Grad school tips
1) A political science Ph.D. is a two-part process. The key dividing point is your comprehensive examinations (or field exams, or qualifying exams, or whatever. At Penn they were called "comps", which is the lingo I'll use). I would recommend a different approach to each side of the comps. Before comps, you should take advantage of what a graduate education is, a chance to get paid to read and explore a field of your interest. I took an eclectic collection of courses during my first two years, including an infamous "Great Books" of comparative politics. But what you want to do here is to scrape the surface of all these different possible fields you could be interested in, to see what works for you.
A mistake that I've seen many graduate students fall into is being held prisoner by the topic they discussed in their graduate school essay. While that topic is important -- you have to sell yourself to get into grad school or you wouldn't be there! -- you really do not know how much you do not know.
2) Start your own research project. There's a paradox in academic research, the interesting questions have either been researched, or there isn't research on them for a reason. For example, I wrote about the potential for social networks to change campaigns (this was in 2010 mind you) -- an issue that we can all agree clearly matters. That being said, it's one of the toughest fields to do research in, because the data is either proprietary and mostly unavailable (from Facebook), or freely available but completely overwhelming in scope (Twitter). So is the impact of social networking on campaigns interesting? Yes. Was it an accessible research question? No, not at that time. Moving on from that idea was one of my best decisions in grad school.
3) Ideally, start a research project that can stand independent of your dissertation. My second summer I started a project on the impact of a ballot reform in Virginia (that eventually became this article). I presented it at one of the large political science conferences, and a very helpful discussant asked if it was part of my dissertation, and I said no. He winced, saying "Beware the side project" and said he prohibited his grad students from such side projects.
I would recommend the opposite. There are valuable lessons to be learned from shepherding a project through the manuscript phase and journal submission and publication process. Being able to do it on a project that is not your dissertation allows you to take a dry run at this, gaining experience that helps when it is time to do the same with your dissertation. My article was solo-authored, and I think this is even better as a co-authored project.
4) Comps. Do enough. You take comps for a reason. Many graduate students think that reason is faculty is just punishing you, and making you do what they had to do. A rite of initiation for academia -- this may in fact be the case.
The more important function of comps is that it forces you to synthesize a lot of information, some books and articles that you know well, and others that you have only skimmed, into a coherent account of a literature, that can criticize existing work, to help us understand the field. This also happens to be the function of literature reviews and theory sections of your prospectus and journal articles. At Penn, this is a closed-book, timed exercise, which is more rare; but no matter the parameters, this is an important skill.
With that in mind, once you know enough to pass comps, you should not spend much more time on it. Sure it would be great to pass with flying colors, and I know some professors that may view passing with honors as important, but the opportunity costs of several of your summer weeks is significant. The sooner you realize that your time is precious, and that every semester, every summer matters, the better off you are.
POST-COMPS
1) With comps in your rear-view mirror, I would recommend the opposite approach to what I said for your first two years. At this point you need to become comfortable with focusing on a single subfield, and ideally a single topic. In these years, you reach the stage where you will know more about your topic than any member of your committee. They will be a tremendous resource, because they know how to produce research, and so much about the adjacent literature to your field, but it is incumbent upon you to know a single literature inside and out.
2) Find an advisor who's style you like. I generally think there's two paths to productive research. One starts with a question that you then track down the evidence to answer. The other one starts with data, and then you use this evidence to answer questions. My advisor, Marc Meredith, fell closer to the second camp. Both sides have advantages, so I won't pretend to know what you'll do, what I will say is to make sure you are comfortable with your advisor's approach, because he or she will need to approve nearly everything you do, so it's helpful if you're on the same page. If you NEED to start with a question because you are laser focused on that issue, it helps to have an advisor that has strategies to work that way as well.
3) That being said, try to be comfortable following your question where it goes. I started out by asking about polarization in state legislatures and ended up by writing about lobbyists in Congress. I would imagine this is a typical trajectory. Being able to move on from a dead-end topic is an important skill to have.
4) Writing a dissertation is a huge endeavor. Try to start by writing a single article. I'm a bit biased on this, because I finished with effectively a three paper dissertation; but even book-length dissertations usually have a key paper at their heart that the supporting chapters build around. And papers always talk about next steps in the discussion section: boom! There's an idea for your next chapter. Do that again and now you can see the structure of a dissertation. Again, this is not the only way to write a dissertation, but it's a way to get comfortable to take the first step.
5) Write like you're running out of time. This tip is not just an excuse for a Hamilton reference. It's a necessity, because you are in fact running out of time. Penn had a generous policy of funding all the students in our department for five years, which is not a given at most PhD departments. But when those five years are up, students would usually need to teach as an adjunct to pay the bills, so the end of that fifth year is a real deadline.
Also, in order to get another job you need to go on the academic job market, which is a six week process, that will also eat into your academic year, so you don't even really have five full years. So rush, rush, rush. Apply to conferences before the paper is done. Keep moving.
JOB MARKET TIPS
1) I won't pretend to be an expert on this, but I have spent a great deal of time the past two years working on this, so I do have some thoughts.
2) Try to pay attention early. Look at departments you would like to work at and see who got hired there most recently and read their work. Try to find your colleagues/competitors on the job market and read their work. See what's in vogue.
3) Cover letters are like anything else, you get better with practice. So ask for samples from people you trust and practice, share them with people, get feedback and keep working on it.
4) APSA interviews: You have to be very proactive to secure these, so I would recommend emailing departments that say they are doing these beginning in early August. You only need to attach your CV. They're short, but do make sure you have good questions prepared.
5) Skype/phone interviews: Be prepared to be more in-depth about a department. Resist the urge to overtalk. Test all of your equipment (especially if it's on Skype), you only have 20 or 30 minutes, don't waste them on "Can you hear me?"
6) Fly-outs. I haven't really processed this enough yet. Have your own bottle of water.
