<p>Too bad we don't get to see the median stats for the apps to compare with, say, the 25th percentile scores for the admits. In other words, how many of all those apps are real hope shots?</p>
<p>This the first year that Holy Cross has not required standardized test scores (SAT, SATII, or ACT). That probably has something to do with the jump in applications.</p>
<p>One could suggest the online common application is the culprit, but even Chicago with its "Uncommon Application" saw a record number of applicants with an increase of 6.9% over last year (still it received only 9,567 applicants, self-selection is still alive and well, at least to a degree).</p>
<p>wow on Reed & Holy Cross.....why so big? Does no SAT requirement really attract that many new app's?</p>
<p>Sure. What's the first question an average kid asks about admissions? Are my SATs good enough to get in?</p>
<p>By not requiring SATs, a whole bunch of lower scoring kids are given new hope.</p>
<p>It's a win/win for the college. Apps go through the roof, acceptance rate goes "down", and they get to cherry pick the best SAT scorers in the freshman class (those who submitted scores) when reporting their median SATs to USNEWS.</p>
<p>Reed requires the SAT or ACT-( although they don't require SAT IIs but they strongly suggest it- D submitted two tests) but even though they dont supply stats to US NEWs- the magazine did profile them in their latest college edition as well as the president Colin Diver and Paul Marthers the admission director have been getting press in magazines and papers like the Atlantic and Washington Post
Also Reed alum of one semester Steve Jobs, mentioned Reed favorably in his commencement address at Stanford-
So while it wasn't perhaps on many students radar 5 years ago, it certainly is now.
( plus while it looks like it could be a school in New England- the weather is much milder)</p>
<p>I think one can give the kids some credit. Reed has not watered down its core, inflated its grading, or dropped SAT/ACT requirements. It is attractive to those kids looking for real academic challenge, and emphasizes that in its materials and during visits. That market segment may be growing as well, in both applicants and number of apps.</p>
<p>I don't think cherry-pick is the term you mean. That suggests that Holy Cross would selectively submit whichever SAT scores it wanted to (the best ones). I think what you mean is that the school will benefit because the kind of applicants who submit their SATs (despite not needing to) may tend to be the kids who knew or suspected they'd have great scores. </p>
<p>It's a matter of self-selection, not cherry-picking.</p>
<p>As for the topic at hand, when Michigan releases its numbers, they will show a healthy increase.</p>
<p>Well, in some cases, it is "cherry-picking". Compare for example, the SAT medians reported on the Common Data Set by SAT-optional Middlebury (those 50% of enrolled students who officially designated SATs be used for admissions) to the median SATs for the entire group of enrolled students in the class profile. There is quite a difference between 1380-1500 and 1230-1400 for the same freshman class. </p>
<p>According to Middlebury's Common Data Set filing, their 1380 25th percentile SAT scores are higher than Williams, Amherst, Pomona, Swarthmore, Columbia, Rice, Duke, Chicago, Brown, Penn, and Northwestern. Of course, those other schools are reporting substantially all of their freshmen while Midd is only reporting the 50% of their freshmen who designated their SAT scores as their "official" test scores rather than their SATII or AP or ACT scores.</p>
<p>It may be self-selected "cherry-picking", but only the ripest fruit is picked in the reporting process, nevertheless.</p>
<p>"I don't think cherry-pick is the term you mean."</p>
<p>Well, does creative reportive or selective use of statistics represent a better description of what is happening at certain schools? There is very little doubt in my mind that US News SHOULD dump all the schools that are SAT/ACT optional in a separate group. However, that would require a lot more from USNews: integrity, an attribute that seems quite foreign to their rankings. </p>
<p>I-Dad correctly points the finger towards Middlebury. It is quite obvious that the school has spend the time and resource to position itself in the rankings, more than probably after a surprising tumble down. Schools that are playing games with the simplest of statistics should not be given the benefit of the doubt.</p>
<p>This is what I had a few weeks ago:</p>
<p>School.. ..2009.. 2010.. Change
Harvard 22796 22719 -0.34%
Yale..... 19451 20903 7.46%
Princeton 16516 17478 5.82%
Brown.. 16900 18250 7.99%
Columbia 18216 19730 8.31%
Darmouth 12615 14000 10.98%
Penn .. 18749 20300 8.27%
Cornell 24452 28012 14.56%
Total 149695 161392 7.81%</p>
<p>Duke... 18089 19282 7.12%
MIT.... 10439 11231 7.59%
Stanford 20194 22223 10.05%</p>
<p>Interesteddad has made the point about the Middlebury SATs before and is plainly right. It was perfectly clear that there is no way that Middlebury's 25% number could be higher that AWSP, just to compare it to LACs. Obviously, including only those enrolled students who designated their SAT scores as the standardized test scores they were relying on would have its greatest effect on the 25% number. You can't make your SATs be higher than they are; but if they are low, you just don't use them; you use SATII or ACT or AP scores, all of which Middlebury accepts instead of SAT Is. </p>
<p>However, Middlebury seems to have cleaned up its act. The 25/75 number cited by Interesteddad comes from Middlebury's CDS for the Class of 2008. For the class of 2009, the CDS reports the following: 78% of students submitting SAT; CR 630-745; M 650-730. So (using the USNews approach of adding CR and M, which isn't quite right) that translates to 1280 - 1475. This is still a bit weird as you can't have a score of 745. But it is a truer picture than 1380-1500.</p>
<p>sidebar on Middlebury....those numbers are all over the board:</p>
<p>class profile for class of '09, "ALL submited SAT & ACT scores reported, even if a different test type was used for evaluation"
SAT V 620-710
SAT M 610-690
ttl SAT 1230-1400
ACT 29-32</p>
<p>CDS class of '09
SAT V 630-745
SAT M 650-730
ttl SAT 1280-1475; 78% reporting
ACT 27-32; 35% reporting</p>
<p>CDS class of '08
SAT V 690-750
SAT M 690-750
ttl SAT 1380-1500; 50% reporting
ACT 28-32; 28% reporting</p>
<p>The '08 numbers look suspiciously even, plus, if they used the same "cherry picking" CDS methodology, I'd at least think that the SAT scores would be similar between '08 & '09, but they are not, especially the 25% being 100 SAT points different. Math is almost double the effect than Verbal between the '09 profile & CDS....guess I'd say that SAT math sub-performance may be a bigger cause to not submit scores. Also, its interesting to see that while the CDS SAT numbers are higher than the profile, the 25th percentile ACT is 2 points lower, implying that the ACT is more opted for admissions submittals by the lower-scoring students than the SAT.</p>
<p>I did a quick look at Bates & Bowdoin, both SAT optional schools, & they keep their public story straight, that is, they don't report the class profile ALL scores like Midd did, but only repeat what's on the CDS. At least Midd does provide some more data, but it sure looks like creative manipulation to me.</p>
<p>I think the moral of the story is that you have to pay close attention to the % of students for whom SATs are reported. When that % closes in on 100%, you can take the numbers at face value.</p>
<p>As the % declines at SAT optional schools, you have to make some educated guesses and make some rough adjustments.</p>
<p>Middlebury's higher scores, when they appear, can also be attributed to a higher income applicant group.</p>
<p>Enuf said. ;)</p>
<p>The irony for Middlebury is that the gamesmanship may not have helped their rankings. The 7.5% of the USNEWS ranking attributable to SAT scores is offset by the 5% attributed to underperformance compared to "predicted grad rate" (a prediction made on the basis of high SAT scores).</p>
<p>I don't think overstating SAT scores is smart business for colleges. It scares off potential applicants who are seriously trying to identify schools where they can get accepted. Those are the kind of applicants you want -- not the "just for kicks" applicants.</p>
<p>For example, suppose you have identified Williams as your "reach" school. Now, you are looking to find similar schools that are more in the match category. Smart applicants aren't looking to fill their dance cards with reaches. If the published SAT numbers erroneously make Middlebury look as selective as Williams, why bother? But, if Williams is a reach and Middlebury's median SATS are 75 points lower, then Middlebury might be a perfect match school. If I'm Middlebury, I want to snag as many cross-applicants with Williams as I possibly can. I'll take all of the 900 Williams waitlist kids as I can get my hands on.</p>
<p>Who establishes the rules for the Common Data Set? Would it fix a big part of the problem if there was a report on the number of total enrolled students (which is already there, of course), with the SAT scores (perhaps combined with an ACT equivalent) of the student 25% and 75% down from the top?</p>
<p>Seperate question: do schools publish seperately their 25 and 75 percentile data for admitted students and for enrolled students? Should they?</p>
<p>eulen, I can't answer your questions, but the CDS as it stands now, by my reading, seems to suffice in its intentions of getting the whole story out there. The instructions seem to be liberally interpretted by the SAT optional colleges. The CDS section C9 test reporting reads:</p>
<p>"Percent and number of first-time, first-year (freshman) students enrolled in fall 2005 who submitted
national standardized (SAT/ACT) test scores. Include information for ALL enrolled, degree-seeking,
first-time, first-year (freshman) students who submitted test scores. Do not include partial test
scores (e.g., mathematics scores but not verbal for a category of students) or combine other standardized
test results (such as TOEFL) in this item. The 25th percentile is the score that 25 percent scored at or
below; the 75th percentile score is the one that 25 percent scored at or above."</p>
<p>The capitalization of ALL, followed by the word "enrolled" means to me that all scores, whether they were submitted for application evaluation or not, should be included in the CDS median tables. I may be misinterpretting here, but if I'm correct, then its a matter of the SAT-optional schools not being called on this by the CDS administrators.</p>
<p>Amhest's Report to Secondary Schools, available on Amherst website, gives SAT I stats for applicants, accepted students, enrolled students. Here's the data for Class of 2008. (Can't access the Class of 2009 report.)</p>
<p>By the way, the percentage figures total by columns, not by rows. That is, for example, it is not that 50% of those with Verbal SAT between 750-800 were accepted; it is that 50% of those accepted had verbal SAT between 750-800. </p>
<p>Distribution of College Board SAT I Scores</p>
<p>VERBAL Applied Accepted Enrolled
750-800 1351 (31%) 496 (50%) 159 (42%)
700-749 1158 (26%) 255 (25%) 106 (28%)
650-699 880 (20%) 143 (14%) 66 (17%)
600-649 570 (13%) 85 (8%) 43 (11%)
550-599 263 (6%) 21 (2%) 9 (2%)
500-549 112 (3%) 2 (%) 0 (%)
450-499 44 (1%) 0 (%) 0 (%)
200-449 25 (1%) 0 (%) 0 (%)
Mean 697 731 722
Mid 50% 650-760 690-780 680-770</p>
<p>MATH Applied Accepted Enrolled
750-800 1116 (25%) 454 (45%) 144 (38%)
700-749 1358 (31%) 283 (28%) 121 (32%)
650-699 992 (23%) 142 (14%) 70 (18%)
600-649 542 (12%) 88 (9%) 38 (10%)
550-599 247 (6%) 31 (3%) 8 (2%)
500-549 94 (2%) 3 (%) 1 (%)
450-499 40 (1%) 0 (%) 1 (%)
200-449 14 (%) 0 (%) 0 (%)</p>
<p>Mean 696 727 721</p>
<p>Mid 50% 650-750 690-780 680-780</p>
<p>Another point of note: </p>
<p>As listed on the Amherst website under enrollment statistics, not as part of the report to secondary schools, for the Class of 2008, the 25/75 for admitted students is 1380/1560 (which is what you get if you add up the V & M scores for accepted students in the above data). By contrast, for the Class of 2009, the 25/75 for accepted students is listed as 1340/1530. Seems an awfully large decline. Odd.</p>
<p>
[quote]
I think the moral of the story is that you have to pay close attention to the % of students for whom SATs are reported. When that % closes in on 100%, you can take the numbers at face value.
[/quote]
</p>
<p>This is true for a school which only accepts one test. If the institution accepts either ACT or SAT, it will always have students who submitted only one but not the other. The # of scores reported can therefore never be 100%, even if the institution discloses everything.</p>
<p>As for the earlier discussion on what institutions SHOULD report, the CDS standard seems unequivocal to me. However, on things like this, institutions sometimes vary on whether they include internationals, students accepted to special "bridge" programs, and so on. It would seem that the CDS standard as reported above would compel institutions to include them. Our CDS drone isn't in today or I'd ask her.</p>