- EphBlog - http://ephblog.com -

More Admission Changes

As a follow up to Harvard’s decision to end Early Action, The New York Times noted:

Officials at many elite colleges and universities said yesterday [last week] that they would carefully consider how to respond to Harvard University’s decision to eliminate early admissions, though none were yet ready to follow Harvard’s lead.

“This will be a big topic of discussion on all these college campuses,” said Richard L. Nesbitt, director of admission at Williams College. “It is something we will consider. Will we change? I don’t know.”

At Williams, Brown, Dartmouth, Cornell and many other highly selective institutions, students who apply early must, in return for early acceptance, give an ironclad commitment to attend.

Nesbitt seems more open-minded to change than I would have expected. Perhaps he is being polite and/or cagey. There is almost no chance that Williams will make a change now. It has too much to lose. It also stands the potential of making some non-trivial gains. First, students who, in the past, would have applied early to and gotten accepted by Harvard/Princeton, will now be tempted by early decision at Williams. Isn’t the appeal of having the whole process done by December 15th as great now as it was 25 years ago? Second, those students will need to apply to other schools regular decision, including Williams. Many will be accepted and some will fall in love with Williams. They will end up at Williams because Harvard and Princeton no longer provide an early admissions option.

Will either effect be large? Tough to know. But if even 25 kids, who would have gone to H/P, end up at Williams instead, that would be important to the overall quality of the Williams student body.

Critics argue that this forces low-income students to commit before being able to compare financial aid offerings from that college and others.

What a crock! This debate is really quite dishonest, and it will be fun to puncture many of the misleading arguments made by these “critics.” (Also, if you’re a New York Times reporter looking for someone knowledgeable who thinks that these “critics” are full of crudola, call me! I am highly quotable.)

Anyway, every Harvard kid whose family makes less than $60,000 per year gets a free ride. The family pays nothing. Princeton is even more generous, with no loans in its financial aid package. If you can get into Harvard or Princeton, financial aid is just not that big of a deal.

But that’s not the misleading part. The key issue is:

The reason rich kids have an advantage in early admissions — whether early action or early decision — is because the institutions themselves (places like Harvard, Princeton and Williams) give them that advantage.

These colleges consciously lower the bar for early admissions. If they wanted, they could fairly easily make the standard for early admissions the same as for regular. (Yes, they don’t know what the full pool will look like until January, but the applicant pools at top schools are remarkably constant in make-up from year to year.) But H/P/W choose not to do so. They choose to accept some applicants early who they know would not make the cut in the full pool. They choose to “advantage the advantaged.” See The Early Admissions Game: Joining the Elite for full details.

Yet it is their choice. Instead of getting rid of early admissions, Harvard could achieve its stated goal by simply letting in fewer applicants, only those applicants who it is certain would be admitted in the spring. That way, no other applicants are disadvantaged by early admissions. It doesn’t affect their chances one way or the other.

But that isn’t what Harvard choose to do because this debate isn’t really (just) about helping poor applicants. There are much more ideological forces at work. Beware of colleges looking out for your best interest.

Comments Disabled (Open | Close)

Comments Disabled To "More Admission Changes"

#1 Comment By hwc On September 20, 2006 @ 1:54 pm

They choose to accept some applicants early who they know would not make the cut in the full pool.

I have seen absolutely nothing to suggest that this is the case at Williams, Swarthmore, Amherst and the like.

It is most certainly not the case at Harvard. EA admits there appear to be the cream of the crop.

The evidence thrown out to support the “sub-standard” theory is usually the higher acceptance rate in ED. IMO, this is a misreading of acceptance rates. The 5% to 10% of total applicants who apply early decision to places like Williams, Swarthmore, and Amherst are the most committed applicants in the entire pool. Well self-selected to match the school and eager to attend. Of course that self-selection and enthusiasm is evident in their application packages. For schools with high-stat applicant pools, those issues of fit and enthusiasm are what differentiates between a good applicant and an average applicant.

Do “below-average” applicants get accepted ED? Sure. Athletic tips. Lower-stat diversity students. “Grandpa gave a science center” admits. That has nothing to do with ED; those “below average” students are going to be accepted whenever they apply.

It is true that the accepted ED groups have slightly lower stats than the overall accepted class. Guess what? So does the final enrolled cohort. It’s easy to understand why. In the yield battles, a school’s top accepted students will tend to have more attractive options and, thus, tend to enroll at lower rates. The weaker accepted students will have fewer desireable options and will tend to enroll at higher rates. Common sense.

Since ED acceptances are, in essence, enrollees, the standard is quite simple. Is an ED applicant as good as, or better than, what we know we will end up with in May after all the yield decisions are finalized? From what I have seen, the answer is yes at schools like Williams.

The real cost of eliminating ED would be that it would make the overall system even less efficient in matching students who most want to attend a college and colleges who most want those students. The real flaw in the current system is that there is too little self-selection, leading to students sending as many as a dozen applications in almost shotgun fashion. The colleges are buried with no way to differentiate serious applicants from those who threw darts at the USNEWS rankings. Anything that increases efficient self-selection is a plus. Nothing is more efficient in that regard than binding early decision.

#2 Comment By David Kane On September 20, 2006 @ 2:13 pm

Much of what hwc says above is true enough. But there is an entire book, “The Early Admissions Game.” about the claim that colleges “choose to accept some applicants early who they know would not make the cut in the full pool.”

Avery et al (see link above) adjust for all the factors that HWC describe above. They conclude that, even for non-athletes, non-URM, non-donor applicants, applying early is worth around 150 SAT points. In other words, a “typical” applicant is much more likely to be admitted to Williams if she applies early than if she applies regular.

At some point, I may expand the Wikipedia entry on the book, but I don’t know anyone knowledgeable about the topic who denies that applying early, in and of itself while holding everything else constant, is a big advantage.

Which doesn’t mean that it is a bad thing! It does, at HWC points out, serve a useful sorting function. But my empirical claim is almost certainly correct.

#3 Comment By frank uible On September 20, 2006 @ 2:59 pm

Why should Nesbitt tip the College’s hand to its competitors? I may be unknowledgeable, but a 150 SAT point advantage seems very large to the point (no pun) of incredulity.

#4 Comment By hwc On September 20, 2006 @ 4:18 pm

They conclude that, even for non-athletes, non-URM, non-donor applicants, applying early is worth around 150 SAT points.

Their conclusion is complete and utter BS. It is predicated on the flawed assumption that the entire college application system can be distilled to a single number: the SAT scores.

Thus, a 1350 SAT applicant has an “X” chance of admission statistically and a 1500 has an “2X” chance of the admission. They, then, make the fantastic leap of logic that, since ED acceptance rates are 2 times higher than RD acceptance rates, ED must be “worth” 150 SAT points.

Of course, they completely ignore the more direct evidence that median SAT scores of the ED acceptances are essentially the same as the median SAT scores of the enrolled class as a whole.

If anyone cares to test this hypothesis, I invite you to find a nice solid unhooked affluent white applicant with 1300 SATs and help them apply to Williams Early Decision. Tilt it in their favor: tell them not to check the financial aid box. Academic 5? 6? I guarantee that ED won’t be “worth” 150 points on the SATs and that applicant will NOT get an acceptance letter! The odds are overwhelming that they won’t even get a polite deferral.

#5 Comment By hwc On September 20, 2006 @ 4:30 pm

It is true that, as a group, ED applicants are much more likely to be accepted to schools like Williams or Swarthmore. Roughly twice as likely.

It has nothing to do with SAT scores. It has everything to do with the fact that, of the thousands and thousands of applications, these are the 300 to 500 students who have come to the conclusion that Williams or Swarthmore is THE school for them. A conclusion that is typically reached after researching the college, visiting the campus, analyzing admissions data to confirm they they are plausible applicants, etc. These aren’t kids who are saying “I dunno, I’ll see where I get in, visit a few places, and make up my mind in April.” These are kids who KNOW they have identified their match.

Of course when you have a pool of the most carefully self-selected 5% to 10% of all applicants with the most enthusiastic applications, you are going to see higher acceptance rates.

#6 Comment By AC 199X On September 20, 2006 @ 5:01 pm

“Critics argue that this forces low-income students to commit before being able to compare financial aid offerings from that college and others.”

I don’t think that this part is intellectually dishonest per se, although I think the correct framing of the issue is slightly different: that lower income kids are fearful to even apply ED since they don’t want to commit without understanding precisely how much belt tightening must occur in order to make it work.

I don’t know what the experience was of others who needed financial aid, but my parents would not allow me to apply early to Amherst, even though it was clearly my first choice for precisely the reason that they did not want to have me committed without knowing what the financial aid package was. And this was in the early ’90s when it was less competitive to get into Amherst than today.

Harvard and Princeton are generous, I’m sure. But I’m not so sure that every other elite school is in position or prepared to be that generous year in and year out. I have no idea what the statistics say the end results turn out to be for low income kids, but I know the current ED structure can affect the decision on when to apply.

Also, I agree that Williams and Amherst will probably wait and see what unfolds with other competitor schools (Columbia, Stanford, Duke, etc.) before joining Harvard. Even Anthony Marx realizes that adding kids via early decision that would have gone to H/P/Y would be a complement to his larger goals, however you might view them.

#7 Comment By David Kane On September 20, 2006 @ 8:47 pm

There is a ton about this in The Early Admissions Game. Consider this tidbit from page 149.

In other words, applying Early Action improves an applicant’s chances of admission by about as much as a 100-point increase in SAT score. (To reiterate, this analysis excludes applicants who are minorities, alumni children, or recruited athletes.

Now, this in not the 150 point difference which I cited above. I think that the larger number is true for Early Decision but I can’t find the cite for it.

Again, there may be nothing wrong with this advantage. It may be worth it to get kids who really want to go to Williams. But the key is that applicants don’t really have to want to go to Williams to get this preference. They just have to apply early. There is no way for Williams to distinguish between the kind of early applicants that HWC likes (well-informed, actively choosing Williams for the right reasons) and those who are just applying to get an edge. (And most every applicant from an elite private or public high school gets accurate advice on this topic.)

Again, the empirical claim has nothing to do with why a student is applying early. The claim (p. 156) is that:

The advantage of applying early to Princeton, Stanford and Yale was especially strong for survey participants with SAT-1 scores from 1460 to 1500. These applicants are attractive but not truly exceptional at these schools. At each of these three colleges, early applicants had double the chance of admission of regular applicants with the same SAT-1 scores.

All the evidence in the book is consistent with the claim that Williams is similar. A non-URM, non-tip, non-donor with 1450 SATs is probably about twice as likely to get in early decision as she is regular decision. Advice your young friends accordingly.

#8 Comment By rory On September 20, 2006 @ 9:00 pm

David and hwc,

you should both be able to see that your points need not be mutually exclusive: a student could be an academically less stellar applicant and get in early–it might be because Williams has decided to admit students who show intense interest in Williams, like students who apply ED and seem particularly knowledgeable about the school and present themselves as a special fit.

That applying early is a benefit to students, all things otherwise equal, is not a bad. What is bad is when that benefit is disproportionately offered only to students who do not have financial concerns. Outside of a rare few schools, (Harvard and Princeton being almost alone) that is true of ED and EA across the country. Further, as AC 199X points out astutely, one of the problems of ED/EA is the perception (right or wrong) that it is a way for the wealthy white kids to get a leg up. Low income students are told and/or believe that they should not apply ED and colleges cannot reach all of them. As such, ED/EA advantages do go to the rich and generally white disproportionately. Harvard and Princeton and Delaware have decided they would best get rid of that problem (and get a PR boost and simplify things for themselves, etc.) by getting rid of EA/ED.

#9 Comment By hwc On September 20, 2006 @ 11:25 pm

In other words, applying Early Action improves an applicant’s chances of admission by about as much as a 100-point increase in SAT score.

The flaw in his kind of statistical argument (it’s also used extensively by the anti-affirmative action crowd) is attributing causal effect to small differences in SAT scores. A 100 point difference in SATs means nothing in a vacuum. You can’t look at admissions in such simplistic terms: “Oh, you have a 1450 SAT, you won’t get in. You have a 1550 SAT, you will.”

You can look at SATs as a descriptor of a group of accepted students. But, there is no validity whatsoever in trying to use small differences in SAT scores as a predictor of admissions odds for any one student.

But, be that as it may, we continue to ignore the fact that there is no difference in statistics between ED admits and the overall enrolled class. Here’s an excerpt from an article written by long time Swarthmore, Stanford, and Princeton admission dean, Fred Hagadon:

In Princeton’s experience, at least, there has been no significant difference in the academic credentials of those admitted ED and those admitted regular decision (RD). For the eight years we had ED classes while I was dean at Princeton (the Classes of ’00 through ’07), the mean SAT scores for ED and RD admits were identical for five of those years. In one year, the mean math SAT score was 10 points higher for the ED admits, and in two years, the mean verbal SAT score was 10 points lower for ED admits. So much for the mythical “100-point” ED advantage that Mr. Karabel proposes as fact.

Note, neither Swarthmore nor Stanford had early decision or early action programs during his tenure. I have read Swarthmore Admissions Dean state the same thing: that the RD and ED cohorts have essentially identical SAT scores and overall academic qualifications. Amherst’s and Williams’ Deans have stated the same thing. I have never come across any evidence that they are all lying.

David writes:

But the key is that applicants don’t really have to want to go to Williams to get this preference. They just have to apply early. There is no way for Williams to distinguish between the kind of early applicants that HWC likes (well-informed, actively choosing Williams for the right reasons) and those who are just applying to get an edge.

What do you mean that applicants don’t really have to want to go to Williams to apply Early Decision? They have to agree to drop their applications to every other college in America and attend Williams if accepted. Sounds to me like they probably want to go to Williams. Dropping that ED app in the mail is the whole enchilada when it comes to making the college decision. That’s it. What else would you propose? A blood oath?

5823 people applied to Williams last year. Fewer than 10% (535) applied early decision. Why is it difficult to believe that these 535 wanted to attend Williams? This is not a box on the application that students just check willy-nilly for giggles. They are smart kids. They know exactly what the checkmark means. As a parent who has been through the process, I can almost guarantee that every single one of them pauses and thinks about it when they make the checkmark.

#10 Comment By current eph On September 21, 2006 @ 2:32 am

I’d encourage you to email admissions about this if you are curious. They will tell you that they use the same system of evaluating ED applicants as they do RD applicants.

#11 Comment By David On September 21, 2006 @ 8:24 am

current eph is correct that the admissions office uses the same “system” in evaluating ED and RD applicants, but they do not use the same standards. Try to get them to say that on the record. They do imply (an applicant I interviewed last year told me this) that there is no particular advantage to applying ED, but you won’t get Nesbitt to clearly say that because it is not true.

Rory is kind to look for agreement between hwc and me. I think that we do agree that ED serves a useful purpose, that Williams should keep the program and that, at least in the short term, Williams will keep it. But HWC and I disagree about the empirical claim made by the authors of The Early Admissions Game. Let me try and state it another way.

Take 100 seniors who are applying to Williams this fall without any hooks. Assume that they fall in the category of strong-but-not-overwhelming applicants — academic rank 2s with sold extra-curriculars and recommendations.

Now, randomly assign 50 of them to RD and 50 to ED. (This is a hypothetical! I realize that those who apply ED in the real world will, on average, know more about Williams, be more interested in attending, be less concerned about financial aid and so on.) My claim is that the admissions rate for the 50 applying ED will be about twice as high as that for those applying RD.

All the evidence presented in The Early Admissions Game is consistent with this claim. I am not saying that this is fair or desirable or the way that the world should work. But this is the way that the world does, today, work. Advise your applicant friends accordingly.

#12 Comment By hwc On September 21, 2006 @ 2:12 pm

Now, randomly assign 50 of them to RD and 50 to ED. (This is a hypothetical! I realize that those who apply ED in the real world will, on average, know more about Williams, be more interested in attending, be less concerned about financial aid and so on.)

David, I think you’ve just proved my point. As we all know, the applicants are NOT randomly assigned and, therefore, the claim of a 100-point SAT advantage is only valid in a hypothetically random world.

Look, I know that there are huge advantages to Early Decision for a solid applicant. Not the least of which is the advantage of 100% focus on writing the application to a school (that the kid actually knows something about) instead of dashing it off as #8 out of a baker’s dozen that have to be sent on New Year’s Eve. It’s a simple, but often overlooked, concept: better app = better odds.

I assure you that my daughter had more ammunition for a quality “Why Swarthmore?” essay (after visiting twice, meeting professors, and so forth) than she would have had for her “Why Safety School?” essay to #8 on her list. That one was going to be pretty damn generic. How many people applying to their #8 school can rattle off specific names of people they’ve met on campus and pull from those meetings specific reasons for excitment about attending? You know…something a little more than “great academics, a pretty campus, and high grad school placement rates.”

The school was her first choice because specific, unique aspects of the place matched her interests and experiences. The flip side of that is obvious: her interests and experiences were a good match for what the school values. That’s why she was accepted. I am quite confident that exactly the same mechanism applies for the vast majority of Early Decision applicants to Williams.

Why would it be beneficial to remove the one mechanism that allows students and colleges that truly are a great fit to identify each other? And, to do so before the avalanche of 12 applications per kid buries everyone.

It’s not the SAT scores. The median SAT score of the entire applicant pool at Amherst, Swat, and Williams is comfortably over 1400. I saw applicants waitlisted at Swarthmore this year with “stats” that would knock your socks off. Valedictorians, 1600 SATs. But, they didn’t really give a flip about attending the school (except as a “safety” if their Ivy apps bombed). I’m sure their applications reflected that lack of interest.

Advise your applicant friends accordingly.

As you probably know, I do advise students serious about a first choice liberal arts college to apply Early Decision if they can. I gave “virtual” advice to 26 Swarthmore ED1 applicants on College Confidential this year. Read and commented on a few of their essays (that’s it’s own story – advising a full-financial aid applicant from Shanghai on a college essay!). 22 of the 26 were accepted ED and a 23rd was ultimately accepted in the RD round. This group was hardly the lily-white, silver spoon cohort imagined by opponents of early decision. It ran the gamut: full-pay, full-need, white, Latino, Asian, Af-Am, internationals, athletes, musicians, you name it. The group was probably a bit more affluent than the overall enrolled class. But, I don’t think it was much less ethnically diverse. For example, it was 9% African-American compared to 6% for the fall 2005 freshman class and 12% for this year’s freshman class. It was more Asian American than the overall class. More international. Not sure about the exact Latino percentage. The three consistent themes were strong high school transcripts, something “interesting” beyond academics, and a high degree of enthusiasm for the school. Virtually all of them had visited and overnighted on campus — many during the minority recruiting event at the school’s expense.

I do add a qualifier to my ED recommendations. Don’t waste time applying Early Decision to a school unless you are at least a solid, plausible candidate for admission. ED doesn’t magically get an acceptance letter for kids who wouldn’t otherwise get accepted. ED does take some of the elements of chance out the equation and improve the odds for a good, solid applicant.