“Our baseline results suggest that increasing class size by 20 students reduces a student’s wage by approximately 6%. "
OMG, I must inform S. He’s ruining his future wage potential in those large CS classes… :((
@privatebanker I don’t disagree with your points. I’m looking for evidence that supports the argument that class size is a credible factor in the consideration of ranking a university even in STEM. Not the only way and maybe not as much in STEM as say Social Sciences or the Humanities, but still a variable that could be used. If I want to study CS at UofM or CMU, could class size be a variable that I’d consider. I think I’d say yes I would, and there is some research to back that up.
@BrianBoiler Good points. I was just interested in the research article and found the conclusions cloudy despite their headline. But who knows. Interesting to debate I guess.
But better in theory doesn’t always translate into better across populations. Implementation and costs outweigh marginal utility.
I.e. considered in isolation, smaller class size may be better.
But does it make the overall college experience better if smaller class size requires the college to reduce offerings of upper level courses to have faculty teach more introductory course sections, or some/more students are blocked from taking desired courses because they are full at the smaller class size limit? Or if the college is more expensive because it needs to hire more instructors to teach more introductory course sections to keep each one small?
@ucbalumnus Exactly.
Ok so as privatebanker alluded, this is not exactly a comprehensive survey:
“For this study, researchers collected data from 17 introductory biology courses from four institutions: California State University, Chico; Cornell University; University of Puget Sound, and the University of Minnesota Twin Cities.”
They just looked at bio courses from three large state schools and UPS, if you want to major in biology at UPS instead of Cornell, go for it, there’ll be a lot of Cornell applicants who were waitlisted that will be happy with that decision so another spot opens up.
De Giorgi, G., Pellizzari, M., & Woolston, W. G. (2009). Class size and class heterogeneity. IZA Discussion Papers, No. 4443. “
So even though two of the authors were from Stanford and clearly they know a lot of math functions, the university studied was Bocconi university, not Stanford.
"Such an intervention [reducing average class sizes to 20 students] would generate a gain of 80 euros x 1,500 students, or 120,000 euros in total each month, which are likely to be more than enough to pay the costs of acquiring the additional resources necessary to activate the two extra classes.”
This means though that the students get a higher salary, how does the students getting a higher salary mean they can add two more instructors? Are they assuming the students will give back in alumni donations to fund two teachers? Seems like a stretch.
Bocconi has 14K students, so not exactly a small college but it’s in Milan, and I won’t criticize anything in Italy. If it means I can have a couple glasses of wine at lunch with my pizza during classes, I’ll attend, large class or not!
Even with improved CDS class size reporting, I doubt the pecking order would look all that different. Roughly speaking, the average class size order still would be: LACs < elite private RUs < Public Ivies < most other state flagships < most directional state universities. But within each of these categories, the intercollegiate differences still won’t be as great within each category (or even across some categories) as the intra-college differences among majors, course levels, or popular vs. unpopular professors. UChicago and Columbia really do tend to have smaller classes than Penn State. However, a linguistics major at Rice is likely to experience smaller classes on average than an econ major at UChicago or Columbia. And CS classes tend to be over-crowded almost everywhere.