<p>
</p>
<p>Excuuuuuse me! Choosing Yale over Harvard is NOT an example of turning down a college with a clear higher ranking. And hundreds of students do it every year (over 1/3 of people admitted to both schools), generally for the reasons cited in the post.</p>
<br>
<br>
<p>Is this true???</p>
<p>Since both schools have such low acceptance records and are “super reaches” for even those with perfect stats, isn’t it strange that 1/3 of accepted students get accepted to both? You’d think the odds would be against that. But, I’m not a math brain.</p>
<p>And, you’re right…decling one over the other is not what this thread is about. And, what scholarships did her D get? I didn’t think the ivies really gave scholarships. I thought they only gave financial aid (which isn’t a scholarship).</p>
<p>I think she meant of the students accepted at both (whatever that number is), 1/3 choose Yale.</p>
<p>MImama, he’s trying to go pro in his sport but isn’t having super success. A Yale sheepskin might have been a good idea in the long run, IMHO.</p>
<p>oops post 142… should say “declining” (I hate that we can’t correct our typos after 20 minutes! 
)</p>
<br>
<br>
<p>A superstar’s son went to a lower-tier school by my home to play a sport. I’m sure he could have gone elsewhere. Weird! (Actually, I should say the son of two Hollywood stars, since both parents are big stars. When the son would have a home game, we’d know that they’d be in town.)</p>
<p>
</p>
<p>I agree with JHS. When the USNWR rankings (or college rankings of any kind) of two colleges are very close, it is hard to say which is clearly higher ranked, and we’ve seen rankings change every year.</p>
<p>Re: competition for cross admits among the elite schools – here is a table from NY Times in 2006. Harvard is clearly the favorite with cross admits.</p>
<p>[The</a> New York Times > Week in Review > Image > Collegiate Matchups: Predicting Student Choices](<a href=“http://www.nytimes.com/imagepages/2006/09/17/weekinreview/20060917_LEONHARDT_CHART.html]The”>The New York Times > Week in Review > Image > Collegiate Matchups: Predicting Student Choices)</p>
<p>Of course Harvard is clearly the favorite. Only about 450 kids/year turn it down – no other college comes close. But the ones who do turn down Harvard generally turn it down in favor of Yale, Stanford, Princeton, or MIT, and enough people do that that it’s not a terribly surprising or interesting choice.</p>
<p>I’m actually surprised by the cross admit pick ratio between MIT & Harvard/Yale. By and large, MIT applicants are interested in science and engineering, which MIT specializes in. Why would Harvard and Yale still beat MIT in cross admit preference? This led me to speculate that many folks still would pick higher prestige, however slightly, over better program, even when both choices are in the rarified realm.</p>
<p>Personally I do not think of Harvard as having more prestige than MIT. Quite the opposite. </p>
<p>Perhaps the cross-admits are choosing Harvard/Yale over MIT because they want a more ‘normal’ social environment?</p>
<p>Perhaps. But, if that was the case, shouldn’t we see the same tilt to Stanford and Princeton?</p>
<p>There is actually no evidence that MIT loses cross-admit battles with either Harvard or Yale (The NYT chart is NOT a compilation of actual cross-admit data). The overlap in admits is simply not very large as can be quickly derived from the high yields of all 3 schools. CC runs a yearly cross-admits battle among its HYPSM admits and has found MIT to run even with HYPS.</p>
<p>
</p>
<p>You have to keep in mind with this chart that “cross admits” are only students who applied to two schools with similar programs. I can’t imagine many, for example, interested in engineering at MIT would have any interest in even applying to Brown. </p>
<p>So, it’s not really an accurate assessment, seeing as many schools don’t have parallel program offerings and only attract applicants interested in their specialties.</p>
<p>It’s quite possible that even more than 1/3 of cross-admits turn down H for Y now. </p>
<p>The Revealed Preference Study that Paperchaserpop links to was conducted back in 2005, and it was based on data collected several years earlier from a group of “theoretical” cross-admits, not actual applicants. (Participants were asked, “If School X and School Y were both to accept you, which would you choose?”) That’s the study that says the cross-admit split is 65/35. There’s newer evidence, in the form of a statistical model, that Yale actually won the cross-admit battle in 2008, 56% to 44%.<a href=“Mathacle's Blog: June 2008…-stanford.html%5B/url%5D”>Mathacle's Blog: June 2008…-stanford.html</a> </p>
<p>Like the Revealed Preference Study, the 2008 statistical model may or may not be accurate. Of course, neither school discloses the data from which to make the determination. </p>
<p>But it’s fun to speculate. S turned down H for Y, too. ;)</p>
<p>I’m not going to take the time to go back and check, but I thought the Revealed Preference Study DID look at actual cross-admits and where they actually went. (That was the “revealed preference” part; otherwise it’s just an opinion survey.) Where it got controversial was in its use of statistical techniques to imply a preference ratio between colleges where there was not a large number of actual cross-admits attending one or the other in its data set. But (at least this is what I thought) it should have been pretty accurate for pairings where there WERE a large number of cross-admits who chose one or the other.</p>
<p>As for MIT-Harvard – there is a huge overlap in the students to whom they appeal. Harvard is very strong in things like math and biological sciences where MIT is also very strong, and Harvard has been working to beef up its engineering. Meanwhile, MIT has generally sought students who were not unidimensional (i.e., students to whom Harvard might appeal despite their general math-science-engineering orientation). I know very few students who applied to MIT who did not also apply to Harvard, and most of the ones who didn’t apply to Harvard were making a judgment that MIT was already enough of a reach for them and they didn’t have a meaningful shot at Harvard.</p>
<p>Based on what I see in real life, I have trouble believing that (a) more than a third of people accepted to both Harvard and Yale who choose one of them choose Yale, and (b) as many as a quarter of the people accepted to both MIT and Harvard who choose one of them choose MIT. I do believe, however, that YPS and MIT account for the vast, vast majority of people who do not accept an offer of admission from Harvard.</p>
<p>“Perhaps the cross-admits are choosing Harvard/Yale over MIT because they want a more ‘normal’ social environment?”</p>
<p>Well, I wouldn’t use the word ‘normal’ : 0 but D definitely found Yale to be a more cheerful place. We just visited her this weekend and she seems quite happy. </p>
<p>As for rankings, i.e. the OP’s question, the ranking of the depts in Ds field of study did matter to her. Not “US News and World Report,” but rather, NRC rankings in her field of study, even though they, admittedly, ancient, but nonetheless give a snapshot of graduate programs… which were relevant to her case, since she knew she’d be taking advanced courses. Now, for the question of what worth do rankings have, whether ancient or not, I don’t know, don’t want to go there : (</p>
<p>Ds experience showed that campus visits really seem to have mattered, as part of a careful and well-organized search. The visits, both prior to applying and after admittance helped her to answer (for herself) that hard-to-quantify question of “where do I think I’ll be happiest over the next years?”</p>
<p>JHS – I don’t recall where I read that the Revealed Preference Study was based on data collected from theoretical cross-admits. My info could be incorrect. I can’t seem to download the entire study; all I get is the abstract.</p>
<p>
</p>
<p>That is actually not true. I am an MIT applicant interviewer and relatively few students I see also apply to Harvard. Among those admitted, the overlap is even lower. MIT is very focused on recruiting the Intel/Siermens/USAMO/IMO finalists and will accept more angular (strictly math/science focused) students than Harvard. Most MIT applicants take the AMC/AIME wile the Harvard applicants most often do not. Between 55% and 60% of MIT students major in engineering and have essentially no interest in Harvard. (There is a stronger overlap with Stanford and Princeton for instance). A substantial number of MIT applicants are also attracted to business and are essentially picking between Penn/Wharton and MIT/Sloan. There is some overlap in the sciences, mostly math and biology. According to the annual meetings we have with the admissions office, MIT gains as many cross admits as it loses from Harvard (It conducts post-admit surveys of admitted students). As far as MIT and Yale, the overlap in applicants is even much smaller than between Harvard and MIT. </p>
<p>In general, I do agree with the statement (and I think the data does show) that the vast majority of students admitted to one of the HYPSM schools but electing to go somewhere else matriculate at another school within the pool.</p>
<p>PS: While the Revealed Preference Study does utilize admitted student data, the cross-admit numbers suggested are NOT based on actual cross admit data but on a modeling of preferences expressed by students. Even the authors recognize their model is vastly underpowered (in terms to numbers) to make any real predictions. Only 17 schools had more than 50 total data points (certainly none were HYPSM). The study was designed to show how a model COULD be constructed if enough data was collected. Even then, the core assumption in their model that if more students prefer school A over B and more students prefer school B over C, then by implication more students will prefer school A over C is largely flawed.</p>
<p>Thanks for the clarification on the Revealed Preference Study, cellardweller. So for the statistically challenged among us, does that mean that the cross-admit numbers were based on a group of theoretical admits? Where in the study was actual admitted student data utilized?</p>
<p>I have a non-engineering kid who would have picked MIT > H. Decided to apply to H the second half of December (after they announced their new FA initiative two years ago), figuring that between the FA and the new funding towards expanding CS, it might be worth it to check out.</p>
<p>Chicago > MIT is probably a math or econ major preference vs. the sciences, but Chicago wasn’t in that original revealed preferences study.</p>