ivyleaves

U.S. News evaluates educational quality by weighting measures like faculty-student ratios, admissions selectivity, financial resources and alumni giving.

However, Forbes evaluates educational quality by weighting such things as the students enjoyment of the college and courses, and most important how successful are they once they graduate!

Full ranking can be viewed on the below link...you may need to skip their advertizement page.

How to Choose a College - Forbes.com

Denzera

"Educational Quality"?

Looking at that list makes me think it's total horse manure. MIT is 17th, but Boston College is 11th. U-Michigan is 39th but Southern Methodist University is 13th.

The Revealed Preference Rankings remain my only real scientific guide to these things. If you asked each person going to a particular school whether they'd prefer to go to a school ranked above them on the list, would they say yes? Measure this by looking at students who were admitted to schools A and B and seeing how often they picked A over B, or vice versa.

ilovebagels

This list = FAIL

charlesjeon

Conclusion that I see =

HYP always 123

ivyleaves

Another conclusion = the 8 Ivy Leagues are in the top 15 Again

Denzera - I respect your views, however isn't the "Revealed Preference Rankings" a popularity ranking? And popular is not always better. I betcha the most popular person from your high school will not be the most successful person in life.

Grades 1-8 prepares you to be successful in High School
High school prepares you to be successful in College
College prepares you to be successful in life.

The end product of the college education must be an important criteria, which is missing USNWR and RPR.

confidentialcoll

"Conclusion that I see =

HYP always 123"

this is probably why forbes did what they did, because unless people see hyp up there, they wouldn't believe the rankings, unfortunately it seems like apart from those 3 up there, they've ordered the rest in a weird manner.

At the same time they're metrics aren't as bad as i've seem: here are the problems:

1)ratemyprofessor is equivalent to using wikipedia in an academic paper, there's no quality standard, I could go to brown's section trash profs, and they'll fall in the rankings next year. There is also no account for biases of perception across different schools.

2) graduation rates are silly, because they tell you nothing about a university. Caltech which maintains a high level of rigor and has a drop out rate of 10-15% is punished for this, ok then, let's give points to schools that don't challenge their students.

3) fulbright and rhodes scholarships look at the top 1% of a batch or a select 1% of the batch, not the rest and is thus both narrow and in the realm of uncertainty.

4) the who's who of america is not bad, seeing that it's a large number of people, but putting people in the who's who probably has little basis.

Overall, there are a few outliers, but the rankings aren't atrocious, my feeling is that people are so caught up with us news, that they forget that usnews is pretty arbitrary to begin with, there's no reason MIT should be number 7 on usnews, they're the best engineering school in the country, and they're good at econ and the sciences, but they offer social sciences and humanities too, and relative to peer institutions, they're really not very good at these. Or brown which is very difficult to get into, focuses on its undergrads, puts out successful alumni gets shafted on usnews, because it doesn't do huge amounts of research and doesn't really have grad schools.

SMU, samford and BC are surprising; removing the engineering and good grad school bias, which us news has in its sat scores and peer review inclusion, it isn't all that strange that michigan is at 39.

kmatimber2

The revealed preference rankings don't mean a lot- it's based on what schools applicants pick BEFORE they attend. The choice that is used as the basis for the ranking is a largely uneducated one, as one can't really know much about what school is like until they go.

sac

I agree with kmatimber2. Also, the revealed preference ranking is based on what schools students pick among those they've applied to. There is an earlier preference at work when choosing where to apply. And, in the end, it just tells you something about your fellow USNWR-influenced high school students rather than the college.

If you're going to use rankings, it's important to know whether the measures they choose as important are important to you.

Columbia was ranked no. 1 in 2007 by the Center for Measuring University Performance.The Center for Measuring University Performance
This is based on research and endowment dollars, number of faculty in the National Academies, among other measures which you can read in their report. Obviously, this excludes LACs. But if you are headed to a research university, it's a ranking that tells you much more than the revealed preference of high school students. Here are the first five, in order, from 2007.
1. Columbia
2. Harvard
3. MIT
4. Stanford
5. U of Penn

sac

Here is how they rank: The Center determines the Top American Research Universities by their rank on nine different measures: Total Research, Federal Research, Endowment Assets, Annual Giving, National Academy Members, Faculty Awards, Doctorates Granted, Postdoctoral Appointees, and Median SAT Scores. (The Source Notes section of this study provides detailed information on each of the nine indicators.) The tables group research institutions according to how many times they rank in the top 25 on each of these nine measures. The top category includes those universities that rank in the top 25 on all nine indicators. The bottom category includes universities with only one of the nine measures ranked in the top 25. Within these groups, institutions are then sorted by how many times they rank between 26 and 50 on the nine performance variables, with ties listed alphabetically. A similar methodology produces a second set of institutions—those ranked 26 through 50 on the same nine measures.

confidentialcoll

"Here are the first five, in order, from 2007.
1. Columbia
2. Harvard
3. MIT
4. Stanford
5. U of Penn"

sac, don't know if you noticed, but these are in alphabetical order, and that's because they all have all 9 metrics in the top 25, the ranking is actually:

1)Columbia, Harvard, MIT, Stanford, Penn
6)Duke, UC-Berkeley, Michigan
...

sac

Thanks for the correction. I can live with that.

I didn't notice because the far right column in the report lists the position in the previous year, and does not take note of ties. But I do see that you're right, proving once more that I -- unlike my S -- am not a number person.

Denzera

This thread is not about the Revealed Preference Ranking, but:

First off, here's the link to the PDF, which I've put up for purposes of discussion.

no. its methodology normalizes out the number of people who got into each school, such that while each data point (one school being picked over another, out of a pool of X to which they were admitted) corrects the numbers towards the information they're trying to reveal, nothing is weighted by the number of data points. The more students who applied (and thus got in) to Penn does not improve Penn's ranking relative to, say, Stanford, IF, given the option, those who were admitted to both Penn and Stanford tended to choose Stanford.

The most popular person in high school was very unlikely to go to Harvard. The "popularity" that this metric includes, if it can be said to include one, has nothing to do with that sort of popularity.

However, in some sense, all of college admissions is a popularity contest. The notion to going to college is a popular one, reinforced by societal pressure. But underneath that societal pressure is a number of very good reasons why a rational person, independent of that social pressure, might choose to go to college anyway. Similarly, there are a number of reasons underlying why people choose to go to Harvard when given the option. You can call it popularity, but they're measuring this in the most scientific way possible, with the most amount of noise removed from the data.

Right, but that is the only point at which they have a CHOICE of school. After they attend, they may rationalize that choice for any of a large number of reasons. But when they are presented with a choice of multiple very good universities, where do they go? I think there's a lot more information (with a lot less noise) in that data than there is in any other set - which is why I keep referring to it like I had written it or something

First off, the issue of bias based on where these students choose to apply comes off the data, if you read their methodology. They're quite brilliant about it. Your latter point is correct, however.

The bigger trouble is that no ranking system that talks about "the college" has fewer objections than this one. Most people find the whole system to be hopelessly out of whack - and I frankly agree. Many of the top colleges are SO different - in campus atmosphere, academic specialties, student-body "personality", etc - that trying to rank them is a fool's errand. However, for those who absolutely must have a sense of how they all stack up, this seems like the, uh, "least unintelligent" way of doing so.

edit: pages 1-3 of the RPRs, just after the "Executive Summary", explain their justification for what they've done much better than I can in a post.

...and Forbes' method is, shall we say, more unintelligent. As other posters have detailed quite well above.

B77

FYI, the CCAP rankings of national universities as published by Forbes.com is inaccurate as to CCAP's actual ranking of SMU. Forbes.com lists SMU as 13 when, in fact, it's ranked by CCAP as 43. Take a look at CCAP's own list at the bottom of its own home page:The Center for College Affordability and Productivity. As you will see, the top 20 schools now begin with Harvard and end with Brandeis. BTW, I confirmed the accuracy of the CCAP-published list by email with CCAP.