<p>When adjusted fairly, all major colleges and universities save a handful will produce an almost identical ratio of PhDs. This entire thread is pointless.</p>
<p>While I agree that these statistics aren’t completely useless, it is very hard to tell just how useful they are when there’s no discrimination among what PhD programs that these schools are sending their students to. No matter the selectivity of a PhD program, there are many high-quality ones and many low-quality ones. This lack of discriminatory power in the interpretation of these statistics seriously hinders their usefulness.</p>
<p>
</p>
<p>Princeton requires its students to do a senior thesis, in addition to 1-2 junior papers, all of which are independent research. Princeton has tons of top graduate programs that are very research-intensive. And Princeton spends tons of money on research, much of it on undergrads. So for all logical causes of PhD productivity, Princeton does extremely well. Yet it’s #10, and its rate of production is virtually indistinguishable from the next large set of schools that follow it.</p>
<p>I think Alexandre’s comment is right: controlling for certain factors, most major colleges are going to be about equal in their PhD productivity.</p>
<p>
This is what is pointless about the numbers. Caltech is the only one that stands out from the crowd. Otherwise forget the numbers and the ranking, except for gross generalizations comparing schools at, say, below 1% vs. above 10% of grads. Only those few HS kids who are interested in future-PhD-tending campus cultures should pay any attention to these lists; maybe the OP HockeyKid is one of them. :)</p>
<p>
10% unknown undergraduates is not that big of a margin of error unless there was a systematic conspiracy by the National Science Foundation to deride the Ivies. I do believe that as a respected government institution, the NSF would partake in sound surveying practice such as random sampling and the 10% is roughly evened out. </p>
<p>If anything, I would think many top universities such as the Ivies would be ranked even lower given that students from lesser institutions would be more “ashamed” of their schools or merely don’t want to put the effort elaborating about their obscure college of choice that no one has ever heard of.</p>
<p>
That’s a horrible analogy. PhDs are nothing like Nobel Prizes or Jeopardy. PhDs are very attainable to the point that 35% of the CalTech undergraduate populace will go on to earn one. This is not like the Nobel Prize or Jeopardy where a dozen or so individuals get to participate every year. Hundreds of thousands of PhDs are awarded every year and many, many more students strive for a PhD. This is hardly a measure irrelevant to students.</p>
<p>Furthermore, I would like to reiterate that I do believe that there is a correlation between PhD attainment and JD and MD attainment, given that doctoral degrees are similar levels of achievement. If Harvard students cannot even muster a PhD, what makes you think they’d be any more competitive for a MD? This data should be relevant for all students aspiring for a doctoral degree.</p>
<p>Regardless of PhD productivity’s relation with JD and MD productivity, a PhD is an area of merit. PhDs are a greater area of outcome-based merit than any other data we have available. In this way, PhD productivity is the best “proxy for academic quality”.</p>
<p>By the way, I’d like to point out that the J.D. is not a terminal degree.</p>
<p>The J.S.D., a research doctorate, is the highest academic degree in law, and is equivalent to the Ph.D.</p>
<p>
</p>
<p>That’s not how statistics works. If you are unable to find data for a subset of a population, then you have no idea what the makeup of that subset is. It may or may not conform to the rest of the data. Considering that 28,000 is a sizeable portion that could cover a school’s total production many times over, several schools could have had a large portion of their graduates in that portion.</p>
<p>And I wasn’t suggesting that it’s the Ivies that are discriminated against. Again, we have no idea what the makeup of that subset is.</p>
<p>
</p>
<p>Nope. I’m really not. Note the qualifiers in the statement you’ve quoted. They are significant. There are other interesting measures that are not outcome-oriented. Academic quality is very important, but it isn’t the only important kind of quality. </p>
<p>Even as an outcome-oriented metric for academic quality, it is imperfect (as most other metrics are). I agree that it would be good to have a comprehensive ranking not only of PhD completions but also of per capita MD and JD completions. Unfortunately, I’ve never seen such a thing. That’s why I think per capita PhD production may be the best we have in outcome-oriented measures of academic quality. I think it is better, for example, than Rhodes Scholar or Nobel Prize production because in those cases we are dealing with such sparse data. The percentage of alumni even at top schools who get PhDs is small, but it isn’t minuscule. I think this measurement also is more significant than alumni giving or (to me) the Payscale earnings data.</p>
<p>As for whether this measurement should matter to you if you have no interest in getting a PhD yourself, I think it should. Would you not care at all to know that your school produces a lot of Nobel Laureates and Rhodes scholars, even relatively speaking, and even if you have no expectation of getting one yourself? Brainy, serious, intellectual kids enrich the campus environment (even, in some cases, if they are a little foolish about their career options.)</p>
<p>Anyway, I think a lot of good points have been raised here on both sides of this one. Too bad there is a lot we simply don’t know about reading the measurement tea leaves.</p>
<p>Phantasmagoric wrote:
</p>
<p>Feh, a virtual tie with Brown #25 (7.4) which is twice the size of Wesleyan but very similar in other ways. This suggests to me, that this is less a proxy for academic quality than a rough proxy for the percentage of the listed college’s student bodies majoring in science.</p>
<p>“Cannot even muster a Ph.D.” at post 65…what an ignorant statement.</p>
<p>I developed my own model for evaluating colleges and included as one of my variables the percentage of a class that can be expected to earn a Ph.D. It is one of many variables in my model, and I think it does reflect to some degree the kind of learning environment at an undergraduate institution. </p>
<p>I also include as one of my variables post-graduate salary over time. </p>
<p>I encourage anyone to develop his or her own metrics for evaluating colleges. We are all different and have different ideas of what a “good” college is.</p>
<p>BTW, the initial thread didn’t indicate how high Cal Tech is compared to the others on the list. The absolute numbers are more revealing than the rank order.</p>
<p>Most students have neither the need nor the desire for a PhD.</p>
<p>To me, this is the number one reason not to consider this an indicator of academic quality. The vast majority of people don’t need a PhD. I think this list says something different about the schools - they have the close-knit, intellectual, liberal-arts-focused environment that often fosters a deep love bordering on obsession with a particular field. Students also have closer relationships with their professors and probably see more of what they do, so that may make them more likely to want to be a professor themselves. At Spelman, I initially decided I wanted to get a PhD and be a professor because there was a professor I really admired and I decided I wanted to be like her. (Luckily, my interest in the field has deepened since my freshman year.) I note with pride that my alma mater is actually ranked at #5 on the social sciences list.</p>
<p>Students at Columbia probably don’t have the same experience because they don’t interact with their professors as much, and besides, their professors aren’t really doing what they would like to be doing - except in the few cases in which they are (by which I mean those students are actually really, really interested in the prospect of working 80 hours a week doing research in a very narrow field).</p>
<p>That said, I would looooove to teach in a place like Reed, Oberlin, Swarthmore, Carleton, Grinnell, or Bryn Mawr.</p>
<p>And actually, Ghostt, for the purposes of most postdoctoral programs a JD is equivalent to a PhD. In my field we often accept law school graduates as postdocs in addition to PhDs in various social and biological science fields and MDs, and typically they have a JD and not a JSD. Also, the vast majority of law school professors have a JD and not a JSD, at least in the US.</p>