<p>OK Sam and esquared, at the risk of going over your heads, here is the problem:</p>
<p>Sam decided to rebut JHS’s analysis by claiming, without any substantiation, two things: 1. that some fraction of the EA applicants to U Chicago don’t have a choice because they don’t apply anywherer else and (2) of those that have a choice, “fewer people have Chicago as their first choice whereas at least a quarter of NU student body makes it their first choice.”</p>
<p>Now these are interesting arguments: (1) depends on unknown information, although SL feels free to speculate and (2) depends on imputing behavioral intent regarding the UofC applicants, that, since they had a choice and chose to apply elsewhere, UofC was not their first choice. We don’t know this. We also don’t know what percentage of the EA applicants applied elsewhere to a ED school. Only the latter would be consistent with SL’s arguments. </p>
<p>But wait, there’s more! Argument (2) confuses “first choice” (as in applying somewhere ED) with yield. This is a particularly troublesome issue because many ED applicants are not applying to the school that is their “first choice” but rather trying to maximize their odds of being admitted to an acceptable school. If you don’t believe this, just read any of the many articles published in recent years about how to pick one’s ED choice and the role of odds maximization in such a process.</p>
<p>I can’t speak for JHS, but many readers would interpret a comment like “reasonable criticism” to mean pointing out issues that could ideally be included in an analysis if only the data were available. Much research is done in this fashion, with critics pointing out what might be done in an ideal world. But the process of research is beyond the scope of this discussion, so I’ll stop here.</p>