Domestic applicant admit rates vs. international application growth

Prep School Admissions
1

<p>Is there any data around the ratio of international applicants to total applicants at specific schools? I ask because it seems that international applications are likely one of if not the biggest factor driving the overall reduction in acceptance rates. Consider the numbers in this snippet form the Atlantic: "In six years, boarding schools like Deerfield and The Hotchkiss School in Connecticut reported a ten-fold increase in the number of Chinese applications. Each received less than 20 applicants in the 2005-2006 academic year and more than 200 in 2011-2012." If I assume that the trend has generally continued, Hotchkiss for example might have had just under 300 applicants from China alone in 2014. Scanning the web, I see total applicant numbers posted in the 1800 range, so if China alone has grown to be 300, and other countries are following a similar pattern, the international applicant pool as a percentage of the total pool should be extraordinarily high. Pairing this with the fact that Hotchkiss (just as an example here) is only 17% international in composition, and it seems like the math has to point to international students having a HUGELY lower acceptance rate than domestic. </p>

<p>In fact, I am having trouble even reconciling numbers...I have no firm numbers to work with, which of course is not a good place to start, but just to test the thinking around the differences in acceptance rates, I ran the following example for Hotchkiss ( I am not picking on this school, but had to pick something for the example). Most are based upon numbers that were documented somewhere, and of course could be wrong. If they are radically wrong it will matter. If they are slightly wrong, I don't think the result will change.</p>

<p>Input data:
Applicants = 1770 (Admissionquest website)
International applications = 600 (best guess)
Domestic applications = 1170 (result of guess above)
Overall acceptance rate = 22% (Admissionquest website)
Yield = 53% (Admissionquest website)
Total new students = 150 (guess based on student body of 600)
International student % = 17% (Admissionquest website)</p>

<p>If I run the yield/applicant/slot calculations, the above data would imply 150 slots at 53% yield would require making 283 offers. With 1770 applicants, this is more like a 16% admit rate rather than a 22% admit rate. To get to 22% you either need to have far fewer applicants, or a much worse yield rate. Not sure I think there is much of a chance that Hotchkiss is way south of 50% yield, so I am going to for the moment assume that the most likely thing here is the 22% is wrong and the admit rate is more like 16%. </p>

<p>With the above in mind, if the application distribution is as noted between domestic and international, and we assume that the 17% international student percentage mirrors the admits (has to be true if the time horizon is long enough), then the domestic admit rate would be just over 20% (126 slots at 53% yield = 237 offers on 1170 applications). By contrast, the international admit rate would be 7.5% (same math). This would imply to me anyway that much of the increase in selectivity is largely a data problem driven by out-sized international application growth rather than a more uniform change in the playing field.</p>

<p>So with the big quantitative preamble, I am curious if anyone has any actual data, or better math. I would further assume the conditional probability of domestic full pay has to only elevate the 20% to something much higher, perhaps as much as 30%. Curious if others have ever run the math, or seen the data...</p>

<p>PS - I spent a little time poking around on the topic and did not see any good information, so if I missed it and it has been covered elsewhere, I apologize...</p>

2

<p>Adding one more thing on the full pay vs. financial aid front:
Given the above numbers, if there were 1170 domestic applications, how many are for full pay? 40%? If so, this would be 468 applications. 37% of the students are on financial aid, so by my yield/slot math I think this would mean something like 150 domestic full pay offers for about 80 full pay domestic slots. Acceptance percentage for domestic full pay with this stack up = 32%…</p>

3

<p>I have no numbers, but I have heard that while applications from China have ballooned, applications from some other countries have dropped. Again, no official numbers-- and I think it’s true that overall international numbers are up, BECAUSE of China-- but maybe not quite as much as you’re assuming. As you say, maybe another reason to need some real numbers.</p>

4

<p>In general, yield rates vary. A student with a generous financial aid award is more likely to attend than a student whose parents can pay full tuition. The full-pay student is more likely (if a strong student, well-suited to boarding) to receive more than one offer, and to have good schools at home. I think legacy students are more likely to yield, unless the kid has multiple legacy ties in the family. </p>

<p>The numbers also include kids applying from feeder schools and prep-preps. By March 10th, the admissions officers may have a good idea where they stand with the families. A kid from Superfragilistic Preprep might have applied to four schools–but the advisors at SP might have passed on the word to Hotchkiss that it’s the family’s favorite. The advisors might also pass on that the school isn’t at the top of the family’s list–wouldn’t Hotchkiss like to consider this other SP student?</p>

<p>So, maybe it’s 32% overall for domestic full-pays, but that doesn’t mean each full pay applicant has a 32% chance. Some kids have a 100% chance. Some have no chance (and academics aren’t necessarily a deciding factor.) Take a look at old threads on results, posted around March 10th of each year. You’ll find many applicants and parents flabbergasted at results. (really, please do this.)</p>

<p>Rather than trying to use statistics to guide your choice of schools, you might want to hire an educational consultant. Very often, schools know the “type” of kid who do well at each school. Some consultants have worked at different schools, so they have connections, and can guide you to strong schools which like your “type” of kid. If your child attends a preprep, the school’s counselors should be able to guide you to schools which like their students. </p>

<p>As a side note, do you know who’s behind Admissions quest? Are their data reliable? Boardingschoolreview is the site people here usually refer to. </p>

5

<p>@Daykidmom‌ - While China may be leading the way, I find it hard to imagine that the rest of the world as an aggregate is not also on the rise. Sure, some countries may be down, but the core reason Chinese applications are rising has to be affecting other countries as well. However, no data, no certainty and I am in the dark…</p>

<p>@Periwinkle‌ - thanks for the detailed and thoughtful commentary. In response, my posting is not so much about trying to select schools through statistics. Rather, I simply have become curious about what is driving the increasing nominal competitiveness as reported in acceptance rates. I guess I just started to wonder if it was really possible that the world had changed so much in so short of a period. After running into the Atlantic article, it led me to want to consider the actual distribution of applicants with the hypothesis that substantial increases in the international application pool may be driving very high levels of overall selectivity without having changed the dynamic for the historic domestic applicant pool. I figured given the popularity and history of this site, perhaps others had already done the research, gathered the data and run the math. The exercise has little to do with my own kids school selection process…As for admissionquest, I cannot vouch for its accuracy, but can at least say that with the Hotchkiss info I used, both sites are comparable in information.</p>

6

<p>

</p>

<p>The key word in your statement is “overall”.</p>

<p>For the very selective, mega-endowment schools, the huge number of applicants from china reduces the admit rate of applicants from china. This reduces the school’s “overall” admit rate when u average all the numbers. But it does not reduce the domestic applicant admit rate.</p>

<p>Domestic applicants & int’l applicants are in separate admissions pools. At the very selective boarding schools the number of int’l applicants admitted from any one country is capped. Therefore, the huge increase in applications from china is only a problem for u if u are an applicant from china. </p>

<p>For the less selective, less well-endowed schools which are struggling financially, fullpay rich kids from china will supplant financially-needy domestic kids.</p>

7

<p>@GMTplus7‌ I agree with your assessment, and this was what I was wondering about when I began the math exercise. I read a lot, on these boards and elsewhere, about how selectivity and exclusivity have increased year after year at the top schools. As is often the case with simple “overall” statistics, I suspect the comments about a never ending staircase of selectivity are really missing the details of the applicant to slot mix distribution. Perhaps only people in the admissions offices can know with certainty, but I strongly suspect that selectivity has not increased in a homogeneous way across all applicant groups but rather is the product of disproportionate growth in certain populations, like international students. As an aside, your comment has now left me curious to go and see how correlated international student % is to financial strength…If you are right, there should be at least a mild inverse correlation.</p>

8

<p>I don’t think anyone knows the answer, other than admissions directors at each school. I have the impression it can vary from year to year, often driven by currency fluctuations. Some schools count US citizens living overseas as international students. I think this also varies. Most schools will list the countries represented in the current student body. However, a number of students will have dual citizenship. Kids living on Bermuda may be US citizens. </p>

<p>It’s rare for international non-citizens to receive financial aid. The international applicants are often very smart, hard-working, and full-pay. They’re often musical, and usually not natural helmet sport athletes. The schools do cap the number of applicants from each country. Overall, though, if a school admits 10% (or whatever) foreign students, that means 10% of the seats are no longer available for domestic applicants. So it could be more competitive for all students, in comparison to the '80s. </p>

<p>The thing to watch out for is how the students mix with each other. If they break into cliques, that’s not good. You can ask about this on visits. Tour guides are good sources of information. At many schools, you can stay for lunch, and observe students as they interact.</p>

<p>HOWEVER, I hate to think of your going through massive data searches to try to judge financial strength. If you’re serious about a school, just look up its 990s. Guidestar publishes them, with a two year delay.</p>

9

<p>@Periwinkle‌ Good advice on the clique front. Duly noted. BTW - I came across a thread you started from some years back around over emphasizing SSAT scores. I have similar statistical curiosities on the SSAT average dimension as well, but tend to think this number may be less misleading than the acceptance rate numbers. Why? Unless I am mistaken, the average SSAT number is only measure on the enrolled students. Therefore, there is a lot less noise in the data. Inflated number of applicants from any source should not grossly skew the SSAT average score. If this is the case, then by and large it would be hard to have an average score around 85% (following on Hotchkiss data) with many more than around 25% in the 70s. Otherwise the remaining 75% of actual enrolled students would have to be into the 90%s for the math to work. So, while SSATs might not be everything, if the averages are as advertised, I have a hard time imagining they are taking a ton of students who are very distant from the average - some yes, but not too many. </p>

10

<p>SSAT scores have been creeping up, but I suspect it’s due to kids nowadays retaking the test many times. I know of some who have taken it 4 times (not my kids!). There was a mother who started a thread asking if her daughter shd retake test w a 98th percentile score-- it’s gotten crazy…</p>

<p>Also, I wonder if some schools are reporting the median, rather than the avg.</p>

11

<p>Could you link that thread? We’ve discussed the SSAT on this site so many times over the years, it all blurs together. In general, international applicants score well on the SSAT (my impression.) It makes it easier for an admissions team to meet SSAT average goals, if they have a supply of smart, full pay students.</p>

<p>Only the admissions teams know the distributions of scores. Some schools post the school profiles they release to colleges online. I gather SSAT score percentiles are not wildly different from SAT score percentiles, so the distribution of SSAT scores probably roughly resembles the distribution of SAT scores 4 years later. (Remember students leave & join the class between 9th and 12 grade.)</p>

<p>I do try to caution people not to assume a high SSAT score is an “Ace in the hole,” particularly at top schools. </p>

12

<p>Link to an article for a well know school. <a href=“http://blogs.groton.org/circle-voice/2014/05/01/groton-admits-12-4-percent-applicants-ninety-nine-admits-return-revisit/”>http://blogs.groton.org/circle-voice/2014/05/01/groton-admits-12-4-percent-applicants-ninety-nine-admits-return-revisit/&lt;/a&gt; Some information and numbers on admits, financial aid, full pay and international applications in the article. Check the graph for the top 6 countries and states for applications. </p>

13

<p>Wow, the data from Groton is eye opening. I can’t believe that my son was one of the 50 admitted with financial aid. This has been amazing experience and he hasn’t even arrived on the circle yet!</p>

14

<p>@Periwinkle‌ - the link to the old thread is <a href=“Caution about SSAT Scores - #4 by warriorboy648 - Prep School Admissions - College Confidential Forums”>http://talk.collegeconfidential.com/discussion/comment/5113968&lt;/a&gt; I mentioned it because as you will see in rereading parts of it, there is some back and forth about the importance all the way from “be above 85% or forget it” to “unless you are in the single digits it does not matter.” My comment about the stats distribution was really around the fact that if SSAT scores are averages and not medians, and they are measure on enrolled students vs. offers, I don’t think you can have much of the population super far from the average if averages are in the high 80s-90s. There just is not enough room above the average to compensate for a larger group significantly below it. I have wondered form time to time about how the SSAT scores distribute over the ages of the applicants. I would expect that the same student as a 10th grader would do better than as an 8th grader on the upper SSAT. If a school has an average of XX%, I would assume the 9th graders being admitted are generally lower on the scale. Not sure if this is valid…</p>

<p>@ForkNotTaken‌ Some of the most detailed stats I have seen made available, and definitely supportive of the overall thesis. There are some numbers we don’t have even with this info (e.g. international vs. domestic split of FA applicants), but you can run some basic numbers on the details still. Here are a few things that struck me from the Groton info:</p>

<p>Full pay “admitted” rate = 97/663 = 14.6%. As these are admitted students vs. accepted students, there is a yield factor missing to get the actual acceptance rate. If I assume FP has a lower yield than FA and set the FP yield to 60%, the acceptance rate for full pay would be about 24%.</p>

<p>Financial aid “admitted” rate = 50/518 = 9.7%. Assume yield per above of 80% for FA students to make higher than FP, and acceptance rate would be 12%. </p>

<p>China + Korea + Hong Kong applicants = 228.
Total Asian applicants enrolled = 14
“admit rate” for this population <= 6%</p>

<p>If 228 applicants had an admit rate of 6%, 518 students had an admit rate of 9.7%, and the overall admit rate was 12.5%, the remaining 435 students would mathematically need to have an admit rate of about 19%…Add yield factor and you can imagine this population has an acceptance rate of closer to 30% than 12%.</p>

<p>Anyway, just more support for the asymmetry in the acceptance numbers, and why averages are very likely misleading to some degree…</p>

15

<p>Very misleading. For one, “full pay” covers a large category. Another fact is the number of activities the schools support. The athletic teams are very important. Few schools are o.k. with losing. So after a certain level of SSAT, really, a great lacrosse player who need a small amount of financial aid might have better chances than a full pay pianist who’s never played team sports. It depends, again, on the sports the particular school cares about. </p>

<p>If your child plays a sport, he or she should try to talk with the coach. I have the impression many parents without experience with the schools think it’s academics first, then money, then extracurriculars. I don’t think that’s the case. It’s more, will this student join the life of the school, on the playing field and the classroom, is there a fit with the child and family, and can we afford him? (For the schools which are not need blind, which is most of them.) Boarding school is 24/7, it’s often a fairly isolated community, and the kids who end up in the community need to be able to get along with others.</p>

16

<p>@blackbeard, I am still trying to wrap my head around the “actual acceptance rate” concept you calculated. The admitted number times the yield percentage gives you a matriculated number, yes, but I think that you may have differentiated “admitted” and “accepted” when they are the one and the same, used interchangeably by these students reporting on revisit day in early April for those previously accepted/admitted. </p>

<p>If we assume a 70% yield overall, the matriculated students come to 103. 20-30 are in eighth grade, about 45-50 would be added to the rising ninth grade, and the rest (20-35) added to rising tenth and eleventh grades (sorry for not writing “II Form”, etc.) for a school total of about 380 students, boarding and day. </p>

<p>What these statistics support is how skewed foreign applications are to China/Hong Kong/S. Korea, and how those nations drive down the overall acceptance rate (as do multiple applications from same students). Around 75% of the foreign apps come from these three, yet only 20 or fewer of the 38 accepted internationals are likely from these places. Meaning, it’s so hard to be accepted from China/Hong Kong given the plethora of applicants from this country for so few spots! And by my reckoning from the numbers given, zero of these international applicants are seeking financial aid. </p>

<p>The implication for Americans is that the acceptance rate for full pay applicants is even higher than 97/663 = 14.6%. I would calculate 97 minus 38 for 59 accepted FP Americans out of 663 minus 318 for 345 FP American applicants, and get a rate of 17%, almost double the rate for the FA students. </p>

<p>The student reporters at Groton gave seriously bad information on the FA budget. It is in fact close to $6 million for about 38% of the students, coming out to a generous average gift of over $40,000 (for a very high tuition).</p>

<p>Finally, Groton is getting about half of the domestic applicant pool (429/863) from just Massachusetts, New York and Connecticut. This supports my contention that even the elite schools that seek diversity will still have a strong regional identity deriving from their location and history (of course) but also from the nucleus of students enrolled from immediately adjacent states. *** The diversity factor also implies that strong candidates from less represented foreign nations and American states may have better odds of acceptance. ***</p>

17

<p>@Charger78‌ well, having read your post, and rexamined the Groton info, I suspect you are having trouble wrapping your head around the numbers because they are screwed up, at least in terminology. As you noted, I think the numbers are the acceptance rates, where I had thought they were the matriculated numbers, and was therefore adding the yield back in to get higher total acceptance numbers. I see this is incorrect, as the 147 is pre-yield application…</p>

<p>However, the adjustment makes the calculated results different, but does not alter the conclusions at all. It is of course well known that being a domestic full pay student from an underrepresented state is advantaged over being a financial aid applicant from China, or Connecticut for that matter. I just wanted to get a rough idea of the magnitude of the advantage. Most of the numbers I have run suggest a 2-3x differential…</p>

18

<p>Students apply to multiple schools. Nothing tracks that side of the equation, to my knowledge. To make up an example, an outstanding football player, full pay, from an underrepresented state, with great academic credentials, may end up with 5 acceptances. A very academic student, full pay, could apply to the same schools, and end up shut out, or with 5 waitlists. It depends on how adeptly the student and his advisors gauge his chances at each school. </p>

<p>Yes, being full pay is a huge advantage, however, if your list is Andover/Groton//Deerfield, you may not end up with an acceptance.</p>

19

<p>^^^ Yes. Still better odds being that second very academic kid and applying from Missouri rather than Massachusetts.</p>