Maybe in Arnold and Megan's experience, but not in mine. At both Princeton and George Mason, I found one of students' main problems to be low morale. This ends up being a self-fulfilling prophesy: Students with low morale don't try to be creative and productive, which leads to mediocrity, which reinforces low morale. I think it afflicts many excellent but under-confident students, who endlessly polish articles that should have been in the mail months or years earlier.
The typical grad student I've known profits more from encouragement ("Your paper is fine the way it is. Put it in the mail right now, while I watch.") than being taken down a peg. At risk of sounding like a self-help guru, the grad students who succeed and eventually make real contributions are generally those who are too proud to fail.
P.S. I sense a skeptical reply from Robin Hanson coming. In anticipatory rejoinder, let me say that I'm not telling grad students to overestimate their potential. I'm encouraging grad students to live up to their potential.
"... the grad students who succeed and eventually make real contributions are generally those who are too proud to fail."
Did you mean stubborn rather than proud?
An unexpectedly empathetic post from Prof Kling ... a post that one wishes one had come across 15 years earlier! Thank you, of course.
Sincere apologies to both Professors Caplan and Kling for having misattributed the post.
I'm starting grad school in the fall. I am torn between high levels of confidence (almost cockiness) and the nagging self-doubt that comes with "the next level." I often wonder to myself if I will be creative enough to add something meaningful to the collective knowledge in my field. It is however, reassuring to hear you Bryan saying that the most successful are the most proud. I have that down, no problem.
Great post Bryan!
I agree - I think the average grad student has too little confidence, not too much. I think there is a difference, though, between the confidence level of first-year Ph.D. students and upper year ones.
My personal experience in a Ph.D. program here:
http://economics.about.com/b/2008/06/19/terrific-econ-grad-school-advice-from-bryan-caplan.htm
I agree with professor Caplan's observations. I am presently a graduate student and well aware of the pernicious consequences of it. I have also observed a relative dearth of arrogance among the students, although the prevailing leftist consensus among them does inspire occasional outlandish assertions about, "capitalist lies," and the like by some students.
Perhaps the attitude of graduate students vary among schools. I would attribute it to status, The Nelson A. Rockefeller College of Public Affairs and Policy (Of SUNY Albany), though a respectable institution, is not an elite one, but given that professor Caplans' applies his observation to Princeton, that seem an unlikely explanation, unless Princeton is somehow exception in ways other than its reputation and academics. I suppose it could be founded upon some other factor or that Miss McArdle's observation is founded upon an idiosyncratic experience.
This grad student agrees with you, Brian: some grad students do swagger with overconfidence, but most suffer from underconfidence.
In my department it seems like either of these two extremes results in failure, as many overconfident students don't study as hard as they should and get tripped up by prelims and, as you say, the underconfident ones tend towards mediocre (at best) work.
Bryan, It sounds like the "when should I ship" dilemma that software developers face. Intuitively, I prefer, "ship early and use the opportunity to correct problems as a chance to show how responsive you are". In practice, I'm a lot more conservative and a terrible perfectionist.
Given that you can submit essentially the same paper over and over until it's accepted, it seems to me that if there isn't any terrible downside to having a paper rejected, it's best to get it out as soon as it has a snowball's chance of being accepted just to get the review system's feedback. Nobody likes rejection out of the box. Learning to use it as a tool of the process that I would guess most grad students just aren't old enough to internalize. And then there's the pride of always doing your best that got them into grad school in the first place...
Guy Kawasaki, in his business book "The Art of the Start" called this attitude "Don't Worry, Be Crappy". It means get it out to market as fast as possible so the market can give you input. It doesn't mean "let it stay crappy".
I think you're missing a distinction between the in group and the out group. Graduate students may be overconfident in their abilities relative to non-academics while remaining underconfident in their abilities relative to others in academia. It's the former I'm talking about.