From talking to colleagues in a variety of fields and different institutions, there seems to be a huge variability in the ways committees handle their responsibilities of reviewing dissertation chapters. I would like to know two things:

  1. What do you consider the best practices for advisors, other committee members, and candidates in order to move the dissertation writing along efficiently?
  2. Does the Journal Review policy (specified below) seem reasonable to you?

Journal Review Method

In this model, the dissertation advisor treats his or her role as the chair of the candidate's committee as if he or she were the editor of a journal and the dissertation chapters articles submitted for review. The student submits a chapter to the advisor who decides whether it is ready to send for review. If it is, then the advisor sends the chapter to committee members for review. The committee members make a brief (2-4 page) report to the advisor that either: accepts the chapter as is, accepts it pending minor revisions, rejects it pending major revisions, or rejects the chapter entirely. The advisor then makes the final determination as to the status of the chapter. It will be the advisor's responsibility to make sure other committee members submit their reports in a timely fashion. The defense is held when the advisor judges all of the chapters to be "accepted".

The primary benefits of this policy are that it creates a clear organizational structure which allows candidates to receive prompt, actionable feedback on their work.

  • 1
    This looks to me like it would be a good Community Wiki candidate... good, relevant question, but asking for a "best prctices" list. Thoughts?
    – eykanal
    Apr 28, 2014 at 12:40
  • My thought is that this isn't a question about the best practices for the site here, but best practices for the departments that the members of this site are members of IRL. My hope is to collect some ideas to suggest to my department about how the process of handling dissertation review might be improved. Hopefully the information would also benefit other graduate students and professors in other departments. I think it deserves an exception.
    – user10636
    Apr 28, 2014 at 12:42
  • I fully believe it would, and I'm not looking to close this question.
    – eykanal
    Apr 28, 2014 at 12:45
  • Sorry! I misread your previous as wanting to move it to the meta board rather than wiki. I'm fine with that, but don't know how to move it. (I'm still kind of new here.)
    – user10636
    Apr 28, 2014 at 12:46
  • No problem. Just FYI, the implication of Wiki is that (1) no one gains any reputation from answers on the question and (2) editing requires lower reputation.
    – eykanal
    Apr 28, 2014 at 12:51

2 Answers 2


The policy is straight forward but an open question is what use a complete reject means if the advisor has deemed the chapter as passable. As I see it the advisor has (should have) both the insight into the details of the problem and research as well as general knowledge of academic expectations to deem when a manuscript is in shape for passing on. If a committee member does not think this is right then the question still remains who may be right and who may be wrong? Is the decision by vote in the end since there will likely be three or five persons involved in reading the chapters? In my system it is possible for a student to defend even if the advisor or committee member advise against it. Only fools do but it is possible. It is however, still possible for committee members to disagree but usually when issues arise it is because the student-advisor communication has failed for one or the other reason.

So in my view the "journal review method" is a good start but you also need to consider what will happen if there is disagreement and how to possibly weigh the input. Do all committee members have equal weight in all aspects or are they experts in some parts and therefore carry more weight in those chapters than in others. I think the system requires some tweaking in order to accommodate the clear differences that exist between a journal publications and a thesis (sensu monograph) where a thesis typically contains larger quantities of more detailed information than would be possible in a published paper.

As a side point, in my system where paper based theses are the norm, it is hard to "reject" a published paper; one can disagree but something that has passed peer review (albeit a poor one) has still passed. The key pint is thus the overall quality of the work and of the emerging scientist behind it. This i snot necessarily covered by the "journal review method" alone.

  • I was worried about what to do if there is a failure on the advisor's part, but I don't know how a general policy can fix that. Those sorts of failings are the kinds of things that the chair of the department, or some other departmental officer would have to have responsibility for.
    – user10636
    Apr 28, 2014 at 12:35
  • Maybe the use of the outright rejection would be to try to show the student that the direction of research on this particular point is a complete dead end? I am thinking that the advisor should have the final say, because ultimately the advisor should have the most say about whether the dissertation is passable or not, since he or she is supposed to be the authority on the topic.
    – user10636
    Apr 28, 2014 at 12:37

Here's what I think from the point of view of mathematics in the US.

For us the usual system is that the advisor is expected to read the thesis carefully, judging its correctness, significance, and novelty, at a level of depth comparable to or higher than would be expected of a journal review (which in mathematics is already pretty deep). The other committee members are expected only to give a more cursory reading, basically to satisfy themselves that the contents of the thesis appear to be mathematics research at an appropriate level.

As such, I think your system would face the following problems:

  1. Expertise. Mathematics as a field is highly specialized, and most of the members of the committee will not be experts in the same area as the candidate. They will not have the necessary background or expertise to carefully read and evaluate the dissertation at the level you propose. For a journal submission, referees can be chosen from anyone in the world, and even so there may only be a few dozen people whom an editor would consider well qualified for that particular topic. The intersection of that pool with the candidate's university is typically just the advisor.

  2. Workload. Even supposing the committee members to have the necessary expertise, what you propose would require a very significant time commitment from them. An average dissertation chapter in mathematics might be 20 or 30 pages of dense computation and logical argument. To read a paper of that length, evaluate its correctness and significance, and produce a 2-4 page referee report, a mathematician might easily spend at least 10-15 hours. Multiply that by 4-6 chapters and that is a lot of work, especially considering that a single person may be on several committees.

  • Let's say 4 chapters @ 10hrs each = 40hrs per dissertation. That seems quite reasonable to me in that: (1) the workload is spread out over a year or more in almost all cases, (2) readers on dissertation committees are standardly given course reductions to compensate them for time expended. Different universities calculate the reduction differently, but something like 1/3 of a course seems common. Hence, by serving as a dissertation reader three time one spends between 120 hours. That's less work than teaching a semester long class. Some dissertations will take longer, of course.
    – user10636
    Apr 28, 2014 at 15:59
  • 1
    @shane: Whether it's reasonable or not is an argument that one could make. It's certainly much more than they are currently asked to do. If committee members really did get course reductions I might agree with you, but no university I know of actually does that. The advisor might get a course reduction (usually not) but the other committee members never do, in my experience. Apr 28, 2014 at 16:06
  • It might also be a good idea to note that this answer is about the American system. As far as I can tell, the comittee has a different role and composition elsewhere (at least in Denmark). Apr 28, 2014 at 16:20
  • @TobiasKildetoft: Thanks, clarified. I see that the asker is in philosophy in the US. Apr 28, 2014 at 16:21
  • 2
    @shane: Maybe this is field dependent, but it's certainly not standard in mathematics. Apr 28, 2014 at 16:41

You must log in to answer this question.