<p>Maybe unlike in Cali being in the top 10% in PA actually means something. Not all Top 10% are created equal.</p>
<p>The most common reason for failing to finish college is financial difficulty. Test scores correlate with income, and so does ability to stay in school financially. It’s quite possible the outliers with respect to grad rates are due to cost factors rather than educational factors.</p>
<p>Compare NYU grad rate to supposed peer schools: 79% vs. high 80s for peers. NYU has notoriously bad financial aid.</p>
<p>
The 25% method would tend to penalize schools that prioritize the admittance of a good portion of their students based on criteria other than GPA/Test Scores/ECs. Those would be, in order of importance, Athletics, Legacy, and SocioEconomic Status and/or Race. </p>
<p>Almost all BCS DI Athletics schools need to feed that money machine… Stanford/Duke/Northwestern/Vanderbilt/Notre Dame/UC Berkeley/UVA/UCLA, Wake, etc. All of them need to specially handle HUNDREDS of admits each year to stock dozens of DI teams, especially football and basketball, but even the non-revenue sports are an implicit advertisement for the University, and losing teams, even in Women’s Volleyball, reflect upon the Institution.</p>
<p>Even within DIII, the importance of college athletics will vary widely from a Williams or Amherst, who highly value winning teams, to a Swarthmore, Pomona, etc. who could care less. </p>
<p>There should be a pretty high correlation between the importance of Athletics and the delta between 75% and 25%. Stanford is the most obvious example to me. Stanford has won the Director’s Cup, or Sears Cup, or whatever the “all sports throughout the year” is called these days, for years and years. Stanford spends more money on Athletic Scholarships than any other school, of that I am certain. Those admits, even in golf, rowing, water polo, soccer, etc. aren’t typically the 740/740 variety. They are usually of the “make sure they’re over 1100” for the two revenue sports, and "make sure they’re over “1250 or so” for the rest.</p>
<p>
</p>
<p>Just based off of what you stated here … you have to be kidding. There are so many top-tier public (and private) schools in CA it’s not even a contest between PA and CA schools. Granted, there are many more bad schools in CA also, as related to the larger CA population nos. between the two states. </p>
<p>With this said, and wrt the unstated, I’m sure Penn State’s calculation of top-10% is more legitimate. The UC calculation of top decile is based on a pooling of all CA hs grads, not wrt students’ native hss. This means top-25%-rank seniors at outstanding hss as Gunn, Torrey Pines, as well as at many other outstanding hss in CA, should be able to achieve a top-decile ranking within this pooling of CA graduates. A 3.8-uwgpa student wouldn’t be top decile at Gunn (it’s a decently complicated recalculation when figuring A-G UC courses, etc., I’m sure… especially since class rank at most if not all public hss in CA is based on wgpa), and might not even be top quintile. But my point is that these students are certainly top-tier and should be ready to step into any college and do well. Even the 50th-%ile rank students would do well at just about any school. Davis and Santa Cruz take a lot of these from Gunn and they do well at their u’s.</p>
<p>The actual per student’s native hs would probably be as follows:</p>
<p>Cal and UCLA, ~80%</p>
<p>UCSD, 60%, maybe less</p>
<p>Everyone else, < 50%, and some significantly so</p>
<p>Less than half of the Notre Dame Class takes the SAT (most use the ACT) so it’s result is skewed articificially high. Other than that, a good and interesting list.</p>
<p>[
</p>
<p>Well, so what? Why isn’t that relevant? A constant refrain on CC is that the bottom of the class drags down what the top of the class can do. I’ve always thought that was more true for small schools than for large schools, where the top of the class will have far more opportunities to separate themselves from the bottom of the class.</p>
<p>But why, at a small- or medium-sized school, isn’t the bottom quartile a more relevant metric than the top quartile?</p>
<p>To the OP:
Any chance you can add: (Percentage reporting SAT + Percentage Reporting ACT), then sort it in descending order?
The reason is that schools that have a summation of 100% of less may need to be “interpreted with a grain of salt”.</p>
<p>NotYourBusiness - I did that (at your request) and the vast majority of schools have a percent summation exceeding 100% of students who reported the SAT plus students who reported the ACT. Evidently, many students report both. Of the 423 schools I looked at, only 8 had less that 80% reporting either SAT or ACT.</p>
<p>Barrons - The high percent of UC students who are in the top decile of HS class does seem very strange. It sounds like there is a complex system for calculating rank in CA as per drax12. Do students in the top CA decile get admitted automatically to UCs? This could create an incentive to fake the class ranks in CA.</p>
<p>Sounds like everyone finds the Penn State grad rate believable and that perhaps the misleading UC classranks are throwing off the US News prediction formula resulting in the +17% overperformance. Correct?</p>
<p>
</p>
<p>The most charitable qualifier for the accuracy of the UC numbers is that they are estimated as opposed to be based on the exact individual high school ranks. The exercise seems to consist in finding enough data points to support the estimates, and ignoring the ones that do not fit the expected outcome. </p>
<p>In the grand scheme of things, this “detail” is trivial. The USNews and other ranking groups are happy to accept the estimates at face value, just as they are happy to accept numbers they know are incorrectly compiled and reported.</p>
<p>
</p>
<p>Class rank as reported by the high school is not used in UC admissions (and probably not even available to UC admissions, since high school transcripts are only used after admission to verify self-reported courses and grades and to make sure that senior year performance is acceptable).</p>
<p>In theory, UC can calculate class rank (using the UC admissions GPA formula, not whatever the high school uses) up to about the top 12% using the data submitted by high schools for Eligibility in Local Context purposes. Whether this is what they actually use to generate the class rank estimates reported in the common data set is something else entirely.</p>
<p>Many of the test-optional schools are not there - I assume that’s the reason why Holy Cross and Bates are absent.</p>
<p>
</p>
<p>No, incorrect. Student finances are a critical reason for failure to graduate. Test scores correlate with income and income correlates with graduation rates. To conclude that test scores are causal for graduation rates exclusive of finances is too simplistic (as is the corollary). It’s actually a multidimensional problem, and many inputs would be required to get an accurate depiction of the relationships, but to evaluate whether schools are “overperforming” or “underperforming” would at a minimum require controlling for income, financial aid, COA, and all other financial variables.</p>
<p><a href=“http://money.cnn.com/2011/11/21/news/economy/income_college/index.htm[/url]”>http://money.cnn.com/2011/11/21/news/economy/income_college/index.htm</a></p>
<p>
</p>
<p>
</p>
<p>
</p>
<p><a href=“http://www.aft.org/pdfs/highered/academic/march06/Gold.pdf[/url]”>http://www.aft.org/pdfs/highered/academic/march06/Gold.pdf</a></p>
<p>This was linked in a recent thread.</p>
<p>I realize some issues exist with the ranking, but still quite useful in early stages of finding score based <em>match</em> schools for my kids. So basically bumping to possibly help others in the college search effort.</p>
<p>The list is useful for determining reaches, matches, safeties.</p>