<p>Yes, let’s do look at those schools. They are succeeding perhaps better than other small schools, but being small definitely weighs in their favor and renders comparison to huge schools in PhD productivity pretty much meaningless. Nine of the ten have undergrad enrollments of fewer than 5000. The top four are all smaller than 2000, and the top two have fewer than 1000 undergrads each. That’s not a random coincidence.</p>
<p>This is a clear example of the effect of small size in per capta analyses. Even research/PhD powerhouses such as Berkeley do not stand a chance in per capita comparisons simply because of their huge enrollments. Consider hypothetical extreme examples to illustrate the point: Tiny College has an enrollment of say 10 students. And they are smart, motivated students so 2 of them go on to get PhDs. Mondo University has an undergrad enrollment of say 35,000 (and there really are schools this big). Those students are also smart and motivated. But no matter what the “academic quality” of the school, to beat Tiny College in a per capita comparison it would need to have 7001 of those students go on to get a PhD. That’s completely unrealisitic no matter how high Mondo’s quality. That’s about the same number of PhDs the entire US produces in a year. Mondo U would have to swamp every PhD program in every department in every school in the US with its graduates to win this “contest.” </p>
<p>If you want to rig it in favor of small schools, analyze per capita. If you want to rig it in favor of large schools, analyze total numbers.</p>
<br>
<br>
<p>What do masters degrees have to do with a comparison of PhD production?</p>
<br>
<br>
<p>This I agree with. What percentage do not do very well is measure relative “academic quality.”</p>
<p>The 10 listed schools all offer small average class sizes. All are very selective. The one liberal arts (non-technical) university in the bunch (largest of the 10) has an exceptional undergraduate faculty, one of the best research library systems on the continent, and stiff core requirements. Reed has unusually rigorous senior thesis and comprehensive exam requirements. Swarthmore has its honors program with outside examiners. None of the 10 have big-time sports or fraternity scenes. More than half of them (if not all) have reputations as “hard” schools with serious intellectual atmospheres. </p>
<p>Now, let’s say we took 500 rising freshmen headed for Carleton, or 1000 headed for Chicago, and enrolled them instead at a large flagship university (one of the tops for PhD production, unadjusted per capita). Would the switch have no effect on the per capita PhD production for just that set of students? Maybe not. Berkeley has good professors, good library systems, and (I assume) tough testing. It also has more huge lecture classes and more extracurricular distractions. So do other large state universities, and none of them are as selective as Berkeley. </p>
<p>Admittedly, it is hard to fairly compare LACs with large universities on this score. The per capita numbers apparently do not isolate the arts and science divisions of the larger schools (which often have engineering, business, nursing, and other pre-professional programs). The small private schools also are more likely to attract more affluent students who can afford to spend years earning a PhD.</p>
<p>I think warblersrule nails it on this one. PhD production isn’t a meaningless figure, but it’s not a proxy for academic quality. Look at the schools on that top 10 list; 7 of the 10 are LACs,3 are top STEM schools, and one is the University of Chicago which in some ways at the undergraduate level is a LAC on steroids. They’re all quality academic programs, but quite narrow in the range of what they offer at the undergrad level. Undergrad business majors by and large don’t get PhDs. Pre-law and pre-medicine students by and large don’t get PhDs. Most engineers at even the very top engineering schools don’t get PhDs, except the engineers at places like MIT, Caltech, and Harvey Mudd, many of whom do go on to get PhDs and become engineering professors, and generally make less money than their colleagues who work in the private sector. Very few education majors get PhDs. Very few people in the fine arts and performing arts get PhDs. Generally speaking, the broader a university’s undergraduate program offerings, the lower its per capita PhD production.</p>
<p>The point about PhD program quality is also a valid one. Yes, it’s impressive to get into any PhD program; it’s even more impressive to complete one. But it’s far more impressive to get into a PhD program at a Harvard, Yale, or Stanford than, say, the University of Kansas. Some schools that rank high in the PhD productivity might be producing a lot of PhDs from mediocre programs whose job prospects are actually quite poor, while a Harvard, say, might be producing a lower percentage of PhDs overall but a higher percentage of PhDs from top programs, along with a higher percentage of JDs, MDs, and MBAs from the top law, medical, and business schools.</p>
<p>Why the linkage between PhDs and professors? Most PhDs work in their industry and are not professors. Consider how few professors retire, compared to the number of PhDs produced, per year. Consider how many PhD earners a professor mentors in her/his lifetime. </p>
<p>In any given field and work setting (college or private sector), PhDs generally make more than BA/BS/MA/MS holders. Clearly professor PhDs are not there because of the money! And why the mention of money at all?</p>
<p>Except MIT. Over half the boys and over a third of the girls at MIT go Greek.</p>
<br>
<br>
<p>Yes! This is exactly my point.
You might reasonably conclude that Swarthmore was of higher or lower academic quality than say Occidental, or Pomona, or St. Olaf based on on PhD production. But comparing Swarthmore to say Berkeley, UCLA, or UT Austin on the basis of PhDs is just too much of an apples and oranges deal - actually more like a cherry and watermelon. </p>
<br>
<br>
<p>Vonlost - you appear to be talking about PhDs in general whereas bclintonk is referring specifically to the engineering field. And I agree with him. In my company we hire a lot of engineers, chemists, and biologists - all to work in R&D. Somewhere between a quarter and a third of the chemists and biologists we hire hold a PhD. But almost none of the engineers do. It’s just not needed to get to the top of the field of practicing engineers.</p>
<p>What does your company do? I’d be very surprised if a company that makes computer chips or airplanes has mostly engineers with bachelor’s degrees doing research and development.</p>
<p>^^The company makes medical electronic instruments. And I didn’t say that most of the engineers have only bachelors degrees. I said that almost none of them have PhDs. Many of them do have masters degrees in engineering.</p>
<p>I don’t think that what your company does generalizes to all companies that hire engineers. I know at least that companies that do electronic circuit design and manufacturing hire engineering PhD graduates for research and development. Companies that sell computer software and systems hire computer science PhDs as well. I’m pretty sure that aerospace and defense companies hire PhDs too.</p>
<p>Well, I wasn’t just talking about PhDs in engineering. The overproduction of PhDs is a well-known and chronic problem in almost every discipline. Engineering and other STEM PhDs can often find private sector jos, but most of those jobs don’t actually require a PhD; often the PhDs are overqualified for the jobs they actually get. In the humanities and social sciences, the situation is even more dire. What is a PD in English literature or sociology actually qualified to do, except to teach at a college or university? And how many of those jobs actually open up? For many, probably most PhDs, it’s a dead-end path.</p>
<p>I highly value higher education. I value the PhD degree. I even more highly value a PhD from a top graduate program that leads to a reasonable prospect of a permanent academic position. But apart from that small, rarefied world, why would anyone need, or benefit from, a PhD from a middle-ranking university? If it doesn’t lead to an academic career track, it’s probably unnecessary.</p>
<p>Here’s a fairly recent story from The Economist:</p>
<p>I don’t consider the biomedical (biotech/pharmacuetical/medical instruments) industry that I’m in to be all that small and rarefied. It employs many thousands of scientists. But as I said we commonly hire a lot of PhD scientists - including me. I earned mine at UC Davis, which I suppose is a middle-ranking university. Yet it has made all the difference in my career. The most that non-PhD scientists (not engineers) in my field can aspire to is to be a senior technician or alternatively leave R&D and go off into Marketing, Manufacturing, or Quality Control. Not all PhDs rise to be head of an R&D department, but very few R&D department heads don’t have a PhD, although a “middle-ranking school” PhD is perfectly acceptable.</p>
I would like to point out at this moment in time that there is no evidence that universities with high PhD per capita are compelled to attain PhDs because they are low income earners.</p>
<p>What dubious career surveys we have available often point to high PhD productivity colleges, such as CalTech, as the most successful workers in the nation.</p>
<p>
Yes, the per capita PhD production for the set of students would plummet.</p>
<p>People overemphasize random alternative factors and completely overlook the effort colleges, such as CalTech or Reed, put in order for their students to succeed. Colleges with high PhD productivity have a number of aspects in common. </p>
<p>For example, they ensure that students partake in a healthy degree of undergraduate research and they make the utmost effort to make such research widely available. This is what other top colleges, such as HYP, fail to offer and what publics flag.</p>
<p>That’s so false - research is far more readily available to students at HYP than at almost any university (I’d say Caltech and MIT are about the same). They have the money to throw research grants at undergraduates, always under the tutelage of a professor. Princeton even requires its students to do a thesis and a junior paper.</p>
<p>IMO it’s all about the type of student that the college in question attracts. It also has a lot to do with the proportion of certain types of majors at the college.</p>
<p>I can’t speak for the other schools on the list in the OP, but Reed at least provides information about where its alumni go to grad school.</p>
<p>The graduate schools most frequently attended by Reed alumni are UC Berkeley, U of Washington, U of Chicago, Stanford, U of Oregon, Harvard and Cornell, in that order. While overall school reputation is meaningless in grad school, where department strength is everything, I’d say most of these schools are quite good at quite a few things (and I suspect Washington and Oregon are so high on this list mainly because of their locations). You can see the full list, along with the statistics for JD, MBA and MD-granting institutions, here:</p>
<p>I have absolutely no reason to believe Swarthmore, Caltech, Carleton and MIT’s grad school placement is any worse–in fact, I wouldn’t be surprised if it was even better.</p>
<p>I don’t understand where the idea that a high PhD production rate = joke PhDs comes from.</p>
<p>Has anyone mentioned that “proxy”, in the thread’s title, is used incorrectly? Proxy may be a person or a document assigning proxy, neither of which applies in this case. Try using ‘substitute’.</p>
<p>Interesting study. However, I’d wonder whether people who CHOSE to go to those schools weren’t already more academic to begin with. In other words, people who are most likely to get a Ph.D. are also more likely to choose a nerdy, academic school like Bryn Mawr over a party school like Radford University or Longwood University. Therefore, regardless of where they go to school, these individuals are more likely to wind up in graduate school. Bryn Mawr doesn’t MAKE you into the type of person who is likely to go to grad school, nor does it necessarily prepare you better to go to grad school. It just attracts more people who are likely to do so.</p>
<p>Ghostt, that link doesn’t prove your point. “Most frequently attended” is ambiguous - it doesn’t necessarily mean that those schools make up a huge portion of those who go on to get a PhD from Reed, but could simply mean that those are the schools who happen to have a higher number despite making up a small portion of those who go on to get a PhD from Reed. So, in theory, the data from your link could be the same even if 90% of the schools that Reed sends students to have unknown PhD programs, e.g. for each unknown PhD program, there were 1-2 students from Reed, while the last 10% (the well-known PhD programs) had say 5-6 students each from Reed and thus make up the top X most frequent PhD programs that Reed sends students to. So under this, the top X schools together claimed, say, 20% of the students who went on to get a PhD, followed by a very long tail of unknown PhD programs, each of which take only a student or two, cumulatively accounting for the other 80%.</p>
<p>In other words, “most frequently attended” could be interpreted as just “the most popular,” rather than your intended interpretation “the most common.” This distinction is important because the topic at hand is the proportion of students who go on to PhD programs, and further the proportion of those who are going to respectable PhD programs. Your data doesn’t tell us anything about the proportion. I’m not saying that your overall point is wrong - it very well could be that both Reed and other colleges do send most to respectable PhD programs. I’m just saying that the data from your link doesn’t support that conclusion. It wouldn’t surprise me, actually, to find that Reed stated it ambiguously on purpose.</p>
<p>It says that it’s “BASED ON THE 2010 ALUMNI DATABASE” - if that’s publicly available, perhaps someone with the time and interest could see just what the proportion really is.</p>
