@Sue22 The acceptance rate is not affected by he yield rate. It’s only a measure of how many students were offered admission. The yield rate can go to 0 and it wouldn’t change the acceptance rate which is only a measure of how many applied we versus how many were offered admission.
Now read some of the posts analyzing why the acceptance rates of many selective schools are the lowest ever this year. There may be multiple factors, but assuming that the most obvious reason is the increased overlap applicants, I predict the following.
Unless these schools used extremely advanced AI algorithms to protect the yield, they will end up having to utilize more of the waitlist pool to fill the class. Many of them must have simply relied on previous years’ statistics.
Of course, I might be wrong. But assuming that (1) the overlap increased and (2) grounds for admissions decision were not very different from previous years’, releases from the waiting list will be greater.
So many knowledgeable sources here…
I understand how colleges use early decision as a way to increase yield (important for institutional planning and a way to game USNWR). Does anyone have an explanation as to why the ED admission rates at certain schools are much lower than the general acceptance rate? For example, SMU has an ED acceptance rate of 36% and a overall acceptance rate of 49%. Does it make sense for a student to apply ED to a school like this even if it is their first choice school?
@civitas, if the yield rate were 0 the class size would be zero. Schools have to adjust their acceptance rates based on how many students they think they’ll yield. That’s why Harvard accepts half as many students as Wake Forest for a larger class-Harvard yields more.
To give some hypothetical numbers, if a school is aiming at a class size of 1,000 and they expect a yield of 75% they need to accept 1,333 students. If they expect a yield of 50% they need to accept 2,000. If their yield drops to 25% they need to accept 4,000. As you can see, a school like Harvard, with an incredibly high yield, isn’t going to be affected as much by this as is a school with a lower yield.
Schools that have done things that raise the number applications, like eliminating a supplementary essay or fees have seen their application number skyrocket, and while it also usually means lower acceptance rates the number are not in the same proportion because some of those application are from those unlikely to enroll, such as students who add the school as an easy to apply to safety at the last minute.
@chable, just a theory, but it may have to do with the different ways students applying to the schools view ED and the populations that make up those ED pools.
At a lot of schools, particularly the more competitive LACs, a large part of the ED pool is made up of athletes and legacies, both categories that have a distinct advantage at these schools. At a NESCAC school (Williams, Amherst, Bowdoin, etc.), for instance, coaches only have pull in the ED round and legacies (at most) are told they will only be given a bump if they apply ED. For that reason the ED acceptance rates are high even if a sizable portion of the ED pool is made up of kids for whom the school is a reach.
At other schools, particularly schools where recruited athletes don’t need to apply early, where there’s no early legacy bump, or where a portion of the class knows they will come in as an auto-admit, the pool of ED applicants is likely to be largely made up of kids for whom the school is a reach, meaning their stats are in the lower end of the overall admission pool.
Just a theory, so someone may have a better one.
I understand yield. My point is it’s not part of the acceptance rate calculation. Your last sentence in your previous posts suggests the acceptance rate calculation would specifically rise or fall based on the eventual yield rate. If that were the case we couldn’t even post acceptance rates yet because they would all change subject to the eventual yield rates when students have all accepted their offers. A school can definitely lower its acceptance rate by receiving more applicants relatively to the number of acceptances. Each school needs to understand if the larger pool is more or less likely to select them if accepted. Stanford will likely yield a similar percent of its acceptance pool regardless of its applicant base while a another school might predict a lower yield and decide to adjust their acceptance pool accordingly. Perhaps this last part is what you are saying but it’s an indirect contributor to the equation not a direct one. We are listing for the most part all highly selective schools here who will do relatively well on yields. It would be interesting to see the acceptance rates and yields on schools not in the top 200.
@tovelove I have been wondering the same and I don’t think anyone knows yet whether the rise in app numbers reflects more applicants or more apps per applicant. I suspect the latter but it could be a mixture of both.
I too suspect schools will need to go to the waitlist more this year and I think they would much prefer to do so rather than deal with the overenrollment that was common for many selective schools last year, when apparently they underestimated yield. I haven’t done so, but it might be interesting to compare actual numbers of accepted students by school for this year and last (e.g. BC accepted 800 fewer students this year than last after overenrolling by 100+ last year).
@citivas, They do change if the school goes to their waiting list. That’s why the numbers released now and the numbers reported in the common data set are often a bit different.
Colleges use sophisticated metrics and historical data to predict yield. That includes overall trends and changes specific to a school. Win a national championship? You can predict a slight rise in yield. Have a scandal or drop a program? Yield might go down. They can’t always get it right, which is why they sometimes end up overenrolled or going to the waiting list, but they’re usually pretty darned close. If schools were simply relying on the previous year’s numbers you’d see every school going to the waiting list every year.
One real-world example. It’s old but it does demonstrate what I’m talking about.
In 2008 BC received 8093 applications. In 2009 it received 8805. Despite the bump in applications the school accepted a higher percentage in 2009 (30%) than 2008 ( 26%) and ended up with a class of the same size (2167 in 2008 vs. 2172 in 2009). Why? Because the yield dropped from 27% to 25%. The enrollment as percentage of applications remained roughly the same, at 7%.
The low admit numbers are clearly a result of students applying to more schools. Yes, there are new additions to the applicant pool: more international applications and more applications from first gen students and likely Pell recipients, largely as a result of top tier schools aggressively courting those groups. The net result is that h.s. students see these breathtaking admit rates and double the number of schools they apply to, just as you’d raise your chances buying another lottery ticket or two. It’s especially bad if you are not a URM, recruited athlete, legacy, first gen, or potential Pell recipient.
Back to test-optional for a minute; does anyone believe what schools say about the reasons for doing this? I’m sensing a pretty cynical consensus here. Does anyone take, say, Wesleyan at their word when they say: “…We are unconvinced that standardized test scores accurately reflect college potential for all students, and believe that test scores unfairly advantage some applicants…”? Surely somewhere in this whole grim business there has to be a place for a few colleges acting on behalf of the stressed-out, endlessly-jerked-around high school kids who are just looking for a school where they can thrive.
@sue22 BC is an interesting example for a lot of reasons. I can’t remember when, but 4 or 5 years ago they added a supplemental essay and their apps plummeted. Their class of 2022 acceptance rate is 27% (down from 32% last year)…finally getting “back to” where they were 10 years ago.
Meanwhile…you have a school like Colby that comes up with a program every year or two to drive applications…no essay…fee waiver…now a commitment to families earning under $60k / year for a free year of tuition…and their applications are going to triple over an 8 year period.
In either case, does the composition of the class change much? I doubt it.
In BC’s case, I think they weeded out the “what the hell…it’s only $60” apps. Kids who really thought it was a place they wanted to be still applied. For Colby, it’s the “what the hell…it’s free and easy”. They didn’t get apps from kids who looked at the school as a match, they got a lot more apps from kids who were hoping for a reach. Now, they’re going to pander to those financially challenged with an opportunity that means almost nothing. If the student with a family income under $60k were to gain acceptance to Colby without the program, they would still get a lot of aid. For 1 of the 4 years the school is turning a small cost into zero, and the applications will bump again.
The impact to the kids is almost zero…but Colby gets to suggest they are more selective than ever.
@penandink If the school’s motive was altruistically to create less stress for students, again why not just go to not asking for test scores before admission, period. They could still ask students to submit them after admission and still report the averages for the CDS and rankings, but students would know with confidence it wasn’t a factor in admission because they only had access to the data after the decision.
Quick and dirty, taking last year’s data mostly from here http://talk.qa.collegeconfidential.com/discussion/comment/20728550/#Comment_20728550 and this year’s data from post #321, here are changes in numbers of acceptances from last year to this year. 11 schools accepted more students, 23 accepted fewer.
school / #accepted 2022 / #accepted 2021 / difference / percent change from last yr
Barnard 1088 1139 (51) -4.48%
BC 8400 9200 (800) -8.70%
Bowdoin 935 963 (28) -2.91%
Brown 2566 2622 (56) -2.14%
BU 14184 15203 (1019) -6.70%
Colby 1602 1750 (148) -8.46%
Columbia 2214 2185 29 1.33%
Cornell 5288 5889 (601) -10.21%
Dartmouth 1925 2092 (167) -7.98%
Duke 3097 3116 (19) -0.61%
Emory 5135 5172 (37) -0.72%
Georgia Tech 7832 7297 535 7.33%
Harvard 1962 2056 (94) -4.57%
Harvey Mudd 594 566 28 4.95%
Haverford 877 859 18 2.10%
Johns Hopkins 2894 3133 (239) -7.63%
Middlebury 1696 1753 (57) -3.25%
MIT 1464 1438 26 1.81%
Northwestern 3392 3371 21 0.62%
Penn 3731 3699 32 0.87%
Pomona 713 741 (28) -3.78%
Princeton 1941 1890 51 2.70%
Santa Clara 7954 8067 (113) -1.40%
Stanford 2040 2050 (10) -0.49%
Swarthmore 980 960 20 2.08%
Tulane 6598 7477 (879) -11.76%
USC 8258 8980 (722) -8.04%
UVA 9850 9957 (107) -1.07%
VIllanova 6545 7578 (1033) -13.63%
WashU 4695 4875 (180) -3.69%
Wellesley 1267 1197 70 5.85%
Wesleyan 2186 1932 254 13.15%
Williams 1163 1253 (90) -7.18%
Yale 2229 2272 (43) -1.89%
@penandink - Bowdoin has been test optional since 1969. I’m willing to give them the benefit of the doubt.
@evergreen5 Is this a change in absolute percentage or relative? In other words, is this admissions were down 2% or the admit number went from 14 to 12%? Etc.?
@citvas - its the change in accepted headcount. Take Yale
Yale 2229 2272 (43) -1.89%
The negative 1.89% a reduction percentage vs. last years headcount (2272).
2272 X -1.89% = 43 less people accepted (2229)
@Sue22: in post 347 you must have meant that in 2008 / 2009 BC accepted 8093 / 8805 (you wrote received). The numbers of applications received must be substantially higher if a class size of roughly 2000 constitutes 7% of the applicant pool.
@inboston, oops, you’re right! That’s what I get for typing too fast.
Is it fair to say CalTech has kids with highest GPA and test scores? Does being most selective have to do with lowest acceptance rate or combination of low acceptance rate and high yield rate?
Whoever is adding to this list can add Vassar College. My husband is an alumni interviewer, and he was just sent an email with the stats. There were 8,312 applications and 24% were accepted.
@websensation Caltech has the highest and Uchicago the second highest. I saw a list last year.