Parchment Student Choice Rankings

Colleges & Universities A-Z University of Chicago ✅
1

<p>For what it's worth:</p>

<p>Parchment</a> College Rankings 2013 | Parchment - College admissions predictions.</p>

<p>Parchment Releases 2nd Annual ‘Student Choice College Rankings’
Unique Ranking System allows comparisons between all colleges based on where students actually choose to enroll</p>

<p>Parchment 2013 Student Choice College Rankings
Quote startParchment ranks colleges and universities based solely on the decisions of hundreds of thousands of students who were admitted to multiple colleges and ultimately chose one to attend.Quote end</p>

<p>Scottsdale, AZ (PRWEB) September 05, 2012</p>

<p>Parchment Inc. today released its 2013 Student Choice College Rankings, the only system that compares colleges based on actual acceptance and enrollment data. Compared to the 2012 rankings, Massachusetts Institute of Technology is now ranked in the Top 5 colleges and University of Chicago, Brown University, and Amherst College are now ranked among the Top 10 colleges.</p>

<p>The Top 10 institutions in the 2013 Student Choice College Rankings are:
1. Harvard University<br>
2. Stanford University<br>
3. Yale University<br>
4. Massachusetts Institute of Technology<br>
5. Princeton University<br>
6. University of Chicago
7. Brown University
8. Caltech
9. Amherst College
10. U.S. Air Force Academy</p>

<p>For the complete 2013 Student Choice College Rankings list, visit: Parchment</a> College Rankings 2013 | Parchment - College admissions predictions.</p>

<p>The 2013 Parchment Student Choice Rankings were based on a sampling of more than 200,000 college acceptances from 2009 to 2012 with data from all 50 states, an increase of 67 percent in the data set since the 2012 rankings. Unlike rankings based on subjective judgments like reputation, or factors subject to "gaming," such as how many applicants a college rejects, Parchment ranks colleges and universities based solely on the decisions of hundreds of thousands of students who were admitted to multiple colleges and ultimately chose one to attend.</p>

<p>When a student is admitted to several colleges, he or she must decide which school to attend. This decision reveals a preference for the chosen school compared to the other schools the student could have attended. After observing enough of these decisions, Parchment ranks the colleges based on the students' revealed preferences. This method for ranking colleges is derived from physicist Arpad Elo’s rating system for chess players, which was later studied by Harvard University professors Caroline Hoxby and Mark Glickman. This results in the colleges ranked not from “best” to “worst,” but instead from “most likely to be chosen” to “least likely to be chosen.” Schools can only move up in the rankings by making their programs more appealing to prospective students. The rankings system aligns the interests of colleges and students to support better match-ups.</p>

<p>A unique result of this methodology is that all colleges can be ranked in a single list unlike the multiple lists often published by other ranking systems separating colleges arbitrarily into “National,” “Liberal Arts” or “Regional.” To compare the Student Choice Rankings of any two schools, try the College Matchup app at Parchment.com.</p>

<p>“Colleges can game other rankings systems – boosting their position by taking certain steps to appear more selective or have higher student test scores,” said Parchment.com General Manager Brent Pirruccello. “Parchment’s rankings provide the only list that is student-centric and equitable in that student decisions are the sole criteria used in comparing a broad-base of higher education institutions, putting them all on a level playing field.”</p>

<p>The rankings are based on student enrollment decision data from Parchment.com, a free service students use to research colleges and discover their chances of admission, see how they compare with peers, get college recommendations, and send official transcripts when they are ready to apply. Students using Parchment.com can opt-in to make their enrollment decision data available for the Student Choice College Rankings.</p>

<p>About Parchment
Parchment is the leader in electronic transcript (e-transcript) exchange. More than 9,000 high schools (over 30 percent of the U.S. secondary school market) and postsecondary institutions have exchanged 4 million transcripts using the Docufide by Parchment™ SaaS platform. Beyond providing the cost and operational efficiencies of electronic transcripts, Parchment works with learners and institutions around the world to unlock the value of transcript and credential data – helping people collect, promote, and share their education credentials in simple and secure ways. At Parchment.com, students can research colleges and discover their chances of admission, see how they compare with peers, get college recommendations, and send official transcripts when they are ready to apply. Founded in 2003, Parchment Inc. is a venture-backed company headquartered in Scottsdale, Ariz. with offices in Roseville, Calif. and Washington, D.C.</p>

2

<p>Good to know, but parchment has no credential.</p>

3

<p>A ranking based on the perceptions of students who have had zero to very little real exposure to any of those schools. How is that not subjective? What it really comes down to is who’s on the hit parade in a given year, which could be very influenced by marketing (what they call here making programs more appealing), and general prestige. Granted, prestige and good reputations are almost always a consequence of quality, and for that reason I approve of US News including peer assessment scores in its methodology, but in those cases its people in the world of academia whose opinions are being taken into account. The only significant factor that this ranking could be indicative of is financial aid, but there is no way to quantify to what degree with the current set up. BTW this is a rant I’ve had in my head for a while, it’s not because I think U Chicago should be higher or anything like that.</p>

4

<p>PMC18, this is an incredibly accurate ranking besides maybe Penn ranking a little too low. All of the schools that are listed as “Peer Colleges” for these universities seem to be absolutely spot on.</p>

<p>The only issue may be the sample size but Parchment accurately captures the desirability of American universities at the undergraduate level. Good job to UChicago…gotta give credit where credit is due.</p>

5

<p>Students are more likely to choose UChicago than Penn? I thought it’s the other way around… Am I reading this ranking correctly? Is that how I should be interpreting it?</p>

6

<p>DivineComedy,</p>

<p>You can also look at specific school vs. school matchups, but yes, generally, this indicates that UChicago is “more desire-able” than the schools ranked below it.</p>

<p>I still have serious issues with the veracity of this ranking. The sampling size tends to be quite small, and I just don’t put much worth in these yield matchup rankings. They might give some evidence of the desireability of a particular school but, for the most part, they seem very suspect. </p>

<p>In some ways, this ranking could just demonstrate that certain admissions committees are very adept at choosing the high-achieving students that are most likely to attend if admitted.</p>

7

<p>Divine Comedy and Cue, there is some selection bias at play here. Chicago is ranked above Penn, Duke, Columbia, etc. despite losing the individual head-to-head matchups because while the majority of qualified high school seniors might prefer those three private schools to UChicago in general, the ones who actually do end up applying to UChicago slightly prefer it overall to those who apply to the other three schools.</p>

<p>It’s kind of hard to make sense of but it might be fair to say that while Columbia beats Chicago in cross admits amongst just those two schools, Chicago is winning more battles with HYPSM and Caltech than Columbia is. That’s one plausible explanation. UChicago appeals extremely strongly to a “certain type” of student so much so that that overriding preference is far greater than the one Penn applicants express to Penn and may lead to greater wins amongst HYPSM admits.</p>

8

<p>Goldenboy:</p>

<p>I think measuring preference is generally useless because of a lack of information and, in the case of parchment, too small a sampling size.</p>

<p>If you run matchups on parchment, some surprising results occur, such as 75% of admits “preferring” upenn over dartmouth, or uchicago and princeton splitting admits about 50/50. </p>

<p>My issue with Parchment is that the sampling size is just way too small. According to Parchment’s figures on UChicago they tracked 1342 of the 3344 students who were accepted to UChicago. There’s a large, large chunk of admits who simply aren’t accounted for in this study.</p>

<p>Further, financial aid probably plays a role in where admitted students go, as does the region the applicants reside, etc. </p>

<p>There are just so many factors that make a baseline ranking of “preference” really problematic. All this really shows is the ability of an admissions office to admit students who want to come to a particular school. There really isn’t such a thing as a student looking at two schools in a vacuum and just selecting one based on “real” preference.</p>

<p>Because of that, I just don’t think it’s worth placing much emphasis in rankings beyond what we already know:</p>

<p>Duke, UPenn, UChicago, Columbia, Brown, etc. are all peers, and a student (all else being equal) should just go based on fit - that’s it.</p>

9

<p>Cue7: </p>

<p>As a mathematician, I really have never understood anyone’s arguments when it comes to the accuracy of these rankings. You say that the sample size is quite small, but quite the contrary! There are over 500 data points for UChicago alone over 2009-2012. Considering that about 13,000 students were admitted around this period, this is mathematically quite a large sample size with which one can make quite a few conclusions. Remember, most voting/etc. polls in the US use a sample size of about 1,000 people to make conclusions about the entire US population (3 million+), and those are usually pretty accurate plus/minus a few percentage points, so why would you have qualms about 500 data points being used to make conclusions about 13,000 students? Mathematically and statistically, the methodology is sound, which is why the schools expected to win usually DO win.</p>

<p>It’s clear from this data that Chicago is now able to compete against the best, including Harvard, for the top students, when it obviously couldn’t do so even 5 years ago. Why wouldn’t this be relevant?</p>

10

<p>Phuriku,</p>

<p>I can’t comment on the math behind it, and perhaps the sampling size is adequate enough.</p>

<p>My larger point, though, still stands. Parchment shows us that Penn “wins” 60% of admits from UChicago, and 75% of admits from Dartmouth. Moreover, UChicago and Brown draw pretty evenly in terms of cross-admits.</p>

<p>What does this tell us? To me, this doesn’t say much of anything, except that generally, all these schools are pretty similarly situated. Perhaps in the voting polls a slight difference can mean a LOT. In this case, differences don’t mean much of anything - these are all peer schools. </p>

<p>Moreover, UChicago’s “big numbers” admissions approach just tells me that UChicago has chosen to play the game. I don’t think there was ever really a question that UChicago was a pretty formidable school - there was just the question of whether it would pay attention to the statistics it could most easily change (e.g. selectivity).</p>

<p>All of these “stats” just point to what we already know - UChicago is playing the game just like everyone else now, and showing a commitment to its college.</p>

11

<p>While the ranking indicates that Chicago loses cross admit battles to Penn, Columbia and Duke. It still does better overall because it loses far less badly to Princeton and Harvard than those schools do. This is a rather interesting phenomena. Chicago does relatively well against Princeton and to a lesser extent Harvard, but loses to school’s considered to be its immediate peers.</p>

12

<p>iameinstein:</p>

<p>It’s an interesting point. Chicago gets 41%, 40%, and 37% of cross-admits against Duke, Columbia, and Penn respectively, whereas it gets 47% and 45% vs. Princeton and Stanford respectively.</p>

<p>Why the discrepancy? It’s really quite simple: Students hedging their bets by applying ED to Duke/Columbia/Penn and EA to Chicago simultaneously, and getting into both. Such cases amount to automatic wins for D/C/P and automatic losses for Chicago. </p>

<p>Since HYPSM don’t have ED, Chicago has “true” matchups against these 5 institutions, and you see the true picture of how students pick if they have a true decision instead of being forced into accepting one of their admissions offers. </p>

<p>I imagine if we took out the students hedging their bets, Chicago would be winning against its immediate peers 60-40, or at least 50-50. The key to Chicago winning more of its cross-admit battles against its immediate peers is quite simple: stop admitting so many students EA. However, this would lead to an overall loss in yield, since EA yield has traditionally been higher than RD yield; plus, EA probably helps us win cross-admits against HYPSM. Therefore, it’s best to just accept the results and move on. The current Chicago admissions model is working fine. :)</p>

13

<p>More unique schools have an advantage in this ranking system, so UChicago has an edge over UPenn, Duke, etc. The same for U.S. Air Force Academy. I am a little bit surprised that Brown is ranked that high.</p>

14

<p>The page keeps saying “not found” even when I google it, but I’m assuming this is based solely on the the current matriculants (class of 16)? I wonder how much Harvard & Princeton re-instating their SCEA programs changed things up from last year…</p>

15

<p>Phuriku and others,</p>

<p>Maybe it’s just me, but do some of the parchment figures really not pass the sniff test? According to this site, 75% of admits prefer Penn to Dartmouth, and 60% of admits prefer Penn to Columbia. Further, 65% of cross admits prefer Duke to Dartmouth, and 67% of cross admits prefer Brown to Dartmouth. </p>

<p>Don’t these numbers appear to be really skewed? Again, this just looks odd to me. Thoughts?</p>

16

<p>Cue7:</p>

<p>In the case of the 75%, 67%, and 50% figures, this usually means there’s not enough cross-admit data to make conclusions for the two schools involved. (That being said, this isn’t to imply that you can’t make inferences with larger scope based on the total amount of data involved. Take the voting example again. You might be able to assembly a clear picture of voter preference by analyzing a sample size of 1,000 in representation of the 300m+ U.S. population. However, you might only have 15 data points for Hispanic viewpoints, so you wouldn’t be able to make any conclusions for Hispanic voters, even though you have enough data to make conclusions for the general population.)</p>

<p>Otherwise, I don’t really see any crazy things in the figures you cited. Duke and Dartmouth are peers, as are Penn and Columbia. In any case, this is where you need to look at the confidence intervals. For Penn/Columbia, for instance, we’re slightly under 95% confident that Penn wins a few more cross-admits than Columbia (since the 95% confidence interval for Columbia is 30.4% to 51.3%). It could certainly be the case that Columbia wins more cross-admits, and there’s a ~6% probability of that.</p>

<p>In any case, statistics don’t lie. As long as you understand that we’re working with probabilities here and not hard numbers for individual cross-admit matches, the data makes sense.</p>

<p>(As a side note, there’s a reason why there are so few cross-admit data between Columbia/Penn/Dartmouth/Duke. These are ED schools, and so the only cross-admits occur when students apply to these schools and get admitted RD to both. Each of these schools have ridiculously low RD rates, so it’s very difficult to imagine too many people being admitted to 2 or more in that group.)</p>

17

<p>I wouldn’t say these results are all that skewed Cue7. Even if Penn’s CAS competed with Dartmouth, Brown, Duke, Columbia, etc. for applicants, the school still has Wharton which essentially only splits applicants with HYPSM. Wharton represents a sizeable chunk of Penn’s population so its going to have a notable cross-admit advantage over its immediate peers. Also, Penn has a Nursing Program which is the best of its kind and affiliated to an Ivy, so I doubt that prospective nurses are really going to turn down Penn Nursing to go a state school program due to prestige.</p>

<p>So that explains why Penn matches up better head to head with Dartmouth and Columbia. As far as Dartmouth and Duke, they are pretty similar schools except Duke has better weather, sports, and faculty/academic programs across the board (I doubt most prospective undergrads care about the last part though). I know a lot of cross admitted Dartmouth engineers who turn down the school for the likes of Columbia, Penn, and Duke since the latter have much stronger offerings in this area and have a better reputation in the industry.</p>

<p>Its tough to explain Brown’s popularity in all these Revealed Preference Rankings but its just seen as an enjoyable and happy place where there’s a lot of grade inflation and freedom to explore different areas of study without harming one’s GPA.</p>

<p>I think what Phuriku said is spot on with regards to Chicago’s appeal except he forgot to mention that Chicago not having an engineering school helps it attract a more self-selected student body that is more likely to choose it. I know a sizeable amount of engineers who choose state schools like Illinois and Berkeley with a half scholarship over Penn, Duke, Columbia, etc. since the former schools actually have better engineering schools.</p>

<p>Chicago doesn’t have the engineering issue.</p>

18

<p>Regardless of the sample sizes there seems to be quite a bit of noise in the data as evidenced by the large year-to-year shifts. It strikes me as extremely implausible that Chicago actually went from 24th to 6th in revealed preferences in a single year. There are a lot of other large jumps as well. 10 of this year’s top 25 schools moved 10 or more places in a single year.</p>