Elite college admissions: What matters and by how much

Because subjectively-graded aspects of applications (essays, extracurriculars, recommendations, etc.) cannot easily be back-checked by outside observers, there it little that outside observers can use to try to reverse engineer how a super-selective college does its admissions beyond looking at academic stats and other non-subjectively-graded criteria (legacy, ethnicity, etc.). So super-selective college admissions will generally remain mysterious and opaque to outside observers, including students, counselors (at least at typical non-feeder high schools), parents, etc…

It’s important to keep in mind that the needs of a college or university change each year, too. For example, the orchestra conductor may be lacking a tuba and trumpet players this year, so applicants who bring those skills to the table have a really good shot of being admitted. The following year, the art department may tell admissions that they just graduated a huge groups of students, so they request more art majors. I would imagine for the LACs this is especially important since their numbers are that much smaller. Yes, these applicants need to have sufficient stats, but this may be what admits them over other applicants - even those who may have higher stats. (Sorry if this has already been mentioned - I’ve been reading this thread over several days and may have missed a post or two.)

I can’t take seriously that someone doesn’t like the food, so rejects a candidate.

For a tippy top, the stakes for the individual applicant are higher and the risks, when he thinks this is all formula. (That is, if he’s serious about college X.) The onus is on the candidate to know what a good whole presentation is,for his college targets. That doesn’t come from putting data into a program. That falls far short of the level of thinking. Eg, some huge number of matriculants at tippy tops were tops in their hs class. That says little about how being top 1 or 2% will work for you, if you miss the rest of what they want, the actual attributes you show, the level of understanding you present, and more. And in the end, your whole app/supp is really what they judge you on.

Some misunderstand my comment about knowing what a college actually wants to see, after stats and standing. Some are willing to look at CC, (often times, a group of seekers, not know-ers) and the few self-edited lines of chance-me or results posts and draw conclusions. That’s incomplete. It’s the bare bones, not the full picture that came across in those kids’ apps and supps. Two kids can have identical records and still present differently, by their own hands, their own choices, their own assumptions- or savvy. That matters. That can make the difference between an admit or a reject.

Is your chance higher, if you start with the stats? Of course. But the goal isn’t really just to get past first cut. It’s to make it to the finish line. Try to learn what that takes, not just what gets you to the door. If you can’t take your exploration past stats, don’t then throw up your arms and call it a crapshoot.

Yes, there is a lot of consensus on CC about what it takes. But don’t assume every poster knows. Do the leg work to understand what Yale looks for, versus Stanford, Cornell, or Duke, Chicago, how environments, college offerings and strengths, etc, differ, and what they seek to continue. what they say, not just what CC says. Go beyond stats and a few accolades. It’s not that hard.

In the end, you can’t control whether they got some handful of great apps from North Dakota or they need a tuba player. But you can rise to the challenge of trying to understand more.

@lookingforward , of course, if an applicant figures out what that college is looking for and delivers it, then yes, that’s the best way to get into a top choice elite.

The article is called “Dirty Secrets of Admissions Officers”. I don’t want the mods to nab me, so I won’t post a link. And I was wrong, it was a restaurant in Buffalo where the AO got food poisoning, and rejected the apps from
Buffalo the next day.

I’m not talking about reading something that got someone a byline. Or a book that attracts lots of attention and makes the author big bucks. If you go back to the article, looks like both “Current admissions officer, state university in the Northeast” and “Current admissions officer, Ivy League university” say they got food poisoning in Buffalo.

Neither of those two is named.

I’m talking about knowing why you want this school (beyond prestige or “I want a top education and you’re a top school”) and why they should want you, beyond stats. Something more than categories in the CDS or that you were hot stuff in your high school. Or what some stripped down summary on a CC results thread shows.

Math mom, you make an interesting point, that schools with new or expanded areas of study will need students interested in those programs. I believe both Dartmouth and Brown are in the process of upgrading their engineering programs so maybe they will be looking for more engineering students in the next year or two? And give preference to those planning to study engineering?

“What matters and by how much”

Strong essays and recommendation letters could give you a chance of about 80%.

We are talking about capable and motivated applicants with 2250 SAT average.

Where’d you get that 80% idea? Most applicants past first cut are capable and motivated and still have a slim chance. And other factors apply. Then final decisions include institutional needs- geo diversity, balance in gender and majors, etc.

“Where’d you get that 80% idea?”

Saw “chance of winning” in the news a lot lately and my number is a wild guess which could vary widely down to 65%.

Comparing to teacher recommendations and quality of essays the other factors on your list are minor, I guess.

For example (hypothetically), if a teacher says a student is her best in 30 years of teaching or her recommendation moves the student to tears, then the student’s chance could be 100%. The teacher certainly doesn’t talk about grades and test scores. The 2400 SAT and 18 APs all 5s kid has to move over. Or maybe the two are one.

The book mentioned in post 9 is interesting, with the obvious caveats that data on here are self reported and do not seem at all to be from a representative sample of applicants.

Nonetheless, the author’s finding that grades matter more than test scores (above ~2200 SAT or 33 ACT he finds little difference in admit rates, but there’s a big difference between 4.0 and 3.8) is consistent with, for example, the data the University of California campuses post on applicants and admits - very difficult to get in with grades much below 4.0 unweighted, much more “forgiveness” for relatively low test scores.

It’s harder to see this pattern at the very top private schools, as the overall admission rates are much lower than at UC, but it makes sense (the best predictor of future performance is generally past performance) and shows up in the College Confidential data.

Certainly lots of other factors at play in admissions, but grades versus test scores stood out as a pretty clear finding.

Is it not reasonable that the data is a little skewed since it is presumably easier for legacies, recruited athletes, URMs, etc to take less rigorous courses in high school and have a good GPA, yet they often would have a lower ACT/SAT? If this is the case, I am not sure someone without a hook should interpret it to mean grades are the most important.

I agree @bluewater2015 that grades trump test scores. I am sure my kid didn’t get accepted to at least two colleges simply because her grades could have been a little higher. Her test scores were great, above 2200. If the grades had been a little stronger, I think the WL at Carleton would have been an admit.

And yet, I sometimes wonder why that is the case, when it’s universally acknowledged that there can be so much variation in grading standards between high schools. At what point does a college admit a kid in which the school has a proven record of low-performing AP scores, but a high GPA, over a kid with perhaps great AP scores and not quite as high grades? Yet most senior posters agree that AP scores don’t really count for much. Which brings it back to standardized test results. Would love to have that one explained.

I hear you, Lindagaf! At some of the privates around here, most kids score 2 or 3 on their AP exams and seem to never have much homework, whereas most of the AP scores at our public are 4’s and 5’s and our top students scarcely sleep. I don’t see it hurts the private school kids in elite admissions, though.

111 I've heard several speeches from admissions deans on the grades vs. test scores question.

They all agree that grades are more important than test scores. But they say that it is much harder to discern and interpret grades. They don’t care so much about high school GPAs, since varying grading policies and grade inflation affect the GPA number greatly. What they REALLY want to know is where the kid ranks in his class by grades.

But since most high schools don’t do ranks anymore, the adcoms and the HS guidance office go through a bunch of shadow ranking information (typically shown on the HS profile) in order to back into a rough guess-timate of class ranking. Once you have the rough class rank, it is pretty easy to compare and deal with top 15% at Swanky Suburban High School vs. top 5% at Tough Inner City High School vs. top 40% at Elite Boarding High School.

Standardized test scores, of course, don’t require the apples-to-oranges exercise. So while the schools don’t want to rely upon them as much as grades, they certainly are much easier to rely upon.

Harvard, after all, invented the SAT test in the first place as a tool that could be used to identify the smart kid at Bohonk High School as compared to just taking one more kid off the list provided by the headmaster at Elite Boarding High School.

For this reason, there appears to be a strong tendency on these forums to assess students’ chances to get into colleges and colleges’ admission selectivity by standardized test scores more than any other factor – even though it can result in inaccurate assessment in many cases, due to it being a lesser factor in many colleges’ admission criteria compared to other factors.

“if a teacher says a student is her best in 30 years of teaching or her recommendation moves the student to tears, then the student’s chance could be 100%.”

Say what? The decision to admit is NOT made by teachers. You seem to have some limited concept of what chances really are. These kids with perfect scores, grades, rigor and even very nice LoRs still have a job to do: make a solid self-presentation in the app and supps. So many are so focused on stats (and their own greatness in their one hs) that they miss the chance to present what the college does want to see. Face it, this is not a chance to transfer to a better high school. It’s the leap to college. And the colleges do the picking.

Too many think ‘hierarchically.’ Ie, best this, best that. And then the kid sleepwalks through the app/supp and everyone gets bent out of shape that he was rejected. Etc.

“still have a job to do: make a solid self-presentation in the app and supps.”

That’s the 1st part of a 2-part equation I mentioned, with LoRs being the 2nd part. The successful kids are very good at doing their part, started years prior. Test scores and numbers of APs are not in the equation. However, these students may be among the top few % in their schools.

“So many are so focused on stats (and their own greatness in their one hs) that they miss the chance to present what the college does want to see.”

Agree with @lookingforward 100%. Too bad to those kids and their parents that for the few dozens of unrepresentative RD applicants to 1 college I eyeballed, hyper high scores reduced their chances. And maybe for years they’ve focused on the wrong things that they lost themselves or reduced themselves to numbers, thus hard to present.

I can’t help but think that schools that offer too many APs are hurting the students.

“All models are wrong but some are useful.” What I present could be just wrong period.

Different colleges emphasize different criteria and have widely varying degrees of selectivity. In theory a student could have a near 100% chance of admission at some selective colleges based on just stats, such a public colleges that have near guaranteed admission on stats. However, selective colleges with a holistic focus often emphasize criteria beyond what you listed. You can get an idea about what is important for a particular school by looking at a combination of what they say is important on their website and CDS, what is known about their process through things like FERPA request, and a history of admissions decisions.

For example, Stanford’s CDS rates 9 categories as very important, so not especially helpful about relative importance, but it implies much more than the above is important. This is consistent with their website, which goes in to more specific information about what they are looking for, emphasizing more subjective areas than are captured in the CDS. They say academic excellence is the most important criteria, which emphasizes HS transcript. However, immediately after that they mention looking for “intellectual vitality”, which includes things like a genuine interest in expanding intellectual horizons, initiative in which you seek opportunities, curiosity, and enthusiasm, among other things. According to FERPA requests for admissions files, Stanford reviewers rate applicants on a scale of 1 to (5?) in six categories, which are largely consistent with the earlier information – tests, high school record, support (includes LORs), non-academic, intellectual vitality + self-presentation, and overall impression. They also have additional stats available in the admissions file like a recalculated 6 semester GPA, rating of HS, a yes/no diversity box, etc. Beyond stats, the reviewers write a summary of why they think the applicant should be accepted/rejected, which may be as long as 2 pages and goes in to depth about subjective reasons. I previously mentioned when I looked at CC decision thread posters, decisions seemed to largely follow out of classroom activities, which probably relates to CC thread posters generally excelling in the in academic areas, but having more variable quality in non-academic areas.

Being in upstate NY, the HS I attended’s Naviance doesn’t have a statistically significant sample of Stanford applicants, so I’ll use Cornell in this example instead. It shows the majority of Cornell applicants with a 96%+ GPA were admitted at all SAT score ranges (very few 96%+ GPA applicants were below ~1200/1600), but the admit rate increased as SAT score increased among this 96%+ group. Admit rate dropped dramatically for below 95% GPA at all SAT ranges, but some were admitted with as low as 90%. However, some applicants were rejected with as high as 99% GPA and as high as 1600 SAT, so they are obviously looking at more than stats. A similar method could be applied to other colleges. Different individual HSs will likely show different patterns.

If we are talking about elite private colleges, I expect you’ll often find the opposite relationship with legacies. For example, Harvard’s freshman survey indicates legacies average significantly higher SAT scores than non-legacies, yet legacies average slightly lower GPAs than non-legacies. This is not surprising considering that legacies are more likely to come from privileged backgrounds and attend high quality HSs. If you look at surveys of admissions officers, such as the NACAP one, the criteria rated as highest average importance in admissions decisions consistently relates to transcript. However, some individual colleges are exceptions. There is not a universal admissions decision method.

who knows. My DD was admitted to 3 out of 4 ivies she applied to plus a bunch of highly selective LACs. She was not the val or sal; in fact probably just barely in the top 10% of her class (but her public HS did not give a lot of weight to APs, etc.) She took (by far) more APs than anyone else in her class and had stellar test scores. So saying GPA matters more than test scores… did not apply in her case. I think there is no simple answer. I imagine her recs were very good.

“She took (by far) more APs than anyone else in her class and had stellar test scores.”

“but her public HS did not give a lot of weight to APs, etc.”

What matters most is the QUALITY of the transcript. That is different than “GPA” or “grades” and obviously has to account for rigor.

Taking “far more” AP classes than anyone else in your entire high school is a really strong measure of quality. And not getting much GPA credit for all those AP classes is an obvious reason why colleges have to look beyond simple numerical GPAs in assessing the strength of the transcript.

On a quality weighted basis, your kid was probably #1 at her school. Which the Ivy League ad coms easily figured out. Helped, no doubt, by the supporting evidence of stellar test scores and great recs. It isn’t really very complicated.