<p>
Each college has a different admissions group, which weights factors differently, and in some cases considers different factors. For example, if you compare the acceptance rate between male and female apps with similar stats among the Parchment data, the acceptance rate for males and females is nearly identical and Stanford, HYP, and most other selective colleges. However, at MIT it’s not even close. In some stat groups, the MIT female acceptance rate is several times greater than the male rate. The degree of increase was similar to URM vs non-URM. This fits with MIT being one of 2 tech schools I am aware of with a well balanced gender ratio. Caltech shows a similar pattern, but to a lesser degree.</p>
<p>Sometimes the CDS lists very different admissions criteria at different selective colleges. For example, Brown’s criteria rankings listed in the CDS mark level of interest, personal/character qualities, and talent as most important… more important than the stat criteria. In contrast, Stanford and HY (not P) say they do not even consider level of interest. The lower weight on stats at Brown fits with Brown having a far lower 25th/75th SAT than one would expect based on the acceptance rate.</p>
<p>The biased sample of posters on this forum fit with the comments above about Stanford placing less emphasis on test scores. Among posters in the RD thread (an obviously biased sample), the rejected posters had slightly higher stats as a whole than the accepted posters, particularly for class rank. This pattern does not appear in RD threads for any other selective colleges I am aware of. The rejected high stat posters generally were accepted at other highly selective colleges (especially Yale). </p>
<p>In spite of all of these differences, there is still a notable correlation between acceptances/rejections at different selective colleges. A student accepted at Stanford has a much greater chance of being accepted at HYPM than the overall app group and vice versa.</p>