. This increases diversity yet at no point is specifically preferring Bs.
I picked this example of high cost/low cost unis purely because it's easy, not to make any statement about whether this kind of thing happens or is important for future performance. this is also not a hugely important point, but it was floating about in my head.
