<p>@Matmaven : Except that your data doesn’t really refute my assertion. I said that going from 35 to 36 would be immaterial. in your data, 35 is in the same group as 34 & 33.</p>
<p>If there is a correlation between higher test scores and ECs, GPA, & essays (and there almost certainly is), then the data does not show that there is a difference between 36 & 35. Also note that there would be far more 34’s and 33’s than 35’s (and even less 36’s).</p>
<p>For instance, just as an example, say that the 36’s have an average of 20 admission-percentage-points (APP) just from their non-test score stuff. 35’s average 15 APP from non-test score stuff. 34’s average 11 APP from non-test score stuff. 33’s average 8 APP from non-test score stuff.</p>
<p>Brown assigns 4 APP to 36 and 35 test scores. 3 APP to 34 test scores and 2 APP to 33 test scores. </p>
<p>Finally, assume there are 4 times as many 33’s as 35’s and 2 times as many 34’s as 35’s (I believe that distribution is about right).</p>
<p>36’s would average 20+4=24 APP. The 35-33 group would average ((15+3)<em>1+(11+3)</em>2+(8+2)*4)/7 = 12.3 APP.</p>
<p>Now, I don’t believe that Brown goes off a strict formula like that (though I’m sure many schools use some formula for at least some candidates), but this is an example showing how an adcom could see a 36 and 35 ACT as the same yet admit twice as many 36’s as 35-33’s.</p>
<p>Interesting, Purple Titan, and it does seem clear that some schools care more about scores than others. I wonder, though, if some of the disparity comes from Wash U and Vandy deliberately playing the rankings. Wash U, at least, has been accused of “yield protection” by waitlisting or denying highly desirable students they see as unlikely to attend, so if JHU is going after those students and getting only a fraction of them, that can make a difference. In addition, if Wash U and Vandy are trying hard to raise their scores, that might mean pursuing some high scoring but otherwise not particularly desirable students simply to gain the statistical boost.</p>
<p>Alternatively, depending on how “Ivy-like achievements” are defined, regional differences between the student bodies of the three schools - particularly in the case of Vanderbilt – might explain the disparity without saying much of all about the correlation between student quality and scores. Maybe the kids from Tennessee just aren’t as likely to go into some of the high-powered professions that get you points in some of the categories evaluated.</p>
<p>I don’t envy people who have to interpret statistics for a living! </p>
<p>As I stated in my first post – I don’t see this as a “myth” that needs to be busted. You just said yourself, OP, that for a kid from a cattle ranch in Montana, a 1400 might be sufficient. Admissions ARE holistic at these colleges. High test scores do give the colleges something (bragging rights, mostly, but provable ones). So they are valued, maybe somewhat more than other factors in the process. I have always stated that I thought test scores are valued more highly than GPA at most schools no matter what the colleges say for this very reason (can brag about the freshman class test score range easier than the GPA). But there are factors that can overcome test scores that aren’t quite tippy top – mostly hooks (geography, athlete, legacy, URM, some great accomplishment).</p>
<p>I just don’t see any of what you are saying as even news to people who have been watching this process for a while. I don’t even see the point of this analysis, really.</p>
<p>I don’t see the point either.
And I think it’s totally disingenuous to look at a school where the admit rate for SAT’s at a certain level is (say) 4% and SAT’s one level up is (say) 8% and say … “Look, there’s a 200% increase in your probability of being admitted if you just get your SAT score up.” </p>
<p>The relevant numbers are that your 96% chance of being rejected now goes down to 92%, not that you have a “200% increase in getting in.” That’s how people mislead with statistics. Your statement is technically true, but it is off such a small base in the first place it’s irrelevant.</p>
<p>I would be crucified by my clients if I reported numbers the way you propose doing. It’s manipulative and misleading, even if factually true. </p>
<p>@Pizzagirl I don’t report the data, the schools do. And it is absolutely not disengenous to say there is a huge increase in the schools I mentioned from Tier 2 to Tier 1. What’s disengenous about it? I get that you don’t like it. You yourself said the data is factually correct. In your hypothetical, going from 4% to 8% does absolutely double your chance of acceptance. I don’t get why that seems almost offensive to people. If you look up the thread, you will clearly see I stated I don’t advocate kids chasing 1 more point on their ACT at the expense of other things. But the data is what the data is. It’s become clear to me, many just don’t like what the data clearly shows.</p>
<p>@intparent This may not be news to you, but the point I was trying to make was that this idea of a Hurdle score is, in fact, a myth. Based on what I have seen on these forums, that is in fact news to a lot of people. </p>
<p>I think the cutoff score is more about rejection - if you don’t meet that min score you better have a REALLY good hook if you apply.
It would make sense that your chances rise in proportion to your stats.</p>
<p>Fun with statistics. Matmaven- I also disagree with conclusions you have drawn. You chose to manipulate one statistic without holding other variables constant. Of course, you don’t have the other variable data either. Likewise an objection to comparing a 36 not to a 35 but to a group with points less scores. Is the threshold 35 or is it lower? Can’t tell.</p>
<p>Therefore- meaningless thread for garnering any useful information.</p>
[quote]
Brown had 197 applicants with 36 ACT scores. 48 were admitted (24.4%). Brown had 3,608 applicants with a 33-35 score. 450 were accepted (12.5%).
[quote]
</p>
<p>A single score breakdown would be more informative. Yes, to a certain extent, scores matter but is it 36 versus 35? 36 versus 34? 36 versus 33? </p>
<p>You are also assuming that the rest of the application mirrors the score. Many kids take the ACT over and over again. It’s likely that the original 29,30 can make it to 32,33 after several tries. Not so many can make it to 35, 36. So that 33 score may mask a less-than-stellar transcript or lackluster student-citizen (no ECs, no passion).</p>
<p>@wis75 Not sure what you are suggesting is meaningless. One question - why do 13 schools in this group choose to show Acceptance Rates by score tiers? I can assume it is because they think it is a very important factor in their admissions decisions. I don’t see the same schools reporting admit rates based on some of the other criteria you and others suggest are potentially also variable between the tiers and could therefore also be influencing the admit numbers - i.e val vs sal, good essays, etc.</p>
<p>I am assuming you and others are suggesting that a big portion of the difference between the Tier 1 and Tier 2 admit rates is not due to the increase in score, but is due to other increased merits of the applicants - better grades, better ECs, better essays etc. Given that the difference between a 35 and a 36 can come down to the matter of 2 or 3 correct answers, I find it just as illogical to conclude that an applicant getting those 2 or 3 questions right translates to a highly likely probability that they have better grades, ECs, and essays than the candidate who gets them wrong.</p>
<p>I continue to believe the data is quite meaningful but that many don’t want it to be. So be it. Have a nice day :)</p>
<p>Ugh, no, the probability of you having better ECs, GPA, etc. does not go up if you get a 36 rather than a 35 (they would be what they are) but the 36’s having on average better EC’s, GPA’s, etc. than the 35’s is highly likely.</p>
<p>Perhaps it depends on the specific high school and experience admissions committees have with that high school. Some scattergrams from some high schools for some colleges show significantly greater acceptances for high test scores regardless of GPA over a certain point (i.e., lots of green at the right side of the graph, a short vertical line, when Naviance plots GPA on the Y-axis and test scores on the X-axis). For some, it’s pretty clear that what counts is grades, that the top of the graph is a green horizontal line. And yes, the right top corner is typically more green than the middle of the scatter plot. That should be no surprise.</p>
<p>@purpletitan Highly likely based on the fact that they got 2 or 3 more questions right on one particular test? I find that mind boggling. I guess I should change the title of this thread to “College Admission: Everything is an Opinion” :)</p>
<p>Sure, they may have gotten in at Brown, etc. at higher rates. They likely got in everywhere at higher rates. Like other mere mortals, however, they can only attend one school. They are likely concentrated at the usual suspects, but they are more likely than many students to have a choice between elite schools. So I would expect them to get in at higher rates, because the college knows they’re less likely to yield. </p>
<p>It’s almost October. Our seniors are under pressure. Please, adults, do not try to make the pressure worse for the students who read the parents forum by obsessing over statistics which make no difference. There is no useful purpose served by making 100% of students worry that their test scores are not good enough. </p>
<p>100%–that’s the cumulative percentile for scores of 36, 35, and 34 on the ACT.</p>
<p>People shouldn’t obsess over test scores because they are statistically unlikely to get in regardless of score, and because going to Cornell rather than Yale is not a fate to be bemoaned. But if a student who doesn’t have an obvious hook scores a 1480 and wants to increase his odds at Yale, it would stil be advisable to retake. </p>
<p>" In your hypothetical, going from 4% to 8% does absolutely double your chance of acceptance. I don’t get why that seems almost offensive to people."</p>
<p>It’s not offensive. It’s just misleading. There is absolutely no difference in my opinion between a 1%, 4%, 8% 10% chance of acceptance, because they are all so slim that I should count on not getting it and being delightfully surprised if and when I do. </p>
<p>Do you make different decisions about carrying an umbrella if it’s 1%, 4%, 8%, 10% chance of rain? Or does your actual feeling about the certainty of your picnic coming off unscathed only change as it gets much higher?</p>
<p>If the forecaster said 4% chance of rain and then revised it to 8%, would you say, “oooooooh, it’s double the chance of rain now”?</p>
<p>Because that’s the way the common data questionnaire is set up. They are reporting information to conform to the way they are expected to report it. </p>
<p>No one is saying that test scores don’t play a part in admissions. The point is that the colleges don’t view them as a linear criteria for admission with the fine distinctions that students on CC think are important in chances threads. There is no material difference between a 33 ACT and a 35 ACT. There is a significant difference between a 29 ACT and a 34 ACT. </p>
<p>There is going to be a cutoff at which a student won’t be considered absent extraordinary circumstances. I’ve always felt it’s fairly easy to discern that rough cutoff by looking at the bottom end of the score ranges in CDS data, when it s made available. I figure that when admissions for a particular, lower score range is reported in the single digits, then the only students being admitted int that “range” are the ones with scores in the very top of that range. That is, if Ivy U. reports that it accepted 3% of students with SATS in the 500-590 range, I’m going to assume those students had scores of 590 and not 510. </p>
<p>But I’m not going to look at the 600-690 range and make the unwarranted assumption that the student with a score of 680 has a better chance of admission than the student with a score of 620. It is very possible that within-range, the scores are viewed the same by the college. It is also very possible that a college has its own internal formula that assigns a certain numerical rank to all scores within specified ranges – for example, maybe there is an admission index that awards 5 points for a score between 700-750 and 6 points for a score between 760-800. It’s quite possible that at some colleges the human beings making the admission decisions might never see actual scores - that is, they might only see the result of their own college’s assigned ranking or AI system – which could very will be a number drawn from a combination of GPA & test scores, rather than test scores in isolation. </p>
<p>You have 100 posts… not to put too fine a point on it, but you haven’t seen enough posts to know what common misperceptions do and don’t exist out here.</p>