<p>
[quote]
You don't like generalizations? We're talking about groups with hundreds of thousands an even millions of members. You want to go through the discussion one person at a time? You can't even discuss this topic without generalizing. Affirmitive Action ITSELF is based on generalized notions of GROUPS being disadvantaged, underrepresented, etc. There is nothing whatever wrong with saying "this group tends to be x or y." Generalizing becomes a problem only when somebody says, "Person A belongs to group B, so he (as an individual) MUST have the the traits that his GROUP in general has."
[/quote]
</p>
<p>You misunderstand me.</p>
<p>It's obvious that generalizations have to be made. A huge amount of generalizations are made about a great deal of topics. However, it's obvious that there's a bad side to it. There are generalizations, and there are over-generalizations. Overgeneralizations tend to be bad generalizations.</p>
<p>You are giving bad generalizations.</p>
<p>Examples?</p>
<p>1) URMs need to stop complaining
2) URMs at Harvard would be at schools like Tufts without AA
3) URMs don't succeed to thier max at top schools (I base this off of your statement that URMs would have higher GPAs, be on Dean's List, etc.)</p>
<p>For Number #2, I'm pretty sure they don't release specific stats or percentages of URMs accepted to schools, so it would be difficult to tell how many URMs are 'underqualified'. Furthermore, even with AA, admissions people very rarely admit students that they don't think will succeed at their school, because in the end it will reflect badly on them.</p>
<p>For #3, there's also no way to tell that a URM/group of URMS would do better at less selective universities because a wide variety of things like student response to an environment, or learning style come in to play. Furthermore, the level of competition may affect the motivation.</p>
<p>
[quote]
Murasaki, your last paragraph points out the differences in difficulty of GPA at different colleges. It might be the case very occasionally that the same effort that gets you a 3.6 at Yale gets you a 3.2 at Fordham, but what I'm talking about is getting a higher GPA compared to others at the SAME college. For example, a URM gets a 3.6 at Yale. You're saying that if that person went to Fordham, perhaps he would have only gotten a 3.2, because Yale has more grade inflation. I'm saying regardless of what the URM gets at Yale, because he's been admitted with USUALLY significantly lower credentials than his "majority" and ORM classmates, he will generally have a lower GPA than his classmates (i.e. lower class rank) and therefore generally be a less-attractive grad school applicant than his Yale classmates would be.
[/quote]
</p>
<p>1) This assumes that lower credentials in admission leads to lower performance in college.
2) If you want to talk about grad school acceptances, then certain GPAs at higher tier schools, even if they are less competitive than at lower tier schools, might work better in the end. In other words, even though a 3.6 at Yale might be less competitive than a 3.2 at Fordham, the mere reputation of the college could work in favor of the applicant.
3) Your generalization on URMs doing better at lower tier schools also doesn't take into account the fact that there is often a disparity between selectivity and competitivity-in-class of a college. College A may be more prestigious, but dropping down to College B because of a lack of AA does not necessarily mean that College B is less difficult to rank at the top of the class in.</p>
<p>But back on my #2 answer, my basic point here is that it's also a generalization to assume that a) the vast majority of URMS are 'underqualified' and b) kicking AA would drop the majority of URMs down a tier and c) that this drop would guarantee an increase in the said URMs' performances.</p>
<p>When put in a vacuum, these generalizions seem perfectly reasonable. But since a lot of your points in your scenario rely on the occurance of a previous one (like AA dropping down URMs down to certain level schools, then those URMs doing better, etc.), the end is an over-generalization of the end result, and that's bad.</p>
<p>I think that I agree with the gist of your idea, but there are just too many unknowns, and those unknowns make for bad generalizations.</p>