standardized dissertations, part deux

I boil down a few arguments, my own and from the last round of comments, in favor of three essays as the default for academia:

1. In all sciences, most professional fields, and most social sciences, articles are standard. There are even humanities areas, like philosophy, where articles are standard. Even for book writers, articles are important. Most book writers do an article or two before jumping to the book.

2. The purpose of the dissertation is to show the ability to conduct research. Creativity is great – and should be rewarded with a degree – but the standard is normal science and competence.

3. The purpose of the graduate program is professional training – not an extended multi-year post-doc.

4. Standards actually protect students. It is too easy for faculty to hold students to unattainable standards, and drag them out for years. Moving goal posts is a real problem in graduate education. I’ve seen it happen too often. If there is a concrete standard, both students and faculty will know when “enough is enough.” There is a basis for appeal if professors are being unreasonable.

5. Three essays is a default, not a requirement. But still, in sociology, for example, the overwhelming majority should probably start with that unless they do ethnography. In other words, try a few articles. If your ideas *really* require more space, ok. But try the basics first.

Bottom line: Doctoral programs are about professional training and most academic professions focus on articles. That doesn’t mean that we should not allow more ambitious dissertations, but that should be reserved for a small minority of cases.

Adverts: From Black Power/Grad Skool Rulz 

Written by fabiorojas

May 31, 2013 at 12:01 am

12 Responses

Subscribe to comments with RSS.

  1. Regarding the issue of creativity, it’s not really clear to me why a “treatise-style” dissertation is, apparently, seen by some as inherently more creative than a three essays type dissertation. You can write a boring, conventional treatise style dissertation just as you can write an innovative and creative three essays style diss.



    May 31, 2013 at 12:33 am

  2. What JD said. Having the form fixed may even leave more room for creativity. You can focus on content rather than on structure and organization.

    More generally, I agree with Fabio on having article dissertations be the default (not only) option. Even most of the qualitative people I know who got great jobs out of grad school published articles first. I’m not going to do an exhaustive search to back this up, but off the top of my head Colin Jerolmack and Alice Goffman both got their jobs partly due to having ASR articles based on their fieldwork.


    Steve Vaisey

    May 31, 2013 at 12:02 pm

  3. In business school OT areas, I see mostly paper-based dissertations, which I’m fine with and agree in general with Fabio’s points. The questions I have though, primarily as a director of a PhD program, are process-focused. I would love to see people’s comments on the following:

    1. Which papers count? Are they separate from papers a student has previously worked on? Only sole-authored? Only lead-authored? Etc.

    2. What about the proposal? How do schools deal with a paper-based proposal. I know in some disciplines the approach is to have finished two of the three papers before defending a proposal. With a manuscript format, it’s relatively simple: the proposal include a question, lit review, proposed methods, discussion of contribution, and sometimes discussion of anticipated results. But with papers, what do you do?




    June 1, 2013 at 2:49 am

  4. Tom:

    1. In the social sciences (inc. management), the standard is clearly that the work is sole authored. Of course, this is loose. People can co-author a paper and then slightly modify it so the new stuff (however derivative) is sole authored. If lead authored – meaning that the student generated most of the material – then something similar can be submitted. This is probably one of those “case by case” issues.

    2. The proposal is very vaguely defined and there are few norms governing proposals in general, though some departments may have specific rules. The default for my students is three essays and I ask that students collect the data and do some preliminary analysis of at least one chapter. In other words, I sign off on the proposal when (a) there’s clearly something to analyze and (b) there’s a reasonable chance that it will yield at least one or two strong results. That means that I can assure employers that the student is well on the way to completing the dissertation by year’s end.



    June 1, 2013 at 3:28 am

  5. Standards can also “protect” people from ambition. IMHO the biggest problem in sociology graduate training at the top programs nowadays is lack of intellectual ambition – FAR from across the board, but it’s still a problem. Too often I’ve had students come to me and say “I’m going to do the three-article option. I don’t know what the research is going to be, but that seems like the easier approach.” I think the remedy lies in the thoughtful encouragement of serious research programmes, not in the format of the product.

    I therefore think we should maintain sufficient flexibility to accommodate ambitious research in sociology–which remains a discipline where books are important.



    June 1, 2013 at 12:02 pm

  6. The difference between a “book” dissertation and a “three article” dissertation is NOT ambition/scope. Many books have essentially one article’s worth of work in them. Either format can be very ambitious or can be minimalist. The difference is between a dissertation with one narrative arc and a dissertation that is a collection of smaller narrative arcs, between a dissertation with one one argument, one methods chapter and one theory chapter and a dissertation with multiple arguments and methods and theory sections in each chapter. In the extreme, the articles in the three article dissertation don’t have to be on the same topic; a legitimate three article dissertation could have the title “Collected Early Works of Jane Doe.” In my experience, the most common dissertation is a blend: there is a unifying theme and an overall master argument, but each analytic chapter stands alone is a publishable unit.



    June 1, 2013 at 2:54 pm

  7. I see a slightly different problem than Andrew: so much attention is put on publishing that students lose sight of the ultimate goal, of doing high-quality research. “Ambitious” is one attribute of high-quality research — per Andrew — but so are “careful” and “correct,” among others.

    We’re all complicit in this, though: if search committees didn’t place so much emphasis on flagship pubs in the search process, and/or if the discipline did a better job of filtering in good research and filtering out bad research from our journals, then the incentives for students would change toward doing good research and not just good-enough research.

    Of course, this would require more agreement on what constitutes high-quality research, which isn’t likely to happen. See the comments on the post about this year’s cultural soc section winners.



    June 1, 2013 at 5:07 pm

  8. Of course the difference between the two formats isn’t ambition or scope. But standardizing on one format — and privileging format over content — could have a negative effect on ambition, that’s my point.

    Agree, too, with Krippendorf. Where’s the post about the cultural sociology section winners? Thanks.



    June 2, 2013 at 12:04 pm

  9. Andrew: here . Comments are closed, and some of the comments that I remember aren’t visible. (I may be misremembering which post they were in, though. I didn’t follow the discussion all that closely.)



    June 2, 2013 at 2:13 pm

  10. I think krippendorf meant the post about the best org. theory articles of 2012, which you can find here.

    We started closing comments after a couple of weeks because they were getting a lot of spam. Comments aren’t closed due to the content of the comments or anything like that.


    brayden king

    June 2, 2013 at 2:57 pm

  11. To the extent that we think book-style dissertation represent more ambition or creativity (which they do not necessarily), we might be conflating an effect of peer review with length of text. A 30 or so page article will be read (hopefully) closely by 2-5 reviewers, probably at least twice. A book manuscript of 200 pages will also be read by 2-5 reviewers, maybe only once. Thus, the author can make more creative and ambitious statements in the book: each word is not as closely scrutinized.

    Of course, each word of a book doesn’t have the same value as each word in an article (as percentage of length), so the effects probably interact.



    June 2, 2013 at 7:20 pm

  12. […] les thèses de plus de 350 pages ! première partie et deuxième partie sur […]


Comments are closed.

%d bloggers like this: