« On Not Having an Advance Abyssal Plan | Main | Formative Youth »

February 24, 2009


As a student teacher I can answer the first question: homework, quizzes, and tests are personally identified to let the teacher know which students need help with what concepts. Their purpose, especially the first two, is to determine where resources (e.g. tutoring) should be allocated to give all students a grasp of the material. For the SAT and GRE, on the other hand, it's assumed that the test-taker is "done" learning the material and the only question is how well it's been learned.

At the college level, most of my professors graded exams and papers by SSN instead of name. Actually, it was usually TAs doing the grading, so even if we'd used our names they would be meaningless to the grader. The exceptions were all small classes with less than thirty students, where the professor graded all exams herself.

Law schools practice anonymous grading for the key first year and award major honors and high-status positions based on the results.

Robin, you missed something else important that high-status schools are selling--the ability to socialize and network with the kind of people who attend high-status schools. Of course, this applies only to people getting "practical" degrees (law, business, engineering, &c.) who intend to go into industry.

Compared to big state universities, many liberal arts colleges charge much higher tuitions and are harder to get into. Yet professors at large state universities on average have far more academic prestige than those at liberal arts colleges. For liberal arts colleges at least soulless is right and students are buying the "ability to socialize and network with the kind of people who attend high-status schools."

I suspect that most of the students who end up going to top research universities also applied to top liberal arts colleges. If so than in general undergraduate students are more concerned about the quality of their classmates than the status of their professors. And since top schools don't give any weight to appearance in admissions, a major driver of status for 18-year-olds, college students care about the quality of their classmates' minds not their classmates' overall status.

Sideways, at the college level profs almost never allocate tutoring, and even if they did they could just allocate it to the anonymous identity that was having the problems.

Soulless, yes schools also sell affiliation with other high status students.

all of the above also explain why we in the UK spend so much money on sending our children to traditional english independent schools.

"all of the above also explain why we in the UK spend so much money on sending our children to traditional english independent schools."

Yes, the key advantage prep and public schools yield for parents is the ability to choose their children's peer-group.

With the letters of recommendation I think people are looking for some sort of human connection. Some proof that a candidate isn't just insanely good at working the system.

I am given to understand that IQ is the strongest predictor of academic and career performance, and yet IQ Tests are usually *not* required or even accepted by Universities/grad programs. Is there a good reason for this? Or is caring about native intelligence academically unfashionable?

Good post as usual.

Generally speaking, I would agree with Robin, that status influences human decision-making to a large and unexpected degree.

However, there seems to be the implied notion that status has little or no value. Status clearly has value. A lot of it in fact, but it is not easy to measure in pure economic terms. Maybe people are paying for this status on the basis of that perceived, yet yard to quantify, value. Every person I know who went to a prestigious school like Harvard, for example, did not go there for the education value alone. In fact, that was not one of the primary criteria they remotely looked at. They went there mainly because of the status benefits of having gone to a prestigious school, and in the knowledge that these status benefit could be converted into economic advantage when they went into the job market later on, and in life generally.

So my issue with Robin’s view on status generally, is that he seems to imply paying for status is unwise or irrational. But that is not always true. In several instances, it could be wise depending on the value of the status and how that value can be converted into other benefits (economic or otherwise) in the future.

I am not aware of any studies that have attempted to measure the economic equivalency or value of status. And it is unclear to me how this could be meaningfully done. (Would appreciate the posting of any studies that you are aware of). But I believe status is becoming an increasingly important economic consideration these days. Some say that the world is changing from a “utility economy,” where purchasing decisions are based on a products physical benefits, to an “aesthetic economy,” where a products value is derived from the status benefits it provides. For example, when was the last time you bought an item of clothing purely for the raw utility of keeping warm and/or providing protection? My expectation is that how it made you look and feel were the primary decision criteria in theat decision. The net result is that products that provide status, clearly have, and should have a status premium associated with them.

Russ, I don't mean to imply status has no private value, nor any social value, but its social value does seem likely to be less than its private value, so we should expect too much effort to gain it.

The "docs" link is broken, it starts with "ttp"

I agree, when it comes to selecting universities, people decide in large part by the desire to be associated with prestige.

But I think there is a different, sensible approach that gives the same outcome. I think there is good reason to believe that one tends to conform to the standards of those around one. If one surrounds oneself by talented, driven people, one will find greater motivation to develop ones own talents. Furthermore, one will learn from said people and develop valuable relationships.

So, given a university is going to attract many very talented students, I think it makes a lot of sense for other talented students to want to attend, even if there is largely no difference in quality of teaching etc, just to be in that environment.

Then all that needs explained is why one school attracts more talented students than another, in the first place. I think there are a number of sensible descriptions in terms of history -- involving not so much "prestige" in the sense referred to above as much as "security" in the sense that one doesn't fear the institution will collapse in the near future. The first school gets an advantage over the second, due to this measure, and that advantage is solidified as future generations follow the lead of those before them.

Also, although the quality of faculty might not be important for most undergraduates, it is important to some. Exceptionally brilliant students who want to become researchers will benefit from faculty expertise from opportunities to work in the best labs. This is a small kernel of the undergraduate body, but it could also be sufficient to create the advantage that is solidified as everyone strives to flock around the existing clusters of talent.

Note that the incentives for choosing a graduate school are different than an undergraduate school. Graduate students are often virtual school employees, and they can work directly with professors on research. Additionally, graduate school level knowledge is frequently specialized to the point where the professor's domain knowledge really matters. (For example, anyone with a technical background and some skill at teaching can teach freshman physics. On the other hand, not many people with teaching skills understand general relativity well enough to teach that.)

But, yes, college is about the people you go to school with and about impressing people with the school's name, not the content of the courses. (Incidentally, I chose which school to go to by rolling a d20.)

I agree that there really is a threshold effect for school -- if you attend a major research university -- and there are probably at least 100 good ones in the U.S. -- you are going to get a decent education if you want one. What you are buying by going to MIT or Stanford is prestige, networking with other high status people, and the possibility to start a career in academia. The latter is the only real contribution to education. The other are more just resume filtering features.
Gladwell makes this same point in Outliers.

"I am given to understand that IQ is the strongest predictor of academic and career performance, and yet IQ Tests are usually *not* required or even accepted by Universities/grad programs. Is there a good reason for this? Or is caring about native intelligence academically unfashionable?" -Anon

The GRE is highly correlated with IQ test scores, but doesn't have the same stigma.

To expand on what Jayson Virissimo said...

IQ, and the assumptions the concept (it is assumed) carries of inborn "general intelligence" are not so much academically unfashionable as politically radioactive, largely because of issues surrounding correlations between racial background and measured intelligence--African folks score poorly while Asians and Jews (hi, Eliezer!) score well and really, nobody wants to even consider dealing with it to the point that they will aggressively ignore the entire concept of IQ and anything that sounds like it. However, given the historical use of culturally-biased "intelligence" metrics to prop up pseudo-scientific racism, the touchiness is understandable, if regrettable.

Therefore, any highly g-loaded test that doesn't look like it's measuring intelligence is much, much safer.

However, given the historical use of culturally-biased "intelligence" metrics to prop up pseudo-scientific racism, the touchiness is understandable, if regrettable.
Was that actually the case? I know there was supposed to be some question on an SAT about regattas and oarsmen, but I don't know if I've yet heard of a test that was found to have poor predictive validity for certain cultures it was intended to test. The Nazis actually prohibited intelligence testing, so I think racism can hold its own quite well without such tests.

In the U.S., at least, use of IQ testing as a hiring criterion was found by courts to be illegal discrimination against minorities and thus a violation of the Civil Rights Act. (Or so I heard; I could be wrong.)

Emotional intelligence is an even better indicator of success in school.

The public knows that "prestige" schools have far more applicants per available positions. The admissions departments at some top schools claim that their entire entering class could be chosen from applicants with perfect SAT scores, if they so desired, or with valedictorians only. Consequently, attending such a school tells everyone that you are smart. Medical school was this way during the Viet Nam war era, when most schools had hundreds of applicants for each spot, down to the military deferment that accompanied admission. Today, many medical schools accept over half their applicants, so this is another reason that medicine's prestige factor is dropping.

@retired urologist: Medicine has defintiely lost its prestige, but I don't think its due to lack of competition to get into medical school. US News still reports low single digit admission rates to top medical schools (< 5%), and I can't somehow seeing it magically being 50% even for the bottom 25 programs, just because many people do wind up going to the Carribean. In my limited experience, at least part of the problem is due to health care economists who devised the lovely reimbursement scheme that devalues physician time (and hence the physician-patient relationship) and over-values marginally useful procedures. Of course that's only part of the problem. Then on top of things, those brilliant economists placed well thought out incentives to make sure the proceduralist doesn't follow the patient he/she intervened on because it isn't "economic".


Here are the stats for all USA med schools for 2007. The numbers are by state of residency, rather than by school, but almost half of the applicants get in somewhere. I think the confusion about the high applicant to matriculant ratio is based on an average of over 13 applications per student. Since each can only go to one school, that means that an average of 12 applications per student go for naught, even if the student is accepted.

USA: 42,315 applied, 17,759 matriculated (42% got in somewhere)

Residents of West Virginia had highest rate of matriculation: 58%
Residents of New Hampshire had lowest rate of matriculation:33%

National ratio of applicants to matriculations: 1.4:1
LA (my current state of residence) ratio of applicants to matriculations: 0.95:1 (more likely to get in than not)

Here's the Web site.

There are no published statistics that I could find that allow us to say how many applicants actually are accepted at the "top" medical schools in order to reach their capacity. We only know how many applied and how many matriculated at that school. In top undergraduate schools, it is not unusual for only 10% of the accepted applicants to attend that school. With the tuition at state medical schools typically being 10% -15% that of top private schools (for essentially the same education and certainly the same union card), I suspect many of the qualified applicants to the top schools don't actually plan to attend even if they are accepted.

Note: Medical College of Georgia $4334; Harvard $39,800 (although actual students tell me their obligation is closer to $60,000).

Wrong right/write, second sentence of the last paragraph.

The comments to this entry are closed.

Less Wrong (sister site)

May 2009

Sun Mon Tue Wed Thu Fri Sat
          1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30