Why it's so hard to get into an under 15% acceptance rate school

Applying to College
121

At a 10% acceptance rate college, different individual applicants will have completely different chances of acceptance that may be very far from 10%. A large portion of applicants will have effectively 0% chance of acceptance. While others will have a high chance of acceptance, in many cases high enough to expect to be accepted. However, such colleges consider many factors beyond just stats and hooks. So if you look at stats alone, a large portion of high stat applicants are often still rejected. Some specific numbers from a Harvard internal analysis and study using the lawsuit sample is below.

Academic Index is a stat based metric derived from a combination of GPA/rank and test scores. Academic Rating is the subjective rating given by admission officers who read the file. The worse academic ratings are very well correlated with not having high stats. There is an obviously a correlation between stats and admission, with some groups having a high admit rate, and not high stat applicants having ~0% admit rate unless hooked. However, it is not as obvious what are the primary drivers of this correlation. If high stat applicants have a high rate of admission, it’s not necessarily primarily because of their high stats. That high rate of admission may also largely relate to being more likely to have other non-stat criteria that colleges value. For example, high stat applicants are no doubt more likely to have excellent LORs, essays, impressive accomplishments outside of the classroom, etc. This makes it unreliable for a particular student to assume his chance of admission is similar to the overall admit rate for his stats or the overall admit rate for the college as whole.

Harvard Admit Rate by Academic Index, White Applicants
+1.62 SDs – 51% admit rate
+1.52 SDs – 43% admit rate
+1.42 SDs – 36% admit rate
+1.32 SDs – 31% admit rate
+1.22 SDs – 24% admit rate
…
+0 SDs – 4% admit rate

**Non-ALDC Admit Rate by Academic Rating/b
Academic Rating = 1: 68% admit rate
Academic Rating = 2: 10% admit rate
Academic Rating = 3: 2% admit rate
Academic Rating = 4: 0.0% admit rate
Academic Rating = 5: 0.0% admit rate

Hooked LDC Admit Rate by Academic Rating
Academic Rating = 1: 97% admit rate
Academic Rating = 2: 49% admit rate
Academic Rating = 3: 18% admit rate
Academic Rating = 4: 3.5% admit rate
Academic Rating = 5: 0.0% admit rate

Hooked Athlete Admit Rate by Academic Rating
Academic Rating = 1: 100% admit rate*
Academic Rating = 2: 96% admit rate
Academic Rating = 3: 87% admit rate
Academic Rating = 4: 80% admit rate
Academic Rating = 5: 50% admit rate*
*very small sample

122

@Data10 , please explain what the abbreviations are. I’ve been here a long time and I have no clue what SDs or ALDC is.

123

Great post data10. I believe that academic rating of 2 would correspond to perfect or near perfect gpa and scores. To get into the 1 category you need research, a significant academic prize/recognition at at least the state level or similar. If you are unhooked and ‘only’ bring straight As, high 1500s and corresponding strong recs and essays, you have a 10% chance - prob lower if you apply rd. Guess this also true for top 8-10 schools.

124

SD = Standard Deviation. +1.62 SDs would correspond to AI stat index in the top 5% of Harvard applicants. in a normal distribution. However, AI is not a good normal distribution, particular at the upper end with rails.

ALDC is grouping of hooked students, as defined in one of the lawsuit sample studies.
A = Athlete
L = Legacy
D = Dean or Director’s Special Interest List
C = Children of Faculty or Staff

125

@Lindagaf, my guess would be:

SD= standard deviation
ALDC= Athlete, legacy, development, child (e.g., of faculty or trustees)

126

Thanks @Data10 . You lost me at standard deviation, haha!

Here is a true story to lighten things up, and demonstrate my complete ineptitude with numbers:
As a bright-eyed and bushy-tailed freshman, I decided to take Statistics to meet my math requirement. Walked into my class on the first day, and within a couple of minutes, I was hopelessly confused. I sat through the hour and then immediately dropped the class.

Why? Well, in my ignorance, I thought that Statistics was just a collection of fun facts. I literally had no clue that it’s a whole discipline in itself.

Oh man, my kids laugh so hard whenever I tell this story. Particularly my D, who has an affinity for Stats and will be a TA in a Stats class next semester.

127

The class of 2023 reading procedures give the following guidelines for assigning academic ratings. Note that readers use additional criteria in assigning academic rating and may deviate from these guidelines somewhat.

Academic

  1. A potential major academic contributor; Summa potential. Genuine scholar; near perfect scores and grades (in most cases combined with unusual creativity and possible evidence of original scholarship, often substantiated by our faculty or other academic mentors.) Possible national or international level recognition in academic competitions.
  1. Magna potential. Excellent student with top grades and, a. SAT and SAT Subject tests: mid 700 scores and up b. 33+ ACT c. Possible local, regional or national level recognition in academic competitions
  2. Solid academic potential; Cum laude potential: Very good student with excellent grades and a. SAT and SAT Subject tests: mid-600 through low-700 scores b. 29 to 32 ACT
  3. Adequate preparation. Respectable grades and low-to mid-600 scores on SAT and subject tests or 26 to 29 ACT.
  4. Marginal potential. Modest grades and 500 scores on SAT and subject tests (25 and below ACT).

That 10% admit rate is an average. Some applicants with this academic rating have a very good chance of admission. Others have ~0% chance of admission. It depends on other aspects of the application besides the academic rating, including how they did on the non-academic areas. As I recall, the academic rating alone explained <10% of variance in admission decisions. Some of the rating categories that had similar or greater explanative contribution were ECs, personal, athletic (including with recruited athletes excluded), and LORs.

128

Indeed. But rule of thumb still holds. Unhooked and near perfect starts you at 10%. Then adjust up or down.

What will really move the needle? State/national recognition for something the institution values. If you’ve founded a million-member political movement or are a national-level violinist, your chances start to look very good (almost) regardless of academics. Harvard especially prides itself on filling its class with kids like this.

129

"are a national-level violinist, your chances start to look very good (almost) regardless of academics. "

This is a major fallacy.

The academic stats of the national level violinists that Harvard et al admit are very, very close to the academic stats of the rest of the class. There is barely a finger on the scale for extreme musical talent in terms of academics. It is a much better EC to be playing at that level than to have a nice tennis serve or be on your towns sailing team. But nobody’s chances are “very good (almost) regardless of academics” if music is the calling card. Especially since so much of the extraordinary musical talent is applying to Julliard and the other conservatories and skipping college altogether.

130

I know two kids in harvard orchestra. Both national level. Both super smart of course. One was near top of class but would probably have been 2-3 rating at harvard and would have had only a small chance in rd pool. One didnt get into state school and would probably have had almost no chance in rd pool.

Dont like to rely on specific examples to prove point but the kid who started gun control group a couple of years back got in with mid 1200s. You will find many of these super talented kids in bottom 25% of harvard class by admit stats and other ivies too. Again still super smart kids.

What do think gets your ap pulled from 10k other kids who have the same stats? They are looking for people who will likely become (or already are) super talented, super rich or super famous. They also want a community full of diverse talents , and to have a great orchestra, newpsaper etc (which they do). Harvard will almost always will give a hundred or two sat points to get a top violinist (gotta be good enough to shine in harvard orchestra of course). Conversely, they reject something like 90% of kid with 1600s. As you allude to they like kids who have both the most, but not many of those - even at harvard.

131

@Lindagaf Thank you for posting this and @Data10 thank you as usual for bringing the data to inform the discussion.

For me, it is really helpful to know that (caution: the following numbers are for illustration only) the 12% published admit is really 3% if you are unhooked, 23% if you are URM, 28% if you are a legacy, and 96% if you can run the 100 meters faster than all but 20 other high school kids. We all have limits in this process - a limited amount of time to spend visiting schools, a limited travel budget, and only one ED bullet. More knowledge helps make better choices.

132

“Certain individual HSs and types of HSs are tremendously overrepresented among students at HYPSM… selective private colleges.”

I’m interested that the automatic assumption is that those HSs are private/prep schools. as in my experience It also seems to apply to certain public high schools. We know plenty of kids who went to top public high schools in Silicon Valley and had vastly more success at prestigious east coast private colleges than kids at less well known HSs a few miles away. That applied even to kids who had been relatively less successful in the same elementary schools and had lower GPAs and SAT/AP scores in HS.

Some of that may be legacies and donor potential, some may be better counseling, but part also appears to be that those top high schools are well known and have a track record amongst the east coast colleges. It also seems to negatively impact the less popular HSs because the east coast colleges can easily fill their quota of middle class ORM Californian kids just from the schools they know.

For comparison the results at UCs appear far more comparable because all the HSs are known and have a track record there.

133

For Harvard private schools and public exam schools are disproportionately represented.

https://www.thecrimson.com/article/2013/12/13/making-harvard-feeder-schools/?page=1

One of the points made in this article is that the kids at most of Harvard’s feeder schools have already gone through a very competitive admissions process.

One of the private schools listed in this article as a feeder school sends an average of 10 kids a year to Harvard, 7 to Dartmouth, and so on. An average of over 1/3 of the class attends Ivies + Stanford, MIT, Chicago.

The HS for the middle class community (avg. household income of 85K) in which this schools is located sent no one to any of these school last year according to the school’s published profile, even though this HS is 50% larger than the private day school. This is a community that sends quite a few kids to the day school.

Another 1/3 of kids from the prep school went to places like Wellesley, Williams, WUSL, Georgetown, Duke, UVA and Middlebury.

The local high school had one acceptance each from Wellesley and Williams, which may or may not have been the same kid.

The question is not which school’s applicants have a better chance of acceptance, it’s how different the profiles of the applicants from both schools are or aren’t. The prep school has a lot of recruitable athletes, including some from the local town. If those kids had stayed at the LPS would they have had just as much success, or did the quality of coaching and academics make a difference? Would the average prep school kid have had better results as top of the heap at the LPS?

I guess about the only way to assess the influence of high school would be to compare 8th graders with matching stats and look at their college results 4 years later.

134

It’s primarily selective HSs with a high concentration of stellar students-- both selective privates and selective publics (magnets/exam schools). Distance from college is also important, as it influences both number of applications and rate of hooks. The link above lists the follow 7 schools with the largest representation at Harvard (in 2012). I believe they are ordered from most students in class to least. The top 6 are all highly selective high schools, with a high concentration of top students that are located in MA or a state that borders MA.

Lexington (#7) is the exception. Lexington is a non-selective public HS, yet they still have a high a concentration of excellent students with a school median SAT of ~1321. They are also located <10 miles from Harvard and in the 2nd wealthiest community in MA, so I’m guessing they have both a high rate of hooks and high rate of Harvard applications. A similar principle applies to wealthy non-selective publics located near other colleges, such as Gunn and Paly having a large representation among Stanford’s matriculating class.

Boston Latin – Selective public exam school in Boston
Andover – Selective private in MA
Stuyvesant – Most selective of the public exam schools in NYS
Noble & Greenough – Selective private in Boston
Exeter – Selective private in New Hampshire
Trinity – Selective private in NYC
Lexington – Non-selective public <10 miles from Harvard

135

The gun control group student became somewhat of a national spokesperson/celebrity of sorts. That was what moved the needle in that case. He had been rejected at UCLA and UCSB (?) the year prior and took a gap year.

136

The question is also how many students applied. The article you linked above mentions >= 138 Andover kids applied to Harvard that year, which was ~45% of the senior class. Most US high schools have <= 1 Harvard application. If 138 students apply at a high school with a median SAT of 1470 and a high rate of strong hooks, I’d expect that HS to have quite a few acceptances.

137

The schools that send the most kids to very selective schools are generally the ones who matriculate the most 9th graders who have been pre-screened to be successful selective school admits. They give them great preparation for sure but they are working with kids who have at least one, if not more, attributes that will appeal to the elite schools. It’s sort of like saying that people who attended US Olympic training camps are far more likely to win Olympic medals that runners who train with local clubs. Well yes! The first group got that opportunity by being exceptional runners before their Olympic camp training. They aren’t turning chaff into wheat!

You can’t believe how much time is spent on the BS admissions thread trying to explain to prospective applicants that just because these schools send a lot of kids th o HYP, they should not assume they will be one of them. They will get a great education but the "average excellent " kid who gets into HYP from BS usually has a
hook. Which may be how they got into BS as well.

138

I think asking whether students attending a selective high school are more successful in college admissions because either (a) they have more raw talent and were going to succeed no matter where they were, or (b) because of the better school preparation and GC’s and other connections is the wrong question.

To get a leg up, having the combination is tremendously helpful, much more so than either factor on its own. You need the raw talent, but also the molding of that talent and the steering from a knowledgeable GC.

The athletic analogy above is a good one. I know several athletes who were born with tremendous raw potential who never made it to the top level of their respective sport. I know several athletes who’ve been given Olympic level training who never made it to the top level either. 99% of the most successful athletes I know have both. If you take a mediocre athlete and put them in the OTC they are going to improve but never going to be world class. Likewise if you take a world class prospect and never give them proper coaching they will never have world class success. You need the combination of both.

for college admissions you don’t necessarily have to have the combination of both, but it is a tremendous advantage if you do.

139

Most high achieving HS students in the US do not apply to HYPSM…, so they have a 0% chance of being accepted. Part of the reason for not applying may be influence from the GC, teachers, peers, or similar HS characteristics. However, I think the more influential factor is it helps to be among the small minority of high achieving students/parent who applies to a highly selective HS. If you are the type of student who applies to a highly selective HS, you are probably also the type who applies to highly selective college. Selective HSs with a high rate of applications to selective colleges are likely to have a good number of acceptances. to selective colleges. Non-selective HSs with few applications to highly selective colleges are not expected to have many acceptances to highly selective colleges.

For example, there have been studies that did something similar to Sue22’s comment above. They compared various outcomes for students who just missed the cutoff for admission to the selective exam school that has highest matriculations to Harvard to students who just made the cutoff. Note that these are not randomly selected students with high scores. These are a special subset of high scoring students who applied to selective HSs. Multiple studies arrived at the same conclusion – few statistically significant differences in rate of admission to highly selective college, SAT score, PSAT score, rate of graduate degrees, or nearly any other evaluated metric.

https://scholar.harvard.edu/files/fryer/files/dobbie_fryer_shs_07_2013.pdf
The chance of attending highly selective colleges forms a steady line based on entrance exam score, with little apparent dependence on whether student made/missed the cutoff for Stuy, Bronx Science, or Brooklyn Tech.

Abstract summarizes, " However, exposure to these higher-achieving and more homogeneous peers has little impact on college enrollment, college graduation, or college quality."

https://economics.mit.edu/files/9518
Abstract summarizes, " Our estimates suggest that the marked changes in peer characteristics at exam school admissions cutoffs have little causal effect on test scores or college quality."

One special case is O’Byrant in the 2nd study. O’Byrant is the least selective of the exam schools, so if students miss the cutoff for O’Byrant, they may go to a non-selective Boston public school, which have a median SAT of only ~5th percentile – a huge gap in student quality and completely different school environment. Attending O’Byrant instead of what is likely a non-selective Boston public school had a statistically significant negative effect on chance of attending both a “selective college” and a “highly selective college”. Attending the other more selective Boston exam schools did not have a statistically significant impact, either positive or negative. However, attending O’Byrant did have a statistically significant positive impact on AP scores, which may relate to teaching quality and have implications on how well students are prepared for college… Again the more selective HSs did not have statistically significant impact on AP scores, either positive or negative.

However, if you compare students who apply to selective HSs to those who do not apply to highly selective HSs, then there are much more significant differences. For example, the study at https://www.brookings.edu/wp-content/uploads/2016/07/2013a_hoxby.pdf found that most high achieving low income kids do not apply to any selective colleges and instead favored local schools near home. However, high achieving low income kids who attended a selective magnet were an exception. All the selective magnet kids (within limited precision of summary table) were in the “achievement typical” group that applied to selective colleges. Part of this effect is influence from the GC, teachers, peers, at the school, etc. However, I don’t think this is the primary driver. Instead I think the type of students/parents who apply highly selective HSs are likely to apply to highly selective colleges, regardless of whether they make the cutoff for the selective HS or not.

140

Except, it’s not random, so some applicants might have 1 in 1000 odds and others 1 in 2. If the average student is only going to do 10 apps making one of them to what is for them a long shot 1 in 1000 or even a 1 in 50 school is a waste of an app that could have gone to a school where they have more realistic odds of admissions.