Grad students who didn't quite finish their dissertations in the academic year but who have postdocs or other positions lined up for the fall are racing to defend and finish by the end of the summer.
These final exams are always kind of strange. In my department, they consist of a talk (30-50 minutes) followed by general questions from the audience, then a closed session of questions involving the student and committee. After all the questioning, the student leaves the room and paces the hall while the committee votes pass or fail. If the graduating student already has a job lined up, it would be bizarre to fail them on this exam, and I can't think of any cases in which this has happened here. As a committee member, I tend to use the exam time to discuss any parts of the thesis that are not yet published -- e.g., manuscripts in preparation or still in review -- so as to give the student feedback on these. Asking questions to probe the student's knowledge is pointless at this stage.
There's also no point in commenting on already published work, unless there are major problems with it.. I was on a PhD committee that encountered this situation not so long ago. It was great that the student had published a few chapters of the thesis, but none of us committee members (other than the advisor) had been shown the chapters/papers until just before the defence. The committee was very critical of the published work, which had some glaring errors that any one of us (other than the advisor) could have pointed out if we'd seen the manuscripts earlier. I felt like my time was being wasted by being on that committee and that the advisor and, to some extent the student, were being disrespectful of the expertise of the rest of the committee.
So, although I am fine with the final exam not really being an exam, I also don't want the whole experience to be a waste of time for me as a committee member. Reading a thesis and going to the defence takes a lot of time.
At the final exam, the talks can't begin to encompass all that a PhD student has done in their graduate career, so the student has to make decisions about what to present and how to present it. For some students, it is painful to leave anything out, as if the audience might think they didn't do much in their research if everything is not described. The best talks are a synthesis of the first-order information, explained in context, and only a limited amount of detailed description of the more interesting aspects of the methods and results.
One of the strangest parts of the final exam process is the introduction by the advisor. These introductions typically involve praise of the student's talents and anecdotes about the student's graduate career. Some faculty are a bit extreme with the warm-and-fuzzy intro. I prefer the introductions that are more professional, though not to the point of being completely bland and impersonal.
My introductions are affected somewhat by whether the student's family is sitting in the audience. One time I scuttled plans to mention (affectionately) an incident in which a graduating MS student, who was a notoriously bad driver, had accidentally smashed into my car in a parking lot while he was trying to park. When I saw his parents in the audience, I fortunately remembered that they didn't think their son's frequent fender-benders (and their effect on auto insurance rates) were very funny.
When I was finishing my PhD, I most certainly did not want my relatives to attend my defence. They already think that my research is strange and useless. Attending my defence would have given them solid evidence for this. I prefer that they just have a vague suspicion that what I do can't possibly be useful (whatever that means).
12 years ago