The Big Squeeze

Michael Winerip provides a plethora of anecdotal evidence for the conclusion that it's becoming much more difficult to get into an elite college. Roughly speaking, he interviews Harvard applicants, they seem much more qualified to him than he was when he successfully applied back in the day, and none of them ever get in. And, of course, these scare stories are based on data. Everybody knows that "several Ivies, including Harvard, rejected a record number of applicants this year." The trick, as Kevin Carey has helpfully pointed out is that this isn't really true. For every applicant, there are some number of applications and the number of applications-per-student has been growing rapidly:

When the number of applications grows faster than the number of applicants, it creates a false sense that admission standards are getting tighter. Imagine 20 students, each of whom applies to five schools and gets into two. Now imagine if the same students each applied to ten schools and got into two. The outcome for the students is the same: two acceptance letters. But the schools report lower admission rates, and the odds of admission seem worse.

In particular, Carey notes that the number of acceptances at the Ivy League increased 10.6 percent between 2002-2006, which was faster than the rate of increase in the number of high school graduates. It was, however, slower than the 28.6 percent rate of increase in the number of college applications. And it's easy to see why students are mailing off more applications -- compared to other things prosperous families do to help their kids get an edge in the admissions process, just mailing more applications is simple and relatively cheap. From a social point of view, however, an escalating arms race in which everyone is applying to dozens of colleges won't be a very happy end point.

Comments

"Nobody goes there anymore; it's too crowded."

-- Yogi Berra

Posted by Tyrone Slothrop | April 30, 2007 8:46 AM

It is indeed harder to get into an elite school than it used to be. I should know - I work in admissions at one.

Posted by Clark | April 30, 2007 8:48 AM

His SAT score was probably 100 points too low — though it was identical to the SAT score that got me in 35 years ago.

Does Winerip actually not know the SATs were recentered?

Posted by J. Ellenberg | April 30, 2007 8:55 AM

The very convenient www.CommonApp.org lets you submit applications to 20 of the most popular colleges without re-entering data.

Posted by Steve Sailer | April 30, 2007 9:29 AM

Carey's article is a classic example of the misuse of statistics. He's engaged in a journalistic three card monte, in which he intentionally confuses two different categories in order to lead the reader to a false conclusion.

Carey says that the number of spaces in the countries "elite colleges" - meaning the 60-odd schools that Barron's ranks as "most competitive" - has increased faster than the number of high school grads. This means, he says, that it is no more difficult to get into the Ivy League, and Harvard in particular, than it used to be. The Ivy League has eight schools, not 60, so the data about 60 schools tells us nothing about Ivy League admissions. By mixing "elite college" and "Ivy League," Carey misleads the reader. I have to think this is intentional - unless Carey is a complete statistical illiterate.

The truth is that Carey's data proves precisely the opposite of what he's arguing. The reason the number of spaces on the "most competitive" list is increasing is that the list is open-ended. Schools that used to be on the "highly competitive" list have become much more competitive in recent years. Emory, the College of New Jersey, Lafayette and many others are now "most competitive." Students who used to think of schools like these as "safe schools" are now unable to get into them. From the heights of Harvard to the great middle of the big state schools, the shadow baby boom has made it tougher to get into college at every level.

So please, Matt, read a little more carefully before you spread this sort of manure around. Often when something seems to be a startlingly contrarian insight, it's contrarian because it just ain't so.

Posted by Bloix | April 30, 2007 9:30 AM

The number of freshmen at the top Ivy League schools has remained esssentially unchanged since coeducation in 1970, while the U.S. population has grown 40% or so, the cost of transportation has vastly decreased (increasing applications from outside the Northeast), and international students are taking up many more slots. Either the application process is more competitive or students are getting dumber.

Posted by arthur | April 30, 2007 9:35 AM

Who cares? Guttenberg made the Ivies irrelevant. You didn't get into Yale or Princeton? Go to the nearest library.

Posted by Jeffrey Davis | April 30, 2007 10:12 AM

This is silly. Let's say you're in one basketball game and you take 10 3-point shots and 2 go in; in another basketball game, you take 20 3-point shots and 2 go in. The end result is the same, 6 points. But that doesn't mean it wasn't harder to score three-pointers in the second game, even if people were taking more shots. The odds of making a three-pointer in the second game is actually lower.

Also, there's a causation issue here. It's possible that the rise in applications per applicant has caused a decrease in acceptance rates. But isn't it just as likely that the increasing difficulty of acceptance has spurred applicants to apply to more places, leading to a vicious cycle.

Posted by jbl | April 30, 2007 10:18 AM

Whether spaces at these colleges are rising faster than high school graduations is not the point -- the Ivies, Potted Ivies, and flagship state universities are increasingly meritocratic.

I see the point about total applications rising, in large part to protect against declining admission rates. The pertinent question is not are these kids necessarily getting into one particular school, but are they getting into a "top" school, at all?

Posted by David | April 30, 2007 10:49 AM

Some factors in the appearance of difficulty getting into college.
++++
More applications due to:
1. The Common App
2. The Common App being online
4. More colleges accepting the Common App
(e.g. Northwestern this year)
and
4. Colleges accepting credit card payment for admissions rather than requiring a check.
++++
Winerip is probably Caucasian. As an alumnus of Harvard he should be aware that legacy status is said to add about 100 points to an SAT score. (some boost anyway).
++++
Colleges have a tough time. If you depend upon donations and if those come from Alumni isn't that some sort of "merit"?

When California law required admission to state universities be based on "merit" the representation of Asians rocketed vs. their percentage in the population. All other groups are now "under represented" especially those of Hispanic and African heritage.

++++
Lake Wobegone syndrome in America. Too many schools graduating kids with high grades. Too many think they are above average.

Posted by JRip | April 30, 2007 11:11 AM

A top student can always get into a top school. It's just that the top school might be a flagship public university instead of an Ivy.

In fact, when you compare the career success of students with comparable resumes who go to top private vs. top public schools, the only difference is that the kids that go to their in-state public universities graduate with a lot less debt.

Posted by Dave | April 30, 2007 11:11 AM

Matt, the increasing number of applications per applicant is exactly why it is harder to get into Harvard now than it used to be. The pool of competition is wider and the number of spots is the same -- whether there's more applicants in aggregate to all schools or simply more applications doesn't make a difference when looking at a consistently high-yield school like Harvard.

Posted by right | April 30, 2007 11:14 AM

"these scare stories are based on data. Everybody knows that "several Ivies, including Harvard, rejected a record number of applicants this year." The trick, as Kevin Carey has helpfully pointed out is that this isn't really true"

He is reporting the 10.6 growth in acceptances. If more kids are applying to more colleges the admissions offices have to deal with the fact that fewer of those who they accept will actually come. This is called the Yield. While top colleges like Yale or Harvard may have a yield of 67% others have 35%.

It might be better to know the increase in the size of the various freshman classes.

If you look at the reported composition of freshman classes at top schools there are more international students than 35 years ago, more people of non-Caucasian ethnicity, more people from other parts of America. So if you,like Mr. Winerip, are a Caucasian living in the northeast your chances are reduced.

and wait... Were the Ivies even admitting women yet when he applied?

Posted by nit picker | April 30, 2007 12:00 PM

When people argue that those who got into elite colleges in the past couldn't get in today, my view is that most (certainly not all) of the same people who got in 10 or 20 or 30 years ago would still get in today. They would just do a lot more crap in high school to burnish their applications.

Posted by old timer | April 30, 2007 12:20 PM

Well, geez... maybe those rejected kids can now go to a school that actually will teach them something. Like, you know, who Kit Carson was. Just for example.

Posted by Collin | April 30, 2007 12:33 PM

It seems to me that the real problem, regardless of how the details of the current discussion playout, is a severe lack of schools that are _perceived_ as elite by the general public. The quality of both the students and the education at Ivies is arguably comparable to the quality of students at other comparably elite but less prestigious institutions.

When my brother was applying to colleges in 2002 he attended a talk by the Dean of Admissions at Harvard, who told a story that went something like this: "If the entire incoming freshman class were being brought to campus on one airplane, and that airplane crashed, and the admissions office had to go to the second string people, and their plane crashed, and so only the third-string students actually made it to campus, the faculty wouldn't know the difference."

Personally, I transferred from Brandeis to Dartmouth half way through my undergraduate and found that the quality of the faculty and the education were basically the same on average (some things better at one place and some at the other).

But in both of these cases, the students who don't end up with the Ivy League degree are at a competetive disadvantage in the job market compared to the ones who got the Ivy degree, and the only substantive difference is the brand name they get to tag their resume with.

Posted by Galen | April 30, 2007 12:44 PM

Good. Our college should be hard to get into. There are way too many kids in college who have no business being there.

The problem is not that there are not enough kids in college, but that class status, not merit is a determining factor in attending college.

Posted by Max | April 30, 2007 1:12 PM

Is there anything more boring than the annual NYTimes college acceptances story? Yes -- two things, in fact. The annual NYTimes college application story and the most intensely boring NYTimes private kindergarten application story.

Posted by ly of brooklyn | April 30, 2007 1:32 PM

As an Ivy League grad and interview for my school (Columbia) let me offer some facts:

1)The number of space at the school is up, but in my era (the l970s) it was male only, so that's deceptive. The school had about 750 students in my class; now it has about 1600 per. Not much increase.

2)It's absurdly competitive. Interviewers are repeatedly instructed to give the top number recommendation to only the best possible students, but the reality is, in my experience, the school only considers the top number. This means that you can be a superb student at a great school, with excellent grades, speak several languages, and edit the school paper, and still not have a real chance because you aren't a charismatic leader who runs a local charity on the side.

3)American students are now competing with much higher percentages of students from abroad than in the past, many of whom attend high school in the U.S. specifically to have a chance to apply to the Ivy League. Some of these students don't really make the transition, but some are very impressive.

It's painful to me to see how many excellent students never have a chance. When my school announces that it has rejected more than 90% of students applying, that is accurate reflection of the situation.

Posted by Kit Stolz | April 30, 2007 3:27 PM

To respond to a few of the posts above:

Bloix: read the original article carefully, and you'll see that I provide separate statistics for the "most competitive" schools and the Ivy League. Moreover, I didn't let the "most competitive" list vary; it's the exact same group of schools for each year of the comparison.

Jbl: The basketball analogy isn't a good one. The number of shots a player or team can take in a basketball game is strictly limited by the time of the game. Every missed shot comes at a cost. The number of applications you can submit is only limited by the number of institutions that exist, the cost per application, and the time it takes to fill them out. Since the per-application cost is relatively low (particularly when compared to the value of an acceptance), and the time it takes to fill out an application has been reduced by the Common Application, students have every incentive to apply more often, which is what they're doing. A rejected application costs the student very little.

Kit Stol: You're right that the story looks different if you compare how things are today to the 1970s. But I'm only looking at 2002 - 2006.

nit picker: A good point, but the number of enrollments at the Ivy League schools has actually increased at the almost exactly the same rate as admissions. At the "most competitive" schools it's lower, so there may be some of what you're talking about occuring there. Hard to tell.

old timer: You're right.

Posted by Kevin Carey | April 30, 2007 4:06 PM

Colleges have a tough time. If you depend upon donations and if those come from Alumni isn't that some sort of "merit"?

If a college acceptance is a financial investment, rather than an educational one? Sure.

Posted by cg | April 30, 2007 4:11 PM

There are a lot of things going on here that I think get people confused about the issue. Rejection rates are an obviously stupid way to look at the issue. But there are perfectly reasonable ways to examine competitiveness of the slots available.

First, the simple ratio of slots to potential college students. There are a lot more potential college students now. You have women, minorities (many asians) and foreign students taking up spots once allocated to white men. My understanding is that the number of freshmen at the Ivies has not grown fast enough to make up for this factor. So yes, there are fewer spots.

Second, how good are the students? Here, I believe, is just a lot of hype. Sure their resumes look a lot better and they talk a good game... but couldn't this all just be improved gamesmanship? As a matter of fact, how does this story possibly square with the complaint of declining educational quality in the US? My opinion is that although its harder to put together a top flight resume, the academic standards have gone down considerably. Professors complain that incoming students are lacking basic expected skills (like writing) and I am none too impressed with modern expectations in mathematics.

Finally, admissions officers always talk up the quality of their incoming classes. Why do we bother listening to these people? They're just trying to hype themselves and the schools they represent.

Posted by mpowell | April 30, 2007 4:12 PM

Kevin Carey, this is why I called what you did ‘three card monte.” In that game, the cards are on the table. But the player is a master of misdirection.

In your second paragraph you conflate “elite” with “Ivy League.” You say, in respect of that paragraph, that the statement that it’s become harder to get into a given school – by clear implication an “Ivy League” school – is ‘not true.” But your evidence for it being “not true” relates to “elite” colleges, not “Ivy League” colleges. Your definition of "elite" colleges includes dozens of schools that a Harvard grad like Matt wouldn't remotely consider "elite." (Grove City College, anyone?) So that’s your misdirection. You carry it out cleverly throughout your article, alternating effortlessly between "elite," "Ivy," and "Harvard," never quite lying but always inviting the reader to conflate and confuse.

And what are these "elite" schools? You now say that you used the same set of schools to make your 2002-2006 comparison. You could not have used the Barron’s list of “most competitive” schools from 2002 and 2006, as you said you did, because the Barron’s list has changed. So what did you use?

But most importantly, you give the percentage increase in the number of “fat envelopes” sent out by your list of “elite” schools as evidence that the number of slots in those schools has increased. This is nonsense. As you note – indeed it is your main point - the number of applications per applicant has been increasing dramatically. Many of these extra applications are going to “reach” schools, but many are going to “safe” schools. Therefore, any given student is going gain acceptance at more schools than a similar student a few years ago. Obviously, schools want to accept the best applicants, but they understand that among them the "yield" - the percentage who will choose to attend - will be much lower than the yield among lower qualified students. To fill up their entering classes with the most qualified students, they must send out more fat envelopes simply to keep the same number of students in each entering class. Even at Harvard, the admissions office understands that some percentage of those admitted will choose Princeton or Williams, and this percentage is higher today than a few years ago. This is a phenomenon that has nothing to do with the difficulty in getting in to any particular school, and is entirely a function of the increased number of applications. Even though a school may be sending out more acceptances, the credentials of the average student who chooses to attend - that is, actually fills a slot to the exculsion of a less-qualified applicant - may well be higher than they were a few years ago.

The key questions are very simple: has your pool of "elite" schools increased the number of slots over the past 4 years? You don't say, presumably because you don't know. Because I can't tell what universe of schools you are using, I don't know either, and I can't even try to find out. Without that key piece of data, it is impossible to say anything at all about whether competition for your self-defined pool of "elite" schools has increased.

A reasonable article would have explained that the increase in applications per applicant is masking the actual trend. Then you would have explored what is really going on. But instead you drew a clever but unwarranted conclusion from partial evidence. You don’t have the data to prove your contrarian position, but you assert it anyway as if it were fact. It’s a criminal waste of your talents and an imposition on your readers’ time.

Posted by Bloix | April 30, 2007 5:28 PM

Broix: Okay, I'll try one more time.

I used the most recent list of "most competitive"colleges from Barron's (the update the list every two years, I believe a new one is due soon). While I can't speak for Matt, the list is entirely comprised of prestigious colleges and universities that we all know and recognized. Grove City College, needless to say, isn't among them. I can't include a link to the list because Barron's doesn't make it available on the Web for free, it's one of the main things they're selling. But I can tell you that the Barron's categories are frequently used in the academic literature as a proxy for institutional selectivity.

I used the same list for each year of comparison because, as you note, doing otherwise would render the comparison invalid. The trends of applications rising faster than applicants holds for every group I analyzed, which is why the use of "elite" applies in all cases.

The data are from the U.S. Department of Education, National Center for Education Statistics, Integrated Postsecondary Education Data System (nces.ed.gov/ipeds). It's all publicly available for download.

To your question: "has your pool of "elite" schools increased the number of slots over the past 4 years?" The answer is, according to the IPEDS data, yes. At the Ivy League schools, the number of enrollments increased at the same rate as the number of acceptances, which takes the "yield" issue off the table. At the larger pool of "most selective" institutions, which includes the Ivy League, the number of enrollments has increased faster than applicants, but slightly slower than acceptances. Whether the difference is the result of institutions making allowances for changes in yield is unknown.

But the larger point is unchanged--in the short term, there's little if any evidence to suggest that the number of qualified applicants trying to get admitted to an elite school is rising faster than the number of slots in those schools. David hits the nail on the head when he says, "The pertinent question is not are these kids necessarily getting into one particular school, but are they getting into a "top" school, at all?"

Posted by Kevin Carey | April 30, 2007 6:15 PM

Kevin Carey, what a bizzare response. I wasn't talking about cost or limitation at all. But fine, let's say I'm playing totally cost-free and unlimited computer solitaire. Let's imagine: in 1987, I play 10 games and win 4, and then, in 2007, I play 20 games and win 4. I have achieved the same outcome in 2007--4 won games--but it was objectively harder to do so.

It wouldn't be fair to say:

The outcome for the solitaire players is the same: four winning games. But the computers report lower winning rates, and the odds of winning seem worse.

The odds of winning don't just seem worse; they are actually worse, 20% vs. 40%. It's harder to win. You have to play more times to win as much. In the case of the colleges, it's harder to get in a door. You have to try more doors to find one that you can get in.

(I take your point that this is probably not because the doors are smaller but because there are bigger crowds at each one. Still, it doesn't follow that the difficulty is illusory.)

I think I can see your abstract point. You're saying college capacity eventually meets applicant demand. That's fine, but you're trying to overprove when you say that "it's not getting a whole lot more difficult" . . . because students today manage to achieve equivalent results with greater persistence. That dog won't hunt.

Posted by jbl | April 30, 2007 6:21 PM

Kevin Carey - Grove City College claims be on the Barron's "65 most competitive colleges" list as of 2006. http://www.gcc.edu/Rankings_1.php Who is lying, them or you?

As for the rest, this is a new article you're writing, not a defense of what you published. Maybe you're right, maybe you're not, but what you originally wrote made no sense at all, IPEDS is impenetrable, and life is too short.

Posted by Bloix | April 30, 2007 6:34 PM

Bloix: That list is from 2003, I used the more recent list, which they're not on. The 2006 reference is to Barron's "Best Buys." Again, read carefully.

jbl: The point of applying to college isn't to getting as many acceptance letters as possible. If it were, your game analogies--basketball, solitaire, or otherwise--would make sense, as would the media's use of institutional admission rate.

The point of applying to college is getting into the kind of college that you want to attend. Whether you get into one, four, or ten of them doesn't matter, because you can only enroll in one. Now, if you've got your heart set on one single college--if you will accept no substitutes for Princeton, for example--then it's possible that the trend of more people applying for the same slots and the vagaries of the admissions process might lessen your odds of getting into that one.

But it's not going to lessen the odds of getting into a college like Princeton, because--again--there are only so many qualified applicants, and that number is not increasing at any thing like the rate of increase in the number of applications.

It does, as you note, take more work. If everyone else is sending out twice as many applications, then you have to too, just to keep up. But that's the point--you've kept up, and your odds of getting into the kind of college you want in the end stayed the same.

Posted by Kevin Carey | April 30, 2007 7:28 PM

“In 2004, Barron’s “Profiles of American Colleges” ranked Grove City College as one of the “65 Most Competitive Colleges and Universities in the Nation.” http://www.gcc.edu/Why_Recruit_at_GCC.php

So Grove City was listed in 2003 and again in 2004 - two years in the middle of your 2002-2006 frame. Yet you claimed that "needless to say" it wasn't on the list. Did you know that Grove City was on the list for those two years but decide that you would stick in that "needless to say" as yet another feint? Or did you not even look at the list for the intervening years before telling us that Grove City couldn't possibly be on it?

Posted by Bloix | April 30, 2007 7:52 PM

to: Matthew Y & Kevin Carey

M.Y. wrote:
"The trick, as Kevin Carey has helpfully pointed out is that this isn't really true."
and Kevin wrote:
"the number of enrollments at the Ivy League schools has actually increased at the almost exactly the same rate as admissions."

Matthew is wrong as his answer relates to Harvard.

I just reviewed the Harvard University Fact Books from 1995 through 2006.
On Oct 15, 1995 the freshman class was 1624 and the undergraduate mix was 56% men.
By Oct 15 2005 the freshman class was 1655 and the undergraduate mix was 51% men. The 2005 freshman class itself was 827 men and 828 women.

The population of the USA was estimated at 262 million as of January 1, 1995 and is about 300 million now. In addition, the count of children of the baby boomers of college entry age is peaking from about last year through the fall 2009 entry.

While there are other factors to consider, the pool of high school seniors has grown and the size of the freshman class at Harvard has not kept pace. Indeed, there are 50 to 75 fewer men now than 10-12 years ago.

Net: More applicants coming from a greater pool = tougher to get in.

Posted by nit picker | April 30, 2007 8:34 PM

Thanks, nit-picker.

Kevin Carey, seriously? So the trend you're talking about would be the historical change in percentage enrollment . . . of some undefined group of qualified applicants . . . in any one of some nebulous and fluctuating list of elite schools? And we're meant to be surprised that your trend and the trend in the raw acceptance rates could arguably be different in light of new trends in application strategies?

Talk about small beer.

Posted by jbl | April 30, 2007 8:58 PM

tracking applicant virtual

Posted by applicant on tracking | May 6, 2007 2:50 PM