Do Ivy League Schools Superscore your SAT Scores?

<p>I've wondered about the ACT superscoring issue for a long time... just another reason to choose the SAT. Maybe it has to do with the fact tht you can pick and choose which ACT scores to send. Still, you can't pick and choose individual sections, only whole tests.</p>

<p>how about a 2350 first sitting? should i retake that if i think i could get a 2360-2380? (although, i could also fall to 2320-2330 lol)</p>

<p>yep they superscore. in fact, many have said they dont even see the other lower scores.</p>

<p>I believe they superscore, BUT...</p>

<p>If you have two equally matched students, and they were to choose one out of the two; one had scores of 2350 and 2400, while the other had scores of 2200 and 2400, they would probably go with the former, as that person shows consistency in excelling.</p>

<p>fyi, brown does not superscore.</p>

<p>
[quote]
If you have two equally matched students, and they were to choose one out of the two; one had scores of 2350 and 2400, while the other had scores of 2200 and 2400, they would probably go with the former, as that person shows consistency in excelling.

[/quote]
</p>

<p>If a school used that reasoning, I'm pretty sure I wouldn't want to go there.
Any school that uses the SAT that much isn't attractive to me. If they're looking at the example that you provided, they should then look towards other facters.</p>

<p>If a school used that reasoning, I'm pretty sure I wouldn't want to go there.</p>

<p>I think there's no way of knowing such a detail about any school's adcom, there's no way to avoid that kind of school.</p>

<p>Better look at ur local CC then if you're afraid they'd use that reasoning MBP.</p>

<p>Anyone care to put up a list of colleges tht superscore?
BTW, I recently read an article tht many schools hv started to superscore ACT.
This a good thread by the way.</p>

<p>"I've wondered about the ACT superscoring issue for a long time... just another reason to choose the SAT."</p>

<p>I don't think so. Since superscoring is the practice for all SAT scorers, it offers no advantage - the whole curve against which any applicant is compared simply moves up. Indeed, it seems that superscoring puts pressure on testers to take the test multiple times which, if you have other things to do, could be a disadvantage.</p>

<p>
[quote]
If you have two equally matched students, and they were to choose one out of the two; one had scores of 2350 and 2400, while the other had scores of 2200 and 2400, they would probably go with the former, as that person shows consistency in excelling.

[/quote]
</p>

<p>I would disagree with this notion. The former appears to be a bit strange--what perfectionist goes and retakes a 2350? Consistent or not, why is this person wasting a Saturday morning taking a test he already did so well on? Doesn't s/he have something better to do? The latter I can understand; this person had definite room for improvement--maybe s/he had a bad day, was sick, etc, and decided to retake (with obvious improvement). </p>

<p>If they're equal, odds are that both of them are getting rejected. Colleges do not compare students side-by-side in making a decision.</p>

<p>I do not often take colleges at their word, but I do when they say they superscore.
The 2400 the first time out is obviously a special case. But to treat someone who got a one-time 2350 higher than someone who got a superscored 2370 is to assume that the one-time 2350 taker could have done better if he/she had just taken the test one more time. So why didn't he or she re-take? Can't assume what they never did. Should the admissions officer just pencil in something higher than 2350? I don't think it happens.</p>

<p>Interesting analysis, enderkin. That's why I don't engage in armchair psychoanalysis of admission committees--there's always more than one way to interpret what would be most "natural" for them. </p>

<p>To review the thread so far, I have posted links to Harvard's viewbook, which says that Harvard superscores the SAT I (and considers the best tests for students who submit more than the required number of SAT II tests). Yale makes a similar statement, linked above. Princeton also makes a similar statement. </p>

<p>Princeton</a> University | Standardized Tests </p>

<p>Princeton</a> University | Standardized Testing </p>

<p>This thread includes a reply indicating that Brown doesn't superscore--does anyone have a link for a statement on that issue? </p>

<p>The other colleges in the Ivy League are Columbia, Dartmouth, Penn, and Cornell. As I recall, Penn superscores, but the former website statement about that doesn't show up in Google anymore. What other links do we have to answer the OP's question?</p>

<p>From Columbia: (Columbia</a> University Office of Undergraduate Admissions - First Year Admission)

[quote]
The SAT Reasoning test consists of three sections, each graded on an 800-point scale; if you take the test more than once, you will be evaluated on the highest score you receive in any individual section.

[/quote]
</p>

<p>Can't find Brown--I think they superscore too, but it isn't obvious by their website.</p>

<p>Dartmouth:
(Standardized</a> Testing)

[quote]
If standardized testing is repeated, the Admissions Committee only considers your highest scores.

[/quote]
</p>

<p>Penn does not specify on their 'required testing' webpage: (Penn</a> Admissions: Required Tests)</p>

<p>Cornell does not specify anywhere that I looked =(</p>

<p>tokenadult wrote:</p>

<p>"But in the absence of published data on the issue, some of the suppositions about the effect size of "superscoring" are implausible to me. ... Someone would need detailed data that have so far not been published to prove me wrong--which I would be glad to see, even if it did prove me wrong."</p>

<p>So, if I understand your argument above, tokenadult, you are correct (other results are implausible) until someone proves you wrong, even if--as you admit--the data is not available for such proof. Oh, and your other argument for why superscoring doesn't change scores is because there was a thread on collegeconfidential debating the topic.</p>

<p>tokenadult wrote: "One reason I don't think I will be proved wrong when I say superscoring makes minimal difference in comparing students within one applicant pool and minimal difference when comparing enrolled classes among different colleges is that I have asked about the former issue: What does "superscoring" do for you?"</p>

<p>Okay, so I guess that means that I can argue that I am the most creative person in the entire world, and since there is no proof to the contrary, my argument is valid. Oh, and since it was discussed by approximately 15 internet posters makes it more valid. Interesting concept, I must say.</p>

<p>There IS proof to the contrary. You see Calcruzer, I am the most creative person in the entire world. </p>

<p>What now =P</p>

<p>tokenadult is right, I believe. Without specific policy in place that actually notes the use of superscoring in admissions decisions, there can not be any supposition that it exists. The burden of proof lies with the claimant then.</p>

<p>Well, based upon that theory, we can thank goodness we have proven that God doesn't exist. I thought for a moment there that it might still be in question.</p>

<p>First of all, my argument isn't related to who superscores and who doesn't. My argument is based upon whether superscoring increases the averages of the schools that superscore versus those that don't. Tokenadult's argument is that it seems implausible to him, therefore it must not matter--and my argument is how exact does not having proof that superscoring doesn't affect the average of scores prove that it doesn't? </p>

<p>Everyone here agrees that Collegeboard doesn't release these statistics--so any question of whether it affects the scores or not is pure conjecture. My argument--which I think makes much more sense--is that it generally does matter in total scores. For my own son the difference was 40 points. In my own informal pole, it appears to make a difference of 50-75 points on average (very informal and not scientific, I agree). Tokenadult says he thought it made a zero point difference, but provided no evidence for his argument other than personal opinion--and then made the statements I quoted in my last post (and that are on page 1 of this thread).</p>

<p>I don't know why people would be taking the SAT multiple times if they weren't expecting a better score through superscoring. And there are lots of examples on the "chances" board where they give the individual and superscored totals and the superscored totals are always better. I'm just asking about why someone thinks this wouldn't be the case--other than that they went to an "Ivy" that is. (Also, why did tokendult post Harvard's scores versus Berkeley's and not Cornell's, Penn's or Brown's).</p>

<p>A last point that should be noted is that people who benefit the most from superscoring are the ones most likely to be accepted to an Ivy. The higher your scores (thanks to superscoring), the more likely it is that you will be accepted to a top school that accepts superscores. Thus, for these people--that is, ones accepted to an Ivy school, my guess is that the superscore is much greater than a one-time sitting score.</p>

<p>Another participant on CC (Sam Lee?) made the interesting argument that if a state university has a lower score range than a private Ivy League university, one way to check the effect size of superscoring the SAT would be to look at the ACT score ranges for each college. If Harvard </p>

<p>College</a> Search - Harvard College - SAT®, AP®, CLEP® </p>

<p>has a higher ACT score range than, say, Michigan </p>

<p>College</a> Search - University of Michigan - U of M - SAT®, AP®, CLEP® </p>

<p>(and you can all see that it does), that suggests that the difference in SAT score ranges is mostly genuine and not just an artifact of superscoring. I don't deny for one minute that College A superscoring puts it at an advantage over College B not superscoring, but I have no objective reason to think that the effect size of that advantage is very large.</p>

<p>I read a description of one college's admissions procedure (sorry, don't remember which one) which indicated that each student's application was gone over initially by one staff person who "distilled" it down to a standard format for the rest of the committee to then consider. The recalculated GPA used by that school, a shorthand description of the relative rigor of high school courses taken, test scores, and a brief description of major EC's and other tip factors was all put on one page, with the student's essay(s) attached. (I can certainly see the value in such a process at schools getting tens of thousands of applications.) The document thus produced was what was actually reviewed by the committee - and it contained only the best test scores, not the others. In such a system the variations in multiple tests not only wouldn't be a factor, they'd be unknown to the decision makers. I don't know if that process is used by any or all ivy league schools in particular, though.</p>

<p>If we were to take a students' poll (sorry about the mistake in typing of the word "poll" in my last post), we might find out how many of these retook the test and improved their superscores significantly. The problem, of course, is this is not likely to be any more scientific than many other methods of comparing schools one against another by using their ACT scores--especially when one school has 74% of their students submitting the ACT and the other school only has 22% submitting the ACT.</p>

<p>Certainly, one would expect that the 22% submitting the ACT at Harvard did so because these were the higher of their SAT versus ACT scores. At Michigan, where the ACT is the normal test to be taken, it is likely that many of the students took the ACT only once or twice and never took the SAT at all. It is also highly likely that many of the students applying to Michigan wouldn't even bother taking the SAT if they knew that it was not going to be superscored.</p>

<p>Look, I'm not arguing that UC Berkeley or Univ of Michigan is likely to have a higher SAT composite score than Harvard if their tests were superscored--I stated already that I consider the difference to be 50 to 75 points on average--and the averages between these schools and Harvard is greater than that. However, this is not true for all the Ivies (one of these days tokenadult will make his argument using all the Ivies and not using the Ivy with the highest SAT scores). And even when one compares Harvard (as well as any other Ivy) with Berkeley and Michigan it bothers me when the 50 to 75 point differential is not considered--that's was my original comment--and the point. </p>

<p>My last point--state schools, unlike private colleges such as the Ivies--have a responsibility to their constituency that requires that they admit a large proportion of in-state students to their campuses. Most state schools admit approximately 15,000 students to the campus each year (with half of these attending)--a total that is usually about 4 times that of the average Ivy-league school. When one takes both of these limitations into account, there is no denying that the SAT scores will suffer accordingly. I'm guessing that top state schools such as UC Berkeley and UC Michigan, and possibly North Carolina and Virginia, could put together a class with SAT scores similar to most of the Ivies if not for such limitations.</p>