UCLA or Claremont McKenna? Or UC Berkeley?

<p>

</p>

<p>My mistake…</p>

<p>UCLA’s admissions website excludes Internationals from all the stats calculations. And I have tried to see how the admin would incorporate the influx of non-residents to the U in more of the federal forms, so up until, say, six months ago, I would try to access its CDS information and wouldn’t be able to do so. The link conveniently wouldn’t work.</p>

<p>I just tried, just now, and the link to its CDS does work. And though the UCLA website does indeed remove Int’ls from the calculations, the U does include them in the CDS link through AIM, for the fall of 2013.</p>

<p>To cross-check with the UCLA site here are the nos:</p>

<p>CA Residents 4,107
OOS 940
Int’l 654
Total 5,701</p>

<p>CDS:</p>

<p>SAT’s submitted, 5027, 88.2%, 5027/.882=5696
ACT’s submitted, 2765, 48.5%, 2765/.485=5701</p>

<p>So the nos. do cross-check.</p>

<p>In its admissions site, UCLA does four things to understate scores: mean scores not median; does not superscore; counts redundant scores at ~ 137% (just a hair more than the CDS); does not include Internationals who have higher mean/median scores. </p>

<p>In its CDS, it does two to understate: does not superscore; counts redundant scores at 137%.</p>

<p>Other cross-checks:</p>

<p>Weighted-GPA per UCLA website Fall 2013: 4.31 (probably fully WGPA, because a tad too high for UC GPA) </p>

<p>UW-GPA per UCLA website Fall 2013: 3.84</p>

<p>Per CDS 2013 in Unweighted GPA mean section the U reports: 4.29 (Therefore, probably fully weighted GPA, 10-11 Grades, a-g courses)</p>

<p>Here’s my reference to my [study](<a href=“http://talk.collegeconfidential.com/university-california-los-angeles/1472470-my-study-on-superscoring-of-sat-and-other-stats-related-things.html”>http://talk.collegeconfidential.com/university-california-los-angeles/1472470-my-study-on-superscoring-of-sat-and-other-stats-related-things.html&lt;/a&gt;) on superscoring. (I found the adjustment to super from “sitting” scores is about 20 points per section.) 3-part, add 60, 2-part, add 40, which is an idea propagated on this board, and for which I found to be fairly reliable.</p>

<p>Mean SAT per UCLA web, 2013, 1923</p>

<p>Per CDS: Splitting the diff. of 75th and 25th, 1930 (but of course higher scores are bunched, and lower scores are more spread, so midpoints would not be reliable, and should push up the difference between the 75th and 25th)</p>

<p>UCLA 50th%-ile scores adjusted for super, excluding redundant scores, etc per my study ~ 2050-2060. I would put Cal’s by adding superscore adjustment to ~ 2110. (I also guesstimated for the 75th and 25th in my study for UCLA.)</p>

<p>The main difference between UCLA’s and Cal’s stats is that UCLA takes more “at-risk” students than Cal because the geography of UCLA for localized admissions has more underperforming hss, and therefore UCLA is “obligated” to take more holistically based admits (who generally have much lower stats), though this is an administrative decision, which is more campus-based than an overall UC policy (because admissions differs between the campuses). </p>

<p>Btw, Cal does conveniently leave out Spring admits from its CDS’s.</p>

<p>Excuse me for manifesting my math challenges.</p>

<p>Another correction:</p>

<p>In trying to estimate the 50th %-ile from UCLA’s CDS per my paragraphs above:</p>

<p>

</p>

<p>Splitting the diff of 75th and 25th would manifest a 1955 score. There probably isn’t a true median (50th) that would be base five as here but would pretty much have to be base ten, or a score ending in a zero.</p>

<p>Further, the true 50th would most likely be higher, ~ 1970 based on the tighter distribution among higher scores nearer the top filtering down to the middle, the 50th, rather than the lower scores affecting upwards to the 50th.</p>

<p>Could the 50th be higher? Perhaps, but I think 1970 is a probably legitimate sitting-scored (as opposed to superscored) SAT with 137% of scores included, based on matriculated frosh,</p>

<p>So the adjustment would be per my notes:</p>

<p>1970+60 (superscore adj)+ 30 (adjustment for one score/student) = 2060 adjusted median. One-score-one-student adjustment could be higher, especially since 137% is quite high, but I cannot make a higher claim, without actual data, which is not available anywhere as far as I know.</p>

<p>So for UCLA, 2013:</p>

<p>UWGPA, 3.84 (probably would fall a bit with senior grades included)
WGPA, 4.29 (probably fully wgpa, would rise with senior grades)
SAT: 2060 (with adjustments)</p>

<p>Cal, 2013:</p>

<p>UWGPA: 3.86
WGPA: 4.30+ (maybe low end based on possible UC GPA calc at 4.20-ish)
SAT: 2110 (Data Cal provides doesn’t show if Cal counts redundant scores so scores adjusted only for super) </p>

<p>Both UCLA and Cal would take a decent to a lot of poorer prepped students who scored lower, but if better prepped, could have scored in the 2100 range because they attended lower-ranked hss – the potential is certainly there for these students. </p>

<p>Btw, I’m not a fan of Wesleyan’s reporting of a mean gpa of 3.7 in its reporting. Is it 3.65 or is it 3.74 and rounded up or down? Its data has to be to two decimal places – use it.</p>

<p>Wow, that was a lot of work, @drax12!!</p>

<p>I agree with you about reporting GPA to 2 decimal places, btw.</p>

<p>It was a moderate amount of work, but I just did a little every day for a week or two. Not that bad. The key was to get the chop off the data from the high school referenced and convert the student data into numerical form from strings in Excel.</p>