US News Ranking of CS Programs
US News and World Report came out this weekend with their latest rankings of graduate programs. Computer Science at Carnegie Mellon received a score of 4.9 out of a possible 5, which in the overall rankings puts us slightly behind Berkeley, MIT, and Stanford.
This is a good result, but I am disappointed by it. Why? Well, I’m pretty certain we’re better than that. We’ve done exceptionally well at recruiting lately, landing new faculty hot-shots like David Andersen, Alyosha Efros, Carlos Guestrin, Anupam Gupta, Ryan O’Donnell, Adrien Treuille, and Luis von Ahn. We’ve added more strength in computational biology, with Ziv Bar-Joseph, Chris Langmead, and Eric Xing. And now we’re on a campaign to hire even more outstanding new faculty. The new hires in the other School of Computer Science departments and institutes have also been just super. Funding, while seeming to consume yet more time and effort from everyone these days, has continued to grow, and in particular we have shown utter dominance in winning NSF grants. We have not just one but two fantastic new buildings being constructed for us with wonderful gifts from the Gates and Hillman foundations. And, of course, we’ve had some banner years lately in major awards, culminating with things like Luis von Ahn’s MacArthur Genius Grant, Takeo Kanade’s Bower Prize, and Ed Clarke’s biggie, the Turing Award. The combination of these and so many other developments has paid off handsomely in our PhD student recruiting — in the last cycle we had our best recruiting yield in several years.
Having said all that, and despite all of this obviously great forward momentum and growth, I’m not so surprised that the US News survey ended up dinging us by a tenth of a point (or thereabouts, due to possible rounding of the scores). Since it is based primarily on a reputational survey of department and graduate program heads, the fact that we’re such a big, complicated place doing a lot of core but also a lot of “wild and crazy” things, makes us, well, hard to judge. We’re not very conventional and we don’t fit neatly into the four classical areas of focus for the US News survey (AI, Systems, Programming Languages, and Theory). Just as it took time for other universities to realize the importance of robotics after we created the Robotics Institute (and similarly with HCI), it may take a while for people to believe that our “big bets” on areas like machine learning, language translation, human computation, DISC, and so on are worthwhile. While outsiders continue to see that we are #1 in Programming Languages, they have not yet had enough time to see how strong we have become recently in Theory.
I suppose there is no way to comment on our ranking without sounding a bit like sour grapes, and I certainly don’t mean to do that. In fact, to my mind the other three top-ranked departments are truly great, each in their own way, and we definitely value the students and collaborations we get from them. But it is also clear to me that there is a gap between people’s perception of us and the reality, and that the size of that gap is about 0.1. ;-)
Peter Lee @ March 31, 2008
I imagine the primary use of the USNews rankings is to provide a metric for graduate school applicants, so having just finished what I think was a very successful visit weekend for admitted students, I found these rankings really irksome. From lots of conversations with different admits, I got the impression that we compare quite favorably with at least one of the top-rated schools. Will this new rating be enough to tip some prospectives on the fence from coming to CMU? I certainly hope not; but it’s really hard to capture the enormity of what CSD does here in a single weekend!
You are so right: we have great professors and brilliant students, and we’ve won so many important awards….. But why, still, we can’t get No.1 even though we are so good?
The drop in ranking might have a negative effect on the prospective students or faculty members. As far as I can see, even though many people don’t believe in the myth of ranking, they don’t have other sources to rely on.
And, it seems to me that you are trying to evade your responsibility by only mentioning how good we are. No one will lose confidence in CMU only because of the 0.1 point. But as the head of the biggest and best department in CMU, you should really think why we lose the point and how we can be better.
I think we should put more effort on theory where we have the lowest ranking.