Archive for November 13, 2009
I wonder if this idea would help computing education, too. If students aren’t getting it, maybe you should redouble your efforts on a smaller set of goals and learning objectives?
Middle school students in every racial, ethnic and income group show greater mastery of mathematics — including algebraic reasoning, statistics and geometry — than they did three years ago.
Educators attribute much of the progress to Oregon’s embracing a national recommendation to drastically scale back the number of math topics covered in each grade.
In Alan Kay’s comment to a recent blog post, he mentioned again that ARPA in the 60′s and 70′s “funded people, not projects.” I’ve been meaning to post a (unfortunately, whiny) blog entry about that phrase, and his comment has goaded me into it.
The current way we fund research is broken. I know that there are researchers who are doing really well under the current system and will disagree. I personally have been blessed with several large grants recently, so I can certainly be accused of ingratitude with these comments. I just keep running into roadblocks that don’t make much sense to me, that seem to me to be a result of our national policy of funding projects, and not people. Some examples:
- Graduate students often take more than 3 years to finish their PhD. NSF grants are almost always 3 years long. There is intense pressure to have graduate students fully funded, or to dismiss them from the program. “If they’re doing good work, you should be able to get them funded.” So what do you do when a student is doing good work, has 9-12 months to finish, and runs out of his NSF grant? Do you apply for another NSF grant? That’s great for his last year, but how do you fulfill your commitments on the proposal for the extra two years after he graduates?
- The whole idea of matching students to NSF proposals is a mismatch of expectations and realities. I have a project where we wrote in a new PhD student before she arrived. It rarely works out that smoothly that we offer a student admission, write a proposal to do what we’ve agreed to work on, then you get the funding as she walks in the door — but in this case, it worked. However, new PhD students don’t really know what they’re going to do, and this one has now realized that her skills and interests are taking her in a different direction. That’s great for her, and I bet she’ll do great work. But now how do we fulfill our commitments to our grant?
- I’ve complained before that the US doesn’t fund computing education research. One of my colleagues just got an NSF BPC proposal rejected, in part (one of the criticisms in the panel review) because there was “too much research.”
- I have a CIFellow working with me, a post-doc. He would like to apply for funding to let him continue his post-doc for an additional year beyond the year funded. Georgia Tech won’t allow him to be the PI of such a grant, so I agreed to be the nominal PI while he wrote the grant and did the work. Turns out that new policies don’t let me. Georgia Tech won’t let me submit a proposal where none of my time is funded. Okay, let’s throw a day or two of me on the grant. Nope, that won’t work either. I currently have NSF grants where my funded time adds up to two months, and NSF (at least in Education and in CS) won’t fund more than two months (summer or academic year) of any faculty’s salary. Until my current grants expire, I effectively cannot submit any more NSF proposals, even as just a co-PI or nominal PI.
A lot of this craziness comes from funding being based on individual projects, not on funding people and programs. However, to be honest, I don’t currently see a way to get it changed. The increased emphasis on accountability and having measurable benefits requires funding to be associated with the outcomes of a particular project. If we funded people and programs, instead of projects, how would you decide who to fund? Even many schools formerly focused solely on teaching are trying to break into research, to help shore up sagging revenue lines. If we only funded people with proven track records, how do new people break in? If the new folks have a great idea for a project, shouldn’t we fund that project?
I don’t have a solution to offer. It just seems to me that there’s a disconnect between the demands of research and graduate work, and the realities of funding and accountability rules.
Okay, I’ll stop whining now. (For this post, anyway.)