I think we–possibly all three of us–are actually saying exactly the same thing. I’m not sure if this was intentional or not, but I admit I didn’t notice it. In essence, saying that “it’s much easier for you to get help when you need it and talk to others” at a small school (you didn’t say this) ignores who the “you” is. That’s what I was trying to point out, maybe I didn’t do a very good job:
If “you” are one of the top 75% RPI, SATs or top 25% Penn State SATs, or the top 90% of Penn State-Honors College SATs* (all of which are the same number), you’re going to see very similar resources, support, training, graduation rates–most everything–at most of the good universities for your program. The amount of effort this takes “you” will not be strongly correlated to the size of the university, how much you pay, or what you mascot is.
Now, if “you” are not one of those people, “you” aren’t likely to end up at the honors college or RPI. “You” are statistically more likely end up at a large state school or something similar. “You” may well feel you have less support, access, opportunities, or whatever it is. This is not entirely the school’s fault: those opportunities are out there, advertised, sometimes even pushed on you. They may even be more numerous at the typical large university. (Inconclusive; I have no good way to quantify this.) Yes, “you” may end up feeling like a number, but much of this is student complacency in ignoring these resources. Such complacency would likely have a similar effect at RPI (in fact, I’ve seen this), but it has fewer students there who are prone to this.
In summary, the statistics on this issue (or any issue) should not be taken out of context. “You”, as far as the statistics I have seen, are not really more likely to do poorly, drop out, or switch majors at a large school than a small school, or vice versa for that matter. “You” won’t even necessarily have a more difficulty accessing resources at a large school. I argue this is not a case of a few, but a trend of the many. Schools’ resources differ, their curricula differ. But in the end they’re just that: different. High dropout rates correlating to certain aspects does not indicate a causation between the characterized systems and the rates.
*I agree, SAT scores are really not a very good way to measure this. Insert some non-existent perfect measure of intelligence here.
EDIT: To clarify on the topic of class sizes, I think I’ve had 2 courses in which grad students did part of the teaching. One was absolutely outstanding–one of the best teachers I’ve ever had–and I took him deliberately. The other was for a lecture class that I took because I wanted to be able to leave early for robotics (this course also went very well). All of the rest of my classes have been taught by completely professors. Almost all of them have been outstanding teachers as well as exceptionally gifted engineers/scientists/whathaveyou. As for being irrelevant to intro classes, I have friends that got into those groundbreaking labs as freshmen–first semester freshmen. Granted, they weren’t doing anything profound, but they were there–and they still are. Only now they’re publishing, writing theses, speaking at conferences.
Another pretty neat place is our combustion lab, where there have been a couple of undergrad rocket engine firings, and a whole ton of ignition research.