A couple of articles recently feature Williams professors. Here they provide insight into peer effects in elite education:

Similar findings [that being with smarter peers has a good effect] have been made by David Zimmerman and Gordon Winston, both professors of economics at Williams College, who discovered that students who ranked in the middle of the pack at three selective universities got slightly worse grades if they roomed with someone in the bottom 15 percent of the class. Earlier, Professor Zimmerman also had found a positive effect on midrange students who had a roommate with higher scores on the verbal portion of the SAT, but in neither case did studies show much effect from roommates on students at the top or bottom of their peer group.

Moreover, experiments in the psychology lab of another Williams professor seem to indicate that, in some cases, studying among smart peers can do more harm than good. In a three-year study, George Goethals assigned students of varying academic ability to work on simple tasks, such as critiquing newspaper articles. Students are very acute in gauging where they fall in the academic pecking order, says Professor Goethals. Often, those who perceive themselves to be intellectually overmatched simply disengage from the experiment.

“The answer seems to be yes, students affect each other a great deal,” he says. “But the effects are complex. You have to tell a complicated story about the influence students have on each other.”

I have blogged before about Zimmerman and Winston’s work on this topic. (Again, it is too bad that, unlike almost all other academic working papers, these are not easily downloadable.) I also think that some of this work has been turned into senior theses at Williams — sure do wish those were easy to read on-line.

Mark Taylor gets some nice press in The Washington Post in a story about the structure of higher education.

Back in 1998, Mark Taylor, a Williams College philosophy professor, and Herb Allen, a Williams alumnus and Wall Street financier, launched a company based on what seemed like a powerful insight:

Every year, they noted, there are thousands of college professors who twice or three times a week offer what is largely the same basic lecture course in a subject like molecular biology or Shakespeare comedies. A few of these professors offer the kind of brilliant lectures that fill auditoriums and provide the kind of educational experience that students remember all their lives. Many of the rest offer something that ranges from mediocre to awful.

So, asked Taylor and Allen, why don’t we identify these extraordinary lecturers, put their lectures on CDs, and sell them to universities that could supplement them with faculty-led tutorials or discussions? The advantages seemed pretty clear. Colleges that employed the celebrity lecturers could help defray the cost of superstar salaries while enhancing their own reputations. And schools that purchase the lectures could lower costs while improving the quality of their educational offering.

In business terms, this was nothing more complicated than bringing the proven benefits of scale economies, outsourcing and high-tech distribution to higher education. But as you might have guessed, it was not exactly welcomed by an establishment that prides itself on remaining a quaint cottage industry.

“To be frank about it, the resistance was astonishing,” Taylor recalls.

Elite universities worried about “diluting” their brands and “contaminating” their mission by joining in a profit-making enterprise. And faculties immediately saw a threat not only to their jobs and salaries, but a lifestyle and teaching model that had cosseted them for centuries. So after several years of trying, including a “no” vote from the Williams faculty, the effort was scrapped.

Williams is solving this problem, via the tutorial program, in precisely the right way. It should do more along those lines. Indeed, there is no good reason, to my mind, why any class should ever have more than 20 students in it. This would require a lot of (small) sections of things like CHEM 101 and the like, but that would be a benefit, not a cost.