<p>UCLA – UC as a whole – concentrates more on gpa, both w and uw, moreso than scores. UCLAs uw gpa is ~ 3.82, wgpa through senior year would probably approach 4.4. The 4.24 wgpa that UCLA reports would be soph and jr. grades.</p>
<p>UCLA does some quirky things wrt SATI (I’ll abbreviate as SAT) and ACT reporting, thereby lowering its reporting of scores: </p>
<ol>
<li>UCLA’s website reports mean SAT, as well as mean ACT.</li>
<li>As Bubbles stated, UCLA doesn’t superscore.</li>
<li>UCLA double-counts a good portion of SAT and ACT scores.</li>
</ol>
<p>If you allow me to concentrate only on UCLA"s SAT reporting…</p>
<p>Wrt point 1:</p>
<p>UCLA reports a mean SAT of 1910. I haven’t looked at component scores, but I’m sure that the Writing component score probably lags in relation to Verbal and Math. I’m sure that this applies to students taking the test nationally; that is, students haven’t quite caught onto and adjusted to this fairly newly added writing component of the SAT … or test-scorers are tougher grading on this particular part.</p>
<p>Many, mainly private, maybe most us like to optimize their reporting of SATs by reporting 25th and 75th percentile medians, and occasionally report the 50th. They would never report a mean as UCLA does, because this would indeed lower their reported nos… When UCLA reports this 1910 mean, this will understate things overall because a mean SAT score will be materially lower than a 50th % median because there are a lot of really lower scores hidden in the bottom 25% of any us entering class, eg, athletes, special admits, etc. </p>
<p>If one notes that the greater concentration of UCLA matriculants is in the 2000 score range, ie, the highest point of the distribution curve, one could guess that 50th % median is probably around 1970 or so. The fenceposts for SAT scores are a little too far apart wrt spacing to really determine a more accurate determination of the 50th% SAT. Similarly finding a midpoint between 25th and 75th %-ile would be an erroneous 50th for any u, especially UCLA, for the same reasons given above … given the shape of the distribution curve. </p>
<p>So the act of converting from a mean/average to 50th %-ile mean is as follows:</p>
<p>1910 -> 1970</p>
<p>Wrt point 2:</p>
<p>UCLA doesnt superscore SATs. Those students who take the SAT more than once, depending on factors of the amount of study prep and comfort the first time vis-a-vis the second, tend to ascend their scores pretty materially, 60 to > 100 points. The act of converting UCLAs unsuper to superscored, is tough, depending on the % of UCLAs matriculants who retake the SAT.</p>
<p>This is similarly hard to find, but in the admit cycle for 2011, UCLA reported > 135% who took both SAT and ACT. This would probably mean that there were a good portion who retook solely the SAT instead of changing from one of these two boards to the other. </p>
<p>I dont have anything definitive on the bump-up of scores from unsuper to superscored, but a rule of thumb on the College Search board stated something to the effect of 20 points per section. Given there are three parts this would mean a 60-point increase for UCLA. But conservatively, Ill use 40.</p>
<p>The act of converting UCLAs scores from unsuper (Im unaware of the term for this) to superscored is:</p>
<p>1970 -> 2010</p>
<p>Lastly wrt point 3:</p>
<p>UCLA reports all of the ACTs and SATs reported by ~ domestic matriculants. Im not sure why, but on UCLAs website, there is a definite lower portion of SATs reported by an amount of > 1,000 of the total who matriculated to the U, which appears to solely domestics. </p>
<p>I would guess that Intl students would have higher scores on average than domestic students becasue scores would carry more weight for them because of often a lack of conversion from foreign-school grades to US, to compare each of these sets of students in admission. </p>
<p>But working with the info the UCLA Admissions website has given us, this double counting will affect the lower tier scores (the 25th %) which were double counted for of the apparent 35% of students who reported both scores. This is because for each student there is only one score that will determine admittance to UCLA instead of both SAT and ACT. </p>
<p>If one scored 31 on his/her ACT and 2050 on the SAT, the 31 will have helped him/her gain admittance to UCLA. Similarly if another student scored 2130 on the SAT and 29 on the ACT, the SAT will have helped this student gain admittance. </p>
<p>One has to figure that 10% of the 35% (~ 30% of the total who reported both) might have like scores on the SAT and ACT. So we have to essentially throw these out, unless they both were super high or super low. This would leave us with 25% (~ 70%) who have one score too low, relative to the other test. </p>
<p>If we assume that these students took, say, the SAT first, and posted a low score, and decided to switch to the ACT later, and scored materially higher, we have to throw out the SAT. Similarly for those who went for the ACT first and SAT second -> elide the ACT.</p>
<p>The essence is this: there are a material amount of scores that UCLA reports for both ACT and SAT that should be excluded based on the one-relevant-score-per-student reporting. What this amount of increase would be would be tough to figure. </p>
<p>Btw, all schools report 15-30% redundancy in SATs and ACTs. There seems to be a downward adjustment for most private us, which could be their partial attempt to reduce the %s who report both tests because one is materially higher for a good portion of students. The CDS and IPEDS however dont force us to report the hard nos. who score within the various intervals to sum up to the %s greater than 100% so one has to assume a best-foot-forward reporting; one has to assume that most if not just about all us (except the extremely transparent UCLA) make the adjustment to a one-score-per-one-student reporting. The UCLAs Admissions website shows that it reports redundant ACTs/SATs for the CDS and IPEDS because the numbers are consistent from its native website to the federal reports. </p>
<p>Anyway heres my adjustment:</p>
<p>2010 superscored, median SAT -> 2030-2040, an additon of 20-30 points, not significantly higher, but a definite increase as there would have to be.</p>
<p>Given the things above, a 2035 median, superscored, one-score-per-student isnt bad. If we assumed symmetry for all parts, I discounted this above somewhere, we have a two part of ~ 1357. This isnt bad based on the fact that there are a significant portion of UCLA students who come from extreme poverty and dont have the funds to really ascend their scores as wealthy students do (because high scores do tend to run commensurate with wealth). Theres a large group of UCLA students who took the boards only once, probably with poor prep, who could probably ascend their scores by a good 200 points.</p>
<p>The only reason why UCLA reports scores on a lower bad-foot-forward mode, Im thinking, is to give poorer students a boost in confidence in applying to the U. You see this all the time, for instance, youll see fairly recently that UCLA received another boost in undergrad applications, with its website noting that there were increases in URM students, over which the administrators of the U are ecstatic. </p>
<p>This similarly explains why kids say how hard UCLA is in which to gain entry, when it seems the SAT/ACT nos. dont necessarily manifest this (but grades do). Besides UCLA dumbing-down the scores of its matriculants, there is a two-tiered admission system at the U, that requires students from wealthy top-tiered high schools to have generally impeccable stats, which includes > 2100 SAT, and those from underperforming HSs to have serviceable but not particularly high scores, even some with really low ones.</p>