Top 10 Yogism. 
I do think we can overthink it.
1 Like
Northeastern IPEDS/CDS numbers don’t include first years accepted to various special programs like NU IN, NU immerse, and probably more. Some school’s numbers don’t include spring starts, for example Middlebury and Tulane are in this group.
3 Likes
However, all of these situations are those where the financials are not that different between the different schools or residencies. To do a matching system for US universities, the universities would have to do financial aid pre-reads for each applicant, so that the applicant can rank them with consideration of the net price.
2 Likes
True. Which is why published acceptance rates are not always a reliable proxy for academic quality and selectiveness.
This!
4 Likes
Lottery effect, systematic mailings from the top-tier colleges to students, and test-optional policies have driven the numerators to the sky.
It will be curious to watch the top Ivy’s shuffle with Princeton and Columbia remaining test-optional while the others have reverted back to testing.
1 Like
They are also very far apart (deservedly so) in the US News Rankings. Interestingly, USNR doesn’t consider acceptance rate as a parameter. However, the rates may have some indirect impact on the “peer assessment”, which is a significant % factor in the rankings.
1 Like
In a broader approach — comprising acceptance rate, ACT/SAT scores and high school class standing — WalletHub arrived at these selectivity ranks as the top 10:
︎1. Caltech
︎2. MIT
︎3. Harvard
︎4. UChicago
︎5. Columbia
︎6. Stanford
︎6. Yale
︎8. Brown
︎9. Princeton
︎10. Dartmouth
1 Like
@CC_Jon why are you even doing this?
2 Likes
Actually looking at that table again I think what’s more interesting is the yield data. I guess I can understand why caltech is quite low (in this list) given the kind of school it is, but people on this list are clearly happier to get into MIT, Harvard, Stanford and Chicago than some of the others on the list, notably Princeton, Yale, Duke and Columbia. Assuming there is some overlap here (in terms of the same student being admitted to more than one of these) the choices of these clearly highly talented admitted students on where to ultimately attend could be insightful.
Chicago’s yield is heavily affected by its overreliance on ED. Without ED I believe its yield would be quite low in comparison to its peers.
Which is why this crude list is an exercise in futility. Way too easy to manipulate.
9 Likes
Oh, that’s a good point. Makes MIT, Stanford and Harvard yield with only EA even more comparatively impressive.
3 Likes
Acceptance rate is tough. Some well regarded schools - the Arizonas, IU, Bama, Oregon schools, Miami Ohio, IU, etc have high rates. UC Merced has a very high rate but is well ranked.
Yet kids still flock to these schools.
My kids got into low acceptance rate schools or in the major and yet chose safety, high admission rate schools.
Everyone sees it differently.
We absolutely looked at acceptance rates in selecting a list. But we still included high ones when realizing - in most cases a well known flagship is a flagship is a flagship.
Is UGA, in the real world, impactfully better than KU or Mizzou or Arizona? Is Florida really seen as superior over Miami Ohio or CU Boulder ?
So it is an overall art, not a science.
But those high admit schools give you a nice balance that allows one to stretch if they do desire, knowing that a win is in the offing no matter what.
I think acceptance rate is a measure of popularity relative to spots available amongst the application pool at large, and that’s about it. It is a partial but not full measure of selectivity, as it ignores self-selection in applicant sub-pools. It is not the best measure of reputation for quality education within academia; the peer scores from USNWR are arguably better for that, but still flawed as smaller schools are less known and the scores imo are likely more skewed by grad programs than intended. More important than any of this, though, is admit rates have almost nothing to do with individual fit, which is what I think matters most when forming an application list. Someone can say they really do care less about their actual education or experience than about exclusivity; I’m sure that happens quite a bit, however misguided I consider such an approach. I suppose it’s possible such views would be more common at the most selective schools, though the extent to which would be a matter of conjecture.
For those trying to gauge the value of these and other factors as they approach their college search, I strongly recommend the below paper out of Stanford’s School of Education. The title refers to rankings, but the scope is considerably broader and includes a discussion of the predictive value of selectivity as well. Here’s a quote: “We explore the research on whether attending a selective college predicts important life outcomes and find no significant relationship between a school’s selectivity and student learning, job satisfaction, or well-being.” There are some exceptions noted in certain contexts, but the paper indicates the poor to non-existent value of selectivity on learning outcomes has been consistently seen in the literature for about half a century.
If not selectivity or rank, what should a family shopping for a college prioritize? In short, the fit of a college in driving student engagement. Page 17 describes various avenues for engagement, from the classroom to independent project to extracurricular activity. Parents and students should ask themselves which schools or categories of schools a student would be most likely to engage in different activities based on what is already known about the student. For some learning styles, it might be a large state university. For others, a small, undergrad-focused LAC. For still others, a private university somewhere in between.
4 Likes
Yep. The NPC system could serve as the core of that, and possibly they could fold in some automatic merit as well (as some colleges already do in their NPC).
A reasonably scientific version of that sort of “revealed preferences” approach was taken in this paper, but it required a special data set and that data is now quite old:
Still, interesting that, say, Chicago did not do so well in this study compared to its normal US News rankings, but maybe consistent with its branding. One might even suggest this sort of study helps explain why Chicago felt compelled to make such aggressive use of binding programs.
2 Likes
UChicago utilizes 2.5 rounds of binding early decision, and it selects over 3/4’s of its class through this route. So very few seats are left, and it doesn’t have to accept many students to fill those remaining seats…it is a calculated way to have a very low acceptance rate and insanely high yield percentage.
MIT has a totally free choice early action program. Nothing is ever binding with them. So MIT’s low acceptance rate and insanely high yield is absent if any gimmickery. No games.
UChicago’s antics are very self demeaning and borders on comical, in the eyes of many
7 Likes
I would be interested in seeing your source for this, TIA.
1 Like
This is a really good point that I’d like to highlight. Down the road from me is CSUN, which accepts ~90% of applicants. It’s got some very good programs and the people I know who went there got a fine education to start their careers. It’s not particularly selective, but there is limited space so some students are turned away. (Hopefully they will go to community college instead.) There’s very little uncertainty around the admissions process.
UCLA is also nearby and is certainly a more prestigious school. It accepts ~10% of applicants. If you are applying to CSUN, it might be sensible to apply to UCLA because you’d rather go there if you got in. As you say, it’s a lottery ticket. For students with competitive qualifications, it’s not really 10%, but much better odds. For other students, it’s much lower odds. But that’s ok because the cost of applying is so low and the potential benefit is (at least in the imagination) so high.
The problem is that so many schools see acceptance rate as a metric of prestige. The means Goodhart’s law applies:
When a measure becomes a target, it ceases to be a good measure.
Encouraging high school seniors to “reach for the stars” sounds like encouraging advice. Go ahead and apply to that dream school. And if a student has a solid idea of their odds (very low unless they are excelling academically) it’s not irrational. But when everyone has an incentive to give students false hope, that’s not helping anyone. Schools shouldn’t get rewarded for artificially increasing applications in order to look better by rejecting more hopeful applicants.
A rating system really ought to account for that.
The study was published 20 years ago. At the time , Chicago had a >= 40% admit rate, compared to ~10% for HYPSM type colleges. Metrics of average academic quality of student body showed a notable gap. It was a different situation from more recent years, with Chicago having increased marketing to the general student population and more focus on rising through USNWR rankings.
Another key issue is self selection. This is an issue for early 2000s Chicago, Caltech, and numerous other colleges who attract a unique applicant pool who is quite enthusiastic about the college, much more so than the general pool of high achieving high school students.
For example, BYU typically has a yield of ~80% – usually on par with HYPSM… , in some years higher. BYU has a such a high yield because it usually the top choice among the largely Mormon application pool who chooses to apply. I expect many high achieving Mormon students would strongly favor BYU to Harvard, and those students who favor BYU generally don’t apply to Harvard as a backup in case BYU rejects them. That wouldn’t make sense. So the kids who favor BYU don’t impact Harvard’s yield. Harvard kids also generally do not apply to BYU as a backup, as typical non-Mormon high achieving students do not have BYU on their radar.
So you end up with BYU ranking high in selectivity among Mormon students and ranking relatively low among non-Mormon students. The study finds this effect in the regional rankings. BYU ranks after HYPSC at #6, among the 8 state region that includes Utah. However, BYU does not rank among the listed top 100 overall. If you split it among Mormon vs non-Mormon, I would not be surprised if BYU ranked as #1 most preferred college among Mormon and not in top 100 among non-Mormon.
Like BYU, Chicago was also self-selective in early 2000s to a unique type of student, much more so than today. The self-selecting students who like Chicago’s often favor Chicago over HYPSM, but don’t apply to HYPSM as a backup in case Chicago rejects them. That wouldn’t make sense with Chicago being 40% admit rate and less selective. So these students who favor Chicago don’t show up in cross admit states. However, the students who favor HYPSM… may apply to 40% admit rate Chicago in case HYPSM… rejects them. That would make sense, and those students do show up in the cross admit stats and revealed preference ranking study, leading to Chicago ranking relatively low overall.
3 Likes
Huh… this is an interesting take. Never thought about this before but it certainly is applicable to how my kid approached admissions last year.
As far as the original question on how one thinks about the acceptance rate when evaluating schools, the truth is, it wasn’t a factor at all in selecting which schools to apply to. I think my kid did a good job in really sussing out which schools he would be happy at and the acceptance rate was really irrelevant to him as far as figuring out the best fit. The only time acceptance rates mattered was in setting realistic expectations and figuring out a Plan B.
2 Likes
Using Yogi Berra and then Yogi Bear as your two quotes… 
1 Like