<p>But I think the question that must be asked (and can't be answered) is this: would elimination of an early round cause a different set of people to be admitted to the school?</p>
<p>I mean, if you eliminate an early round, how does that solve the problem that Harvard claims to have -- less economic and ethnic diversity in the early pool? You still have the same total pool of people, and you still have to choose the same total number of them. I would predict that the same set of people (well, a set of people with statistically indistinguishable characteristics, both academic and economic/ethnic) will be admitted to Harvard next year. They'll just all be admitted at the same time.</p>
<p>I feel MIT's current approach is sensible, as they accept a limited number of people in the early pool, then are free to evaluate the strength of the RD pool -- if the RD pool were unusually strong, they would be admit lots of RD applicants and ~0 EA deferees. Since EA deferees are evaluated in a fresh light during RD (their applications get read again and summarized by a different reader; they are likely to be discussed in different committees), I think it's more appropriate to treat them like RD applicants than like EA applicants.</p>
<p>Perhaps at this point I am preaching to a choir on one side and brick walls on the other.</p>