Return to Richard's Grad Advice Page
As with all of my advice, this should be taken as supplementary to the official advice you can get from the University Dean of Graduate Studies web pages and our own School of Communication web pages for graduate students.
As with my general advice on making progress, don't let the comprehensive exams grow into a mountain that you cannot see yourself climbing. You don't want to take forever doing this. It is just a step you have to take in the path to getting a PhD. It is something that every PhD student has taken, and it doesn't need to be that scary. A semester to prepare and a semester to finalize your reading and then write them (i.e., 8 months) is all you need. Don't drag it out. Seriously.
I think a very good approach is to do a theory comp exam and a methods comp exam. In each one you should consider the following three elements, at least: the origins, including "classic" authors in the field, the current "state of the art," and the emerging issues.
In terms of length, I usually advise my students to prepare their comp areas as if they were preparing a course for a group of senior undergraduates or new graduate students. Cover the material that could reasonably be covered in 13 weeks - in other words, no more than 2 or 3 articles, chapters per week. Prepare a "lecture" for each week, that covers the material in the readings and draws out the main issues. Prepare an opening statement, as you would in an introductory lecture, and a closing statement - perhaps akin to the wrap up before a final exam. That kind of preparation has several benefits. First of all, you contain the project and your comp doesn't grow out of control since you have to think about what can be learned in a single semester by a group of students new to the topic. Second of all, you are preparing something that you can actually teach when you do graduate. Nothing like having a couple of courses "in hand" when you go on the job circuit. In the case of my students I ask that they actually prepare the course outline and all the lectures. This is outside the bounds of what is required in our department, so not all of them do it, but I think those that do find the investment of time well worth it.
I often work with my PhD students to see if they can actually teach one of these "comp courses," so it isn't just an academic exercise. This has worked very well in the past and I continue to encourage students to do this.
Many people like to think that they would prepare their comps as if they are chapters in their thesis. I think this is false economy. You inevitably change your topic/orientation and so the "chapter" you have created doesn't survive into the thesis and if it does it does so in a "cut and paste" form which results in a choppy and ill-formed document. And believe me, no one at a job talk wants to hear a chapter from your thesis...