<p>My take on research is that you need to be able to demonstrate a certain amount of skill in order to prove to the professor that you wouldn’t turn out to be a liability for him if he offers you a position in his lab.
There are two ways to do this: Personal projects and/or high GPA. If you have outstanding personal projects that really shine through, then you have a high chance of getting some meaningful research. If not, then you have to rely on your GPA. This being said, I’d go so far as to say that a low GPA and minimal personal projects wouldn’t get you very far in a well-funded private institute either (I might be wrong in general, but I’m certain this is the case with Stanford, since I’m in touch with quite a few people there).</p>
<p>@yeah: I’d like to clarify that I didn’t mean to sound condescending at all towards Cornell or UPenn when I mentioned them like that - I meant to say that they have many undergraduate schools that are preprofessional programs. There’s nothing wrong with “core training” or “industry focused” education that they (and state schools) provide. My point is that when choosing a school, you should consider what your attitude towards learning is, too. </p>
<p>Also, you tend to have a more focused peer group at smaller institutions than state schools, simply because state schools are required to admit almost everyone from the state who applies (this does not apply to Berkeley).</p>
<p>An engineering professor at a state school recently told me that graduate students from private, undergrad-focused institutions are better off for engineering grad school on average, in his opinion. Also I’d think it would be easier to fall into the shadows at a state school, if the students’ academic quality is as high as you say it is.</p>