Self-reporting is fine, but we’ve seen UChicago go from disdaining playing the rankings game to being one of the hardest USNews ranking gamers around. They already have deferred admissions and just added ED. The only other frontiers are accepting more transfers (and/or guaranteed transfers), sending kids with weaker stats abroad first (or delaying until after the fall), and taking a big chunk of the student body off the WL.
Then again, many schools already do some/all of the above, and I expect more to do so in the future.
Chicago is already taking a large number of students off the waitlist (or they did this year, anyway). It’s pretty clearly designed to manage yield, but not something to be up in arms about.
I don’t like the switch from EA to ED+EDII+EA, because ED programs generally favor wealthier applicants and will lock in some students who won’t be happy at Chicago despite a change of heart, but I’m glad there are changes like these to mitigate the effects on low-income students.
@runnersmom Last year at least, application fees were waived for any student who applied for financial aid. I saved $75 this way (though I spent the $75 and then some on CSS profile fees).
We just received email from UChicago admission personal that they require us to end official test scores either from the testing agency or from high school guidance counselor. This is contradicting to what is said in Uchicago website. I guess we will send the scores from testing agency
@concernedmom17
So true. No more soul. Just stupid Ivy wannabes.
If Chicago (or anyone else) really wanted to do some playing field leveling, they’d (i) get rid of super-scoring and/or (ii) require applicants to submit official scores from each test taken. Those two are the ways to cut down on the gaming and costs associated with standardized tests. Rice, Stanford, Penn, Yale, Georgetown do this.
Super-scoring doesn’t give any advantage to applicants of equal means (since the benefit is available to all). But super-scoring does benefit the school. Since it inflates the test scores of the enrolled class that the school can report to the ranking services.
Do you have any evidence that only certain demographic segments benefit from superscoring?
@bluebayou, it’s as obvious as it seems:
People who can afford the time, energy, and prep to take more tests would benefit from superscoring.
^^yeah, but since this IS college confidential, I’m just asking for some real, hard data to support the pov. Otherwise, supposition and speculation, even if considered ‘common knowledge’, gets a C in my course.
@bluebayou I think there’s a big difference between math/logic and “supposition and speculation.” Math/logic tells me that applicants “benefit” from superscoring only if they take multiple tests and have their highest score on each element of the test on different test dates, and there is a greater chance of that the more times an applicant takes the test. So the benefit, if any, for applicants goes to those who take the test most often. Probably more than twice: I think there is hard data that most test-takers improve both scores when they take the test a second time, so relatively few of those benefiting from superscoring will have taken the test “only” twice. And I think it’s completely fair to notice that it take both money and sophistication to start taking the tests early enough to take them three or more times. If everyone had an equal opportunity to do that, and behaved rationally, then superscoring would benefit no applicants at all; it would just raise the bar for everyone. So maybe it’s not actually helping any students, although I think we know that relatively few students take any standardized test more than once, much less three or more times. But if it is helping some students more than others, it can only be students with money and sophistication.
On the other side, colleges have a prisoner’s dilemma. If they use superscoring in their admission process, they can use superscoring in their Common Data Sets, which feed into NCES and USNWR. That makes their entering classes look smarter, and gives them a few ranking points. When a few colleges started doing it, everyone was doing it within a year or two, because the effect was so obvious. So at this point, superscoring probably isn’t actually benefiting one college over another, although it is probably exaggerating the difference between colleges that admit lots of people who take the tests three or more times compared to colleges whose students take the test only once, or at most twice. But any college that abandoned the practice unilaterally would have an immediate disadvantage relative to its main competition.
JHS – agree that if your peer competitors are inflating their reported numbers via super-scoring, it is tough to unilaterally disarm.
Seems like the most reasonable position on this is (i) to allow super-scoring (so as not to unilaterally disarm) but (ii) require all scores to be sent to the school for admissions purposes.
I’d guess that the policy requiring all scores to be sent would have a bigger impact than the policy of allowing unofficial scores to be submitted initially to admissions. Although both should help kids that don’t have unlimited resources to devote to polishing their standardized test scores.
Here’s what Rice says about that approach:
“Rice uses the highest scores from any sitting on the same version of the SAT in order to consider each applicant’s most positive test results. Recognizing that this policy could disadvantage those students who cannot afford repeated testing or expensive test prep coaching, we believe a comprehensive testing history provides us with the appropriate context required for making a fair judgment of what the test scores mean in a holistic admission process. Therefore, we require all applicants submitting the SAT to submit all scores to Rice.”
I’m not sure what I think about that. If I were an Admissions Director, I might say, “Rice is sending a message to people who take the test multiple times that they shouldn’t bother applying, and that’s not what they mean to say at all. Meanwhile, I am putting all sorts of resources into recruiting, admitting, and funding low-income students. I am being evaluated all the time on how well I am doing with that. So I am morally certain that low income kids as a group aren’t being disadvantaged; quite the opposite. Why should I do something to discourage a group of applicants who, in fact, are very attractive, since they are relatively affluent, smart, and care deeply about college? Sure, that group of obsessive test-takers may be disproportionately Asian, but if it gets too overwhelming we have other ways to deal with that!” (j/k)
All of this also starts from the premise that relatively small differences in standardized test scores are important in selective college admissions. But I think that’s not the case at all. I think test scores are one factor out of several that constitute an initial screen, and at the very high end they may combine with other factors to get an automatic admission for a very small group of applicants. Small differences in test scores may make a difference as to which of two or three athletes recruited to the same team will be admitted (or whether they are even eligible for admission – another reason I’m sure why superscoring is popular with admissions departments and isn’t going away anytime soon). But for most applicants, once they pass the initial screen they will be accepted or rejected for reasons that have nothing to do with their test scores, and their test scores may hardly be mentioned, even.
That may be true in some or most cases, but there are clearly situations in which a one point difference in test score will make a large difference in outcomes. For example, at Alabama the difference between a 32 and a 31 on the ACT is worth about $36,000 in tuition benefit. There was an earlier thread on Brown’s applications that showed that their acceptance rate nearly doubled for students with an ACT score of 36 compared to a 33 to 35 range.
I took the SAT I exam only once. A number of my classmates took the test 2 or 3 times. From everything I heard there was a very minimal change in score the second or third time. Some actually scored slightly less the second time. The only real major difference was one friend who was sick the first day and did substantially better the second time around. This, of course, is anecdotal and an insignificant data pool.
What is the average increase in taking the SAT a second time. I don’t think it would be very much unless the student was sick or unprepared the first time.
JHS;
With your closing sentence, you supported my question. To wit: “But if it i…”
I bolded the salient preposition.
Sure, that is a no-brainer. But where is the data to show what the score improvement is for 3 times test takers, 4 times, 5x? And then, what is the end result?
Money, most probably, but sophistication? Hmmmm. 
@bluebayou Superscoring is heads-I-win, tails-you-lose. On my nth test session, if I score higher than I ever have on any portion of the test, that becomes my official score for that portion of the test. If I score lower lower than I have in the past on any segment of the test, that score is disregarded. I’m sure the likelihood of a positive outcome declines the more times one takes the test, but since there is never any likelihood of a negative outcome, each additional test has an expected positive outcome, however small, until one’s superscored score is the maximum possible score.
No data showing the score improvement for 3x test takers, and 4x, and 5x changes that. My understanding is that, on a population basis, there is no/u expected improvement beyond the second test. But on an individual basis, each score is one specific point in a range of scores I might have gotten, given my skills, on that test, and if I take the test again I will get another score, and that score will likely be in the same range, lower, higher, or the same, but if it happens to be higher in the range,I get to claim the higher score.
As to sophistication – it takes a great deal of sophistication to start taking the test early enough to get three or four tests in before the music stops, if you wind up deciding you want to do that. Most kids, I believe, take the test once in the fall of their senior year in high school. By the time they find out how they did, as a practical matter they are stuck with that score. Or they take it at the end of their junior year, and maybe if they think they can do better they take it again in the fall. Someone who takes it three or four times has to have started in the fall of his junior year, and maybe in some cases registered for the next test before getting the scores for the prior one. That’s sophisticated in my book. Misguided and foolish, too, but sophisticated.
Re: merit scholarships. I meant to address that before. Sure, superscoring favors better-off, more sophisticated kids in looking for merit scholarships that are often heavily based on test scores. But isn’t that the point of merit scholarships, from the standpoint of the university? Attracting wealthy, committed students with discounts? It seems odd to criticize an element of a program designed to attract well-to-do students by saying that it systematically favors well-to-do students.
There are students w/ACT 22 when the midrange is 32 - 35???