Click here to go to the NEW College Discussion Forum

About two months ago, I started a thread which compared Harvard and Princeton along several statistical dimensions. Although I focused on common variables such as graduation rates and alumni giving rates, I parenthetically referred to the sharply contrasting rates of suicide on the two campuses.

I just noticed that The Harvard Crimson website is currently running a feature story on mental health issues on campus. The article contains information which allows me to update my earlier data, obtained from another Crimson story about a year ago.

http://www.thecrimson.com/article.aspx?ref=357115

According to the essay linked above, in the past 10 years, at least 15 Harvard undergraduates have, tragically, taken their own lives. The article cites a recent Crimson poll in which 10 percent of undergraduate respondents said they had seriously considered attempting suicide in the past year and 1 percent said they had actually attempted suicide in that time. The story also mentions a 1998 incident in which a female undergraduate was stabbed to death by her roommate, who then committed suicide. Finally, the Crimson quotes the Director of University Health Services who estimated that more than 20 students take time off from school each year due to mental health problems.

I don't have any new data on Princeton subsequent to my original thread. According to The Daily Princetonian, the campus has suffered three suicides over the past 30 years: one undergraduate in 1989 and, later, two graduate students.

Copied below is my first post. [This thread is at least somewhat dated in that, in the 2004 admissions cycle, for the first time in at least three decades, Yale had a lower admissions rate than either Harvard or Princeton.]

A few days ago, I was motivated to respond to “Brownalum,” a frequent College Confidential poster who seemed to be calculating his or her own statistics to support the contention that Yale University is the best college in America. I composed a thread entitled, “Yale Ranks Below Harvard and Princeton Statistically,” containing a variety of common and not-so-common statistics which in their totality show that, from a quantitative perspective, Yale clearly ranks below its two most direct peers.

Thanks very much to all the posters who have subsequently added to my thread. Some of you have asked whether I was “NYCFan,” a well-known avid supporter of Harvard. I think I can answer that question definitively by taking the numerical analysis to its logical conclusion, namely differentiating between Harvard and Princeton. In particular, I believe that the weight of the evidence points toward Princeton.

First, allow me to repeat the foreword to my other thread: A compelling QUALITATIVE argument can be made that any one of several universities is the best college in America. A reasonable case could be compiled for at least half a dozen names, maybe a dozen.

But once the debate shifts to QUANTITATIVE criteria, it is virtually impossible to advance a comprehensive analysis which leads to any conclusion besides Harvard or Princeton. Indeed, after examining variables as disparate as admission rates, enrollment yield rates, graduation rates, SAT scores, high school ranks, faculty salaries, alumni giving rates, endowments per student, and US News & World Report academic reputation rankings, it is remarkable how thoroughly these two institutions dominate the academic landscape.

So how to distinguish between the two leaders?

The most conspicuous reflection of Harvard’s primacy is its considerable success in attracting students, as reflected in its enrollment yield rate:

A. Enrollment Yield Rates (2002)
1. Harvard 78.8%
2. Princeton 73.4%
3. Stanford 69.2%
4. Yale 63.6%
5. Columbia 63.4%
6. Penn 62.1%
7. Brown 59.0%
8. MIT 56.7%
9. Dartmouth 51.1%

Some of the respondents to my original thread pointed out that the gap between the top schools is narrowing over time. That is absolutely true, although the order has not changed in a long time. Harvard and Princeton have ranked #1 and #2, respectively, on this criterion for over twenty years.

So the enrollment yield rates suggest that Harvard’s name still attracts high school students most effectively. Moreover, Harvard wins the direct competition for overlapping admittees with the other colleges at the top of the list. Tellingly, Harvard’s lowest head-to-head “winning percentages” are against MIT and Princeton.

But what of the two schools’ performance once the students arrive on campus? US News & World Report likes to focus on graduation rates as a reflection of institutional excellence. The magazine uses six-year graduation rates because all NCAA Division I member schools have to report this data in a uniform manner.

B. Six-Year Graduation Rates (Freshman Classes of 1993-1996, Graduating 1999-2002)
1. Harvard and Princeton 97%
3. Yale 95%
4. Brown and Dartmouth 94%
6. Stanford 92%
7. Penn 91%
8. Cornell 90%
9. Columbia 88%

Thus, six-year graduation rates do little to distinguish between Harvard and Princeton. I prefer four-year graduation rates which are not reported to the NCAA but are available for all the Ivy League colleges. I have no quarrel with students taking off a year or two away from school but a high four-year graduation rate reflects the confluence of several positive factors: (i) an extremely motivated student body, (ii) a school that provides educational programs which engage students and keep them engaged through graduation, (iii) generous financial aid policies, which minimize the number of students who take time off due to monetary constraints, and (iv) a supportive social environment, where students enjoy being there and want to graduate on time, with their peers.

C. Four-Year Graduation Rates (Freshman Class of 1998, Graduating 2002)
1. Princeton 92%
2. Yale 88%
3. Dartmouth 87%
4. Harvard and Penn 86%
6. Brown and Columbia 83%
8. Cornell and MIT 82%

Based upon this information, I would hypothesize for the time being that Princeton students are, on average and to a slight degree, happier to be there. To corroborate or refute this premise, I cross reference with alumni giving rates.

Some observers have rightly criticized alumni giving rates as an evaluative measure because they unfairly penalize public universities. That is likely true, as an in-state graduate of California-Berkeley or Michigan-Ann Arbor may feel as though he or she is already supporting alma mater by paying state income taxes. But the eight Ivy League schools, Stanford and MIT are all private universities so they compete on a level playing field. Every year, when they solicit alumni donations, they are essentially taking a poll of their graduates: What percentage feels that their experience on campus and the benefits of the degree after graduation are sufficiently valuable that they are willing to vote with their pocketbooks?

D. Alumni Giving Rates (2002)
1. Princeton 61%
2. Harvard 49%
3. Dartmouth 47%
4. Yale 44%
5. Penn 40%
6. MIT and Stanford 38%
8. Brown 36%
9. Cornell 34%
10. Columbia 32%

Finally, I make one more quantitative observation which admittedly borders on being qualitative or, at a minimum, anecdotal in nature. Since 1990, fourteen Harvard students have, tragically, taken their own lives. Over that same period, the corresponding figure for Princeton is zero (although, sadly, two Princeton graduate students committed suicide during that span). The Boston Globe conducted a study in 2002 which concluded that, of the nation’s elite universities, Harvard had the second highest rate of suicide on a per capita basis, after MIT.

Of course, it is stretching the train of thought to extrapolate from such small numbers. And, obviously, not every student is unhappy at Harvard nor happy at Princeton.

But I think the logic goes something like this: Nobody questions Harvard’s greatness in the literal sense of the word: the breadth of its educational resources and the prestige associated with the name. But Harvard’s very size and scope, and the fact that its glittering name compels many students to enroll without questioning whether there is a good fit with the individual – these ultimately conspire against it to a degree.

Harvard is famously demanding in terms of attracting the brightest students possible and then assuming that they are independent and motivated enough to take advantage of the resources scattered throughout Cambridge with relatively little in the way of support services. And for the bona fide geniuses in the world, that may be the right model for college. But based upon graduation rates and alumni giving rates and all the other quantitative data that I could find, I conclude that for the “average” high school super-achiever, Princeton is the best college in America.

By Asterix (Asterix) on Wednesday, May 26, 2004 - 09:56 pm: Edit

This year is going to rewrite the books for HYP with the shift in admissions stats, so last year's stats aren't really important now.

By Collegeguy (Collegeguy) on Wednesday, May 26, 2004 - 11:15 pm: Edit

Not bad

But I still chose Harvard over Princeton. Eating clubs are a real turnoff.

By Foreigngrad (Foreigngrad) on Thursday, May 27, 2004 - 01:17 pm: Edit

"The most conspicuous reflection of Harvard’s primacy is its considerable success in attracting students, as reflected in its enrollment yield rate" (Hoopitup)


Hoopitup,

searching myself for meaningful data, I came across this scientific paper from 2002 ("A revealed preference ranking of US colleges and universities" authors: Avery (Harvard KSG), Glickman (Boston U), Hoxby (Harvard) and Metrick (Wharton)).

http://www-1.gsb.columbia.edu/divisions/finance/seminars/micro/spring_03/Hoxby_paper.pdf

In my view, this is the most meaningful, not to say objective, analysis of relative preferences of American (?) students.

The authors claim: "... our ranking is unbiased and non-manipulable, unlike crude indicators of preference such as the matriculation and admission rates..." (for the statistical methodology read the paper)

By interviewing 3240 highly meritorious students, they analyzed relative considerations for 79 undergraduate institutions in the US and in Table 6 display the following ranking according to the following mean ratings:

1) Harvard 0.692
2) Yale 0.673
3) Princeton 0.601
4) Stanford 0.458
5) Brown 0.306
...
13) Swarthmore 0.007 (scores then become negative)
...

79) Indiana U -2.358

According to this study, which is more intelligent than all the other superficial statistics traded here, the following can be inferred:

a) H is the top choice for most high school seniors

b) HYP are perceived to be a league apart from the rest of the pack, S already trailing with some distance

c) Double admits will by tendancy pick H over P and Y over P (to a lesser degree H over Y), although the distance in the statistic is too small to infer this would systematically (nearly) ALWAYS happen.

By Foreigngrad (Foreigngrad) on Thursday, May 27, 2004 - 01:39 pm: Edit

"Eating clubs are a real turnoff" (Collegeguy)

Collegeguy,

I am not that literate about clubs at H or Y.

From what I heard, 'Porcelian' is rather exclusive. So is 'Skull and bones' at Y I heard. Most eating clubs at P are less exclusive in the literal sense of the word. I would really like to understand why the exclusivity of maybe 4 clubs at P (6 are bicker clubs altogether to my knowledge) taking in every third or even second aspirant should be more of a 'shame' to a college than the ones named above.

Or does it really make a difference in your pov, whether such groups of students own the HOUSES where they convene (for eating mainly)?

I would really appreciate, if somebody could explain to me why clubs at H are good (or not bad at least), while at P they are supposedly spoiling the qualitiy of the university (?)

By Asterix (Asterix) on Thursday, May 27, 2004 - 02:35 pm: Edit

Skull and bones is nothing like what is at Princeton. At Princeton they don't give a meal plan after two years, forcing students to pay extra money for eating clubs. They ARE exclusive, sometimes friends are split up. There are many articles about the process, some involve insane amounts of alcohol and others have an extensive interview process (like 10 interviews). You get all your meals there and it's a big part of social life as a result.

Skull and bones is a small, 15 person organization at Yale. It's difficult to get into, but it's basically a discussion group that meets a few times at week at most. It's not a HUGE part of the life there. Members still eat with everyone else. It's mainly for networking.

It's stupid to compare S&B with Princeton's eating clubs because they are so different. When I was at Princeton I heard one Princeton girl, as she was walking into an eating club say "We aren't exclusive, haven't you heard of the Skull and Bones at Yale?" This comment was completely ridiculous - it's apples vs. oranges.

One reason I chose Yale over Princeton was that Princeton's eating clubs, a main component of social life, ARE exclusive. I don't want to be split from my friends and have to pay an extra couple of thousand for more limited dining options. No one cares about the S&B at Yale, it's just a small thing that, unless you are involved, doesn't influence you.

By Deferreddude (Deferreddude) on Thursday, May 27, 2004 - 10:23 pm: Edit

"HYP are perceived to be a league apart from the rest of the pack, S already trailing with some distance"

Hmm....I disagree with that. The gap is not HYP--gap--S. In reality, the gap has always been: H--gap--YPS. Stanford is definitely not trailing behind Yale or Princeton in prestige, but Harvard has always been Harvard and it stomps out Princeton and Yale any day. It may not necessarily be a superior school, but there is a MUCH biiger gap between Harvard and Yale, Harvard and Princeton, and Harvard and Stanford than there has been a gap between Yale and Stanford or Princeton and Stanford.

By Foreigngrad (Foreigngrad) on Friday, May 28, 2004 - 02:10 am: Edit

"but there is a MUCH biiger gap between Harvard and Yale, Harvard and Princeton, and Harvard and Stanford than there has been a gap between Yale and Stanford or Princeton and Stanford" (deferreddude)

Deferred...,

The so-called Hoxby paper (see above) on the issue, which nobody requires you to read of course, talks about gaps in preference of high school seniors, regardless of their actual applications, thus measuring 'hidden' cross-preferences.

According to their data, the gap between H and Y is marginal, and the gap between P and S twice as big as the 'gap' between Y and P.

Whatever you are talking about, it is something else than this scientific study is. Nobody contests that H is the top dog in popularity, but the popularity gap to Y is very limited and to P is still minor.

qed

By Deferreddude (Deferreddude) on Friday, May 28, 2004 - 06:18 am: Edit

Foreigngrad, surely you don't think that a sample of "3240 meritorious high school students" is more reliable that the true sample of ALL admitted people in the aggregate.

This is the yield rate for ALL high school students that were admitted--not just the 3240 students picked out for the sample:

A. Enrollment Yield Rates (2002)
1. Harvard 78.8%
2. Princeton 73.4%
3. Stanford 69.2%
4. Yale 63.6%

Stanford apparently beats out Yale, and Harvard beats Yale by a longshot.

By Icansmile4u (Icansmile4u) on Friday, May 28, 2004 - 08:57 am: Edit

The very interesting thing about Foreigngrad's dataset Hoxby paper is that it actually puts Princeton on "4th" position without price consideration. and it's ranked "3rd" with price consideration. Foreigngrad obviously didn't mention even a bit about the condition assuming nobody will bother to read the 48 pages long paper. Typical example of "data manipulation" by deliberate omitting. Even, he leaves this psychological trick in his post "The so-called Hoxby paper (see above) on the issue, **which nobody requires you to read of course**)

1.Harvard 2.199
2.Yale 2.112
3.Stanford 2.052
4.Princeton 1.992

This is the actual dateset from Hoxby paper.
So the correct argument after revising ") Foreigngrad's "HYP are perceived to be a league apart from the rest of the pack, S already trailing with some distance" is

"HYS are perceived to be a league apart from the rest of the pack. (P didn't crack 2.00 ), P already trailling with some distance, and P only manages to join the league with the back-up of its financial aid policy."

By Foreigngrad (Foreigngrad) on Friday, May 28, 2004 - 02:04 pm: Edit

Icansmile4you,

I posted a link, so everyone interested in the paper could catch up with Hoxby's (et al.) arguments. I consider Table 6 the "final" ranking of that paper, not table 3, if only for the reason that money DOES matter in life. Read pp.30 f.on their comments on that one. You better be careful insinuating 'manipulation' based on the fact that I did not discuss at length all the INTERMEDIATE analyses of a study here on a board.

But while we are at it: if you think, table 3 is really the most meaningful ("the 'actual' dataset"...), then please note that the variance is significantly less between ALL schools than in table 6 - table 3 sure still has a ranking, but no real (relative) gaps between schools as table 6 does - so with regard to table 3 on which you chose to dwell, your statement that "HYS are perceived to be a league APART from the rest" is plainly false and easy to check for anybody looking at the WHOLE ranking and its intervals.

Btw, deferreddude still obviously did not care to read the paper, as he did not understand that Hoxby tried to take preferences into account of people who did not apply to a particular institution in real life (which is why the authors - I repeat myself - consider their analysis as less biased than one-dimensional statistics such as yield figures). Furthermore the composition of the sample chosen by Hoxby et al, btw does not differ much in quality from the composition of the pools of accepted at HYPSMetc.- so much for his flawed argument that Hoxby's sample did not represent 'all' students.

Deferreddude, relax, no matter by what statistic, H is leading the pac, nobody contested that!

By Icansmile4u (Icansmile4u) on Friday, May 28, 2004 - 03:23 pm: Edit

"If only for the reason that money DOES matter in life" haha sure. What an argument. Yeah I admit money matters in real life. However, when you talk about a school's educational quality from both social and academical scopes, do you take "how much I can squeeze out of this institution" into consideration? probably not. When you evaluate a school's prestige, does the amount of financial aid it gives matter? probably not.

The idea is that for the statistical purpose (which is already quite far from reality), money doesn't have as much as bearing as you think. The given statistics itself is unrealistic (sample of 3000 students probably)? and you seek 'realistic' factor in it? That's quite an irony.

Table 3 has a ranking which DOES have "RELATIVE" gap you are referring to. When you substitute Harvard 2.199 as 100%, you get 96%, 93%, and 90% for YSP order. That actually DOES tell something about 'RELATIVE" mean value of preference you are talking about.

By Icansmile4u (Icansmile4u) on Friday, May 28, 2004 - 03:32 pm: Edit

Let me give you a note from that article word by word. "The table(which is table 6 you support) shows the "price-considered"ranking of colleges and universities. That is, we ***have not***(key word) attempted to control for the effect of variables that might influence admissions or matriculation. but which are ***not***(another key word) indicators of college quality or student merit."

Do you comprehend what is meant by this sentence at all?

Since this thread is specifically on the preference of college-bound seniors interested in the "overall college experience," we actually have to pick a dataset which attempts to reduce "unrelated" variables. To serve that purpose, Table 3 actually does a better job of reducing statistical bias.

Plus, if we need a preference order with "price-consideration," we just look at yield rate as Deferrdude provided. Why bother to quote another source?

By Anonymous1989 (Anonymous1989) on Friday, May 28, 2004 - 05:53 pm: Edit

well since this year nobody on the yale waitlist got accepted and ppl from harvard did...maybe there's gonna be a difference this year with the yield.. *shrugs* not like any of this matters...

By Jessc (Jessc) on Saturday, May 29, 2004 - 08:43 pm: Edit

Statistics that Mean Nothing
Harvard 548678.98876
Yale 45362.987
Princeton 766544333.99876

By Foreigngrad (Foreigngrad) on Saturday, June 05, 2004 - 11:22 am: Edit

To correct some misperceptions here (quoting Icansmile4u in sequence):

“Let me give you a note from that article word by word.”

Icansmile4u, it is good common practice to properly reference quotations, no matter what school you go to... This allows others to check the originally intended context of a quote. (Icansmile4u was quoting the footnote of table 6 on page 47 of the Hoxby paper in this case).

"The table(which is table 6 you support) shows the "price-considered"ranking of colleges and universities. That is, we ***have not***(key word) attempted to control for the effect of variables that might influence admissions or matriculation. but which are ***not***(another key word) indicators of college quality or student merit." (i4u)

It is not me who 'supports' table 6, but the authors offer table 6 in the said study as their last ranking. If table 6 did not matter at all in the context of the Hoxby paper, there would be no point for the authors in establishing tables 4 and 5 (explaining differences in ranking between tables 3 and 6), let alone publishing table 6.

”Do you comprehend what is meant by this sentence at all?” (i4u)

I am not sure, you do... For your enlightenment, here is a definition of the (undiscovered by you) ‘key word’ you seem to have trouble coming to grips with:

>>> "Controlling" for something in statistics – Definition:

The overall relationship is seldom the main focus when you have more than one predictor variable in the statistical model. Instead, these models are used to address a much more important question: what is the effect of something when we take into account something else? It's such an important concept, statisticians have some jargon for it: they talk about controlling for something, adjusting for something, or partialing something out. For example, what is the effect of sex on weight, when we take height into account? Think about it. Boys are heavier than girls, but boys are taller than girls, and taller people are heavier, so if we take into account the difference in height between boys and girls, is there any "real" difference in weight between them? A trivial question here maybe, but not if your outcome variable is an athlete's performance or health, and you control for time spent training before you look at the effect of sex or sport or region or whatever. And of course, it's also important to know about the effect of training on performance or injuries when you take into account differences between sexes or sports or regions. <<<

Coming back to Hoxby’s table 6: “not controlling for the effect of variables that might influence … matriculation” means nothing else than NOT filtering out the effects of such influencing factors collectively called “price” by the authors (although anybody who looked into table 4 could see that this does encompass more factors than just net tuition ). In other words: Hoxby et al. think that these variables “not controlled for” in table 6 (=not adjusted for) DO influence matriculation decisions (I am dwelling on matriculation, because that is what was discussed here, while Hoxby takes into account the admission side as well (see p. 10, inter alia: “We want to use students’ revealed preferences to rank colleges AND colleges’ revealed preferences to rank students…”). This is exactly what I said beforehand.


”Since this thread is specifically on the preference of college-bound seniors interested in the "overall college experience," we actually have to pick a dataset which attempts to reduce "unrelated" variables. To serve that purpose, Table 3 actually does a better job of reducing statistical bias. “ (i4u)

A couple of things have come to be discussed in this thread. The context in which I quoted the Hoxby paper, was the mere matriculation preference theme, which has been discussed at many times in various threads in this forum. When college students decide among colleges they have been admitted to, they not only decide on the basis of “pure (abstract) desirability” (which – I think there is no disagreement – is rendered by table 3) but also on the basis on a couple of other “real” factors pictured in table 4. The resulting ranking is shown by Hoxby in his table 6. In table 6, as I said, you have asignificant relative gap between Y and P, and even more so between P and S in the HYPS order; in indices: 100, 97, 87, 66. The gaps are significantly more than in the order of indices in table 3, rendering a HYSP order of indices 100,96,93,90 as to your “gap” argument”. I would not necessarily call a distance of 3% a real relative “gap”, if you comprehend the statistical implication…

”Plus, if we need a preference order with "price-consideration," we just look at yield rate as Deferrdude provided. Why bother to quote another source?” (i4u)

If you want to discuss Hoxby et al, you ought to start by reading their paper. The authors explicitly contend that matriculation rates (as they call it) contain systematic bias in comparison to their ranking.

Without going further into a scientific study for people who seem to lazy to really read it: 3000 is generally not too small a sample for the purpose of the study. You might want to take statistics 101, if you care to find out why…

By Hoopitup (Hoopitup) on Tuesday, July 13, 2004 - 07:48 pm: Edit

Thanks to all respondents for the very interesting posts. Professor Hoxby's study is a very creative, not to mention labor-intensive, attempt to measure the desirability of various colleges. His analysis, though well designed, is constrained by the fact that his sample size is, by necessity, limited.

Hoxby's survey of 3240 college matriculants is spread among 79 universities attended. His data is a small subset of the actual students who had the enviable choice of selecting between two or more top colleges.

Listed below is the data for ALL students who were choosing between Hoxby's top ranked college, Harvard, and four of its closest competitors.

A. HARVARD COLLEGE ENROLLMENT YIELD RATE BY CROSS ADMITTED SCHOOL (Class of 2006, admitted in 2002)

1. MIT 55%
2. Princeton 73%
3. Stanford 74%
4. Yale 84%

That is, of students admitted to both Harvard and MIT who eventually enrolled at one of those two, 55% percent selected Harvard. Similarly, of admittees choosing between Harvard and Princeton, 73% opted for Crimson over Orange and Black.

This data reveals that (1) Harvard has a "winning" enrollment percentage against each of its top competitors, and (2) Harvard's greatest "winning" percentage is against Yale.

This data is from the 2002 admissions cycle. As mentioned at the beginning of this thread, Yale in 2004 for the first time passed EITHER Harvard or Princeton in achieving a lower admissions rate, although early matriculation data suggests that the relative enrollment yield rates will be in the same order that they have been for over two decades, namely (1) Harvard, (2) Princeton and (3) Yale. However, this year the distribution will almost certainly be tighter.

Recognizing that MIT with its engineering-focused curriculum does not compete as broadly with HYPS in the same way that those four compete with each other, Harvard's "winning" percentages against Princeton, Stanford and Yale are in the same order as their overall enrollment yield rates.

B. ENROLLMENT YIELD RATES (2002)
1. Harvard 78.8%
2. Princeton 73.4%
3. Stanford 69.2%
4. Yale 63.6%
5. MIT 56.7%

So the Harvard-specific results in 2002 were consistent with data from the universe of all students admitted to Harvard, Princeton, Stanford or Yale, as might be suggested by the law of large numbers.

By Hoopitup (Hoopitup) on Friday, July 30, 2004 - 05:56 pm: Edit

There is a thread on this message board about whether Harvard students are unusually unhappy or merely typical of ambitious, high achieving college students. I agree that undergraduates at Ivy League and other elite schools are probably more anxious in general than those at less selective colleges.

Nevertheless, as pointed out at the beginning of this thread, Harvard averages 1.5 suicides PER YEAR. Meanwhile, Princeton has had one undergraduate suicide IN TOTAL over the past 30 years. That seems to be a statistically significant difference to me.

Therefore, I have a difficult time accepting that Harvard unhappiness is just "typical" of goal-oriented undergraduates. On average, it must be the case that either Harvard students are unusually unhappy or Princeton students are unusually happy. Maybe both.

By the way, I was not able to find statistics on Eli suicides by searching the Yale Daily News. It seems to me that The Harvard Crimson and The Daily Princetonian are much more willing to air their campuses' dirty laundry. The Yale Daily News reads more uniformly positive.

By Signup (Signup) on Friday, July 30, 2004 - 10:58 pm: Edit

People love to hate Harvard--you guys are so funny.

By Jazzyjeff (Jazzyjeff) on Thursday, August 19, 2004 - 05:04 pm: Edit

Does anybody know when US News & World Report will release its 2005 ranking of America's Best Colleges? Thanks.

By Hoopitup (Hoopitup) on Tuesday, August 31, 2004 - 05:39 pm: Edit

In Hoxby's study A Revealed Preference Ranking Of U.S. Colleges and Universities, the term "price variable" refers not solely to "price" in the literal sense.

"Price" as used by Hoxby means not only tuition paid but also any other independent variable besides the school itself. In this context, "price" includes whether the student's father or mother is an alumnus/a of the university, how far the college is from the student's hometown, etc.

Since the purpose of the study is to rank the universities themselves, one would always attempt to control for these other extraneous variables. Thus, Table 6, the "price-considered" ranking, is a more precise list than Table 3, the "price-not-considered" output.

By Baghdadbob (Baghdadbob) on Tuesday, August 31, 2004 - 08:08 pm: Edit

Cetirus Peribus.

By Hoopitup (Hoopitup) on Wednesday, September 22, 2004 - 08:52 am: Edit

Ceteris paribus, indeed.