The following links offer excellent advice about selecting graduate programs and advisors, a timely topic this time of year (thanks to the Nanopolitan blog for the pointers):
- Choosing a graduate program, by Greg Mankiw
- Choosing a graduate school, by Diana Davis
- Toward a jerk-free world, by Nanopolitan
To complement their posts, I want to emphasize that you don't really have good advisors - instead, you have good student-advisor relationships. By the same token, you occasionally have bad advisors, but often you have bad student-advisor relationships. While some advisors certainly are more successful than others in helping their students develop, it should not obscure the fact that what represents a good relationship for one student might be awful for another; in addition, some will thrive in an environment where others falter. They should recognize their strengths and find the advisor who can make the best use of these skills.
An old Nanopolitan post describes the "advisor from hell", who inflicted verbal abuse on his post-doc and tried to hurt his career. The post-doc ultimately realized the relationship was not salvageable and moved on, but that was obviously a very difficult decision to make.
To summarize (using my own insights and the posts above), here are a few things to keep in mind:
- Job prospects after graduation, especially in academia, heavily depend on the name of the university you get your PhD from (and the university where you do your post-doc, if applicable), the ranking of the department, and the recommendation letters your committee members will write for you.
- Everything else being equal, an advisor with a strong record of supporting students through Research Assistantships is better than an advisor with a strong record of supporting students through Teaching Assistantships. That's because you want time to do research. Being a TA a few times is a wonderful experience if you want an academic position down the road, but it is also very time-consuming.
- Some advisors want to meet with their students almost every day, others every few weeks, others once a week. Make sure that your advisor's habits match what you are comfortable with. Ideally, your advisor should be open to meeting at least once a week if you are getting started in your research, and a little less often as you become more senior and develop independent-thinking skills.
- An advisor with lots of students might produce a lot of attention-generating papers, but he might also be short on time to spend with each student individually.
- Everything else being equal, you should avoid working for an advisor whose students take much longer than the department average to graduate. You should also look into the jobs the students accepted after graduation, especially industry vs academia.
- Failure is no fun. If a department has a history of failing many graduate students at the qualifying exam, you want to know that before you accept the offer. You don't want to start the process again somewhere else. Besides, often students who fail the exam do not want to start again from scratch elsewhere and will simply change career paths, but they might have been outstanding researchers if they had joined another institution.
- Don't choose a program based on a single faculty member. He might decide not to fund you or he might leave. Make sure you have a backup plan if the person you want to work with is not available.
- If there is a faculty member you're interested in working with and you're already on campus (for instance in the Master program), try to take his or her classes so that he/she can see your worth first-hand. Plenty of students with excellent undergraduate transcripts end up being disappointments because they cannot adjust to graduate-level work. But also, students who look average on paper sometimes end up as success stories.
- Ask students what they think and ask professors about their advising style, but be aware that people might have an agenda.
- Don't forget you're in this for the long run. You need to maintain a good working relationship with your advisor for four to five years, while many of the friendships you make will ebb and flow and sometimes fade. Pick an advisor you can see yourself working with - not just someone whose name will look good on your CV.
- Understand most advisors do the best they can, and you're only aware of a tiny part of what they're doing (writing grant proposals, attending committee meetings, preparing lectures, editing their students' papers, etc). They're not trained in how to advise students - often they only have the example of their PhD advisor (or possibly their post-doc supervisor) to provide inspiration and develop their own advising style by trial and error. Be willing not to take things personally.