Wednesday, December 26, 2007

Research advice

Some solid advice for grad students considering major academic research projects:
1) Use graduate school to tech up. You'll have time to learn how save the world later, when you're actually in it. Learn all of the theoretical, statistical and other difficult-to-acquire skills you can while in grad school, because you won't have the time later on. You, your cause, and your job prospects will be well-served by the technical skills you build.

2) Hang in there. In the first year of any grad program you will encounter a lot of required material that will feel too theoretical, too divorced from social change, and (occasionally) like too much nonsense. Much of it is good for you (see point 1), even if it doesn't feel like it at the time. After a year of metrics and micro theory, I was ready to run to the real world to do what I thought I really wanted to do. The best advice I ever got (from one of my pre-PhD advisers) was, "Shut up and hang in there; by your second to third year you will discover all the people doing interesting applied work soon enough and be free to work on whatever you want by your third year." He was right.

3) Take chances. The second best piece of advice I ever received came from my dissertation chair, shortly after my oral examinations committee told me that my prospectus was poorly thought out, uneconomic, and overly risky. They were 100% right, and I benefitted from hearing it (although at the time I was miserable). Where I think they were wrong is that they told me to abandon my plans for risky and expensive field work. They favored the less risky route that could get me to a completed dissertation faster. My chair's response: "Hey, if you really want to do this, why not? Give it a shot. If it doesn't pan out after three months, then come back and work on something else. Worst case scenario: you lose a few thousand dollars and a summer, but you have a great experience." I plan to give the same advice to my students.

4) But minimize your risks by being prepared. Don't embark on a big project, especially field work, without a solid hypothesis, research design, and plan. Think through the theory beforehand. Write down your assumptions, your logic, and your econometric regressions before you collect data. Especially write out your regressions. I am still guilty of rushing to the field too quickly, and am continually reminded of the costs.
The author (Chris Blattman, an assistant professor of political science and economics at Yale) has six additional pieces of advice. The post is intended for people considering economics-related research as part of a PhD program, but some of the tips can be applied to what ALM students are doing at the Extension School. Tip #4, above, seems especially relevant -- in my proseminar and in the ALM thesis writers' workshops, other students would sometimes propose complicated paper-based surveys, or ask extremely broad research questions (e.g., "does religion encourage war?"). Fortunately, the ALM program has processes intended to help candidates find realistic, solid research topics -- namely, the proseminar and the thesis proposal.

(Thanks to Greg Mankiw for the link)

Related entries:

No comments: