Who are the smartest graduate students? You’ve probably already guessed that one: physicists. Second down in the brains ranking are mathematicians, then computer scientists, then economists and practically any sort of engineer. Such are the results of an analysis made in 2004 by Christian Roessler, a lecturer in economics at the University of Queensland, in Australia, of the mean scores on the Graduate Record Examination of Ph.D. candidates in 28 different academic fields. Roessler’s findings, recently linked on the Carpe Diem blog of Mark J. Perry, an economics and finance professor at the University of Michigan-Flint’s business school, and also by education blogger Joanne Jacobs. Roessler derived his rankings by looking at doctoral candidates’ average scores in 2002 on the three components of intellectual ability tested by the GRE: quantitative, verbal, and analytical (the analytical section, then a multiple-choice test like the quantitative and verbal sections, has since been replaced by a written test of analytical reasoning).
And if physicists (No. 1), mathematicians (No. 2) computer scientists (No. 3), economists (No. 4), and engineers (Nos. 5, 6, 7, 8, 12, and 13) are the smartest young people, judging from their test scores, to enter graduate programs that will train them to conduct scholarly research and teach the next generation of scholars in their fields, who are the dumbest? The answer to that question may well be easy to guess, too: grad students in communication (No. 26), education (No. 27), and public administration (No. 28). The dismal mean scores for doctoral candidates at education schools (467 in verbal ability, 515 in quantitative)—giving new meaning to the adage “Those who can’t, teach”–prompted a commenter on Jacobs’s blog to write, “The fact that the dimmest bulbs in our colleges self-select themselves as being the ones who should influence the education of future generations explains many of the edu-fads we see, as well as our continued failure to improve educational outcomes across disadvantaged populations….”
By contrast, the physicists, mathematicians, computer scientists, economists, and engineers consistently scored on average either above 700 or close to it in quantitative ability, although their verbal scores tended to be mediocre (the top-ranking physicists, for example, scored only 536 on average in verbal ability, while civil engineers, ranked at No. 13 on Roessler’s list, scored a mere 469, just two points higher than the educators). The scientists tended, however, to make up for lost verbal points by their high scores—typically above 600—on the analytic component of the GRE, a feat the educators, testing on average at 532 in analytic ability, could not match.
One of the most fascinating aspects of Roessler’s rankings is how poorly doctoral candidates in the humanities fared on the GRE compared to their counterparts in scientific and mathematical fields. Of the various humanistic fields, only philosophy (No. 9) ranked in the top half of the 28 academic fields surveyed by Roessler. (Graduate students in three other scientific fields also made it into the top half: chemistry, ranked at No. 10, earth sciences, ranked at No.11, and biology, ranked at No. 14). Indeed, not a single humanities field surveyed by Roessler boasted a mean score of more than 591 on any component of the GRE, with the sole exception of architecture (No.21), whose doctoral candidates averaged 606 on the quantitative component of the test but sank near the mean bottom with a mere 491 in verbal ability and 555 in analytic ability. Even doctoral candidates in English (No. 15)—whom you might imagine to possess stellar verbal skills if nothing else upstairs—failed to distinguish themselves on the GRE’s verbal component, with average scores of only 573. Grad students in religion (No. 16), history (No. 18), and anthropology/archaeology (No. 20) fared even worse overall.
There is an elephant in this GRE living room, a female elephant. The academic fields that yielded the highest overall scores in Roessler’s survey are the so-called STEM (for science, technology, engineering, and mathematics) fields in which women are famously—and to many observers troublingly–underrepresented. At the undergraduate level, it would seem that the higher the Roessler ranking for an academic field, the fewer women seem to concentrate in it. For example, although half of all undergraduate majors in biology (Roessler’s No. 14) and related life sciences are women, only one-fourth of undergrad majors in No. 1-rated physics are women. The number of women in STEM fields thins out even more in postgraduate studies: While nearly half of all college math majors are female, women account for only 2 percent of math doctoral students. By contrast, tank-ranking education, and, increasingly, the humanities and social sciences, are dominated by female students at all post-secondary levels. Some might argue that such disparities are due to different kinds and levels of intellectual aptitude for the two sexes (boys consistently outscore girls in the quantitative section of the Scholastic Aptitude Test that is linked to success in math and science, but also, if less dramatically in the SAT’s verbal section)–or women may simply be less interested than men in scientific careers. Or a combination of both aptitude and interest may be work.
Whatever the reason, the Roessler rankings suggest that, if GRE scores are a valid measure of intellectual aptitude, the brightest college graduates are getting their Ph.D.’s in male-dominated scientific and mathematic fields, while doctoral programs in the humanities seem to be the province of those more modestly gifted in the brains department. As for doctoral candidates in education, perhaps the less said, the better
As a retired physicist who did quite well on the GRE exam more than four decades ago, I would put precious little stock in the analysis of either Roessler or Allen with regard to who are the most “intelligent” applicants to graduate school. Undergraduates who major in the science and engineering disciplines have had four or more years to hone their quantitative and analytical skills in courses that require them in heavy measure. This puts them at a great advantage in comparison to their colleagues in the humanities and social sciences when they take the GRE.
Whether or not this makes us physicists “smarter” than sociologists, say, is an open questions. Over the years I encountered a fairly large number of colleagues who were adept at quantum mechanics, but who would have had trouble finding their way out of a phone booth in Fresno with a road map.
I know of one Caltech physics professor who was so lacking in common sense that he had to call his secretary half a world away when he missed an airline flight to find out what to do.
There is more than one type of intelligence.