My advice for thinking about thesis and dissertation writing

My thoughts on how to approach and write a thesis or a dissertation can be found here. I particularly recommend downloading and viewing the presentation file.

Perlmutter's advice for PhD dissertations

Here is a series of articles on writing your dissertation and using it to advance your career from David Perlmutter, who has written on the topic for The Chronicle of Higher Education for many years.

  1. The Completion Agenda, Part 1
  2. The Completion Agenda, Part 2: The Best Defense
  3. The Completion Agenda, Part 3: Revising Your Dissertation
  4. The Completion Agenda, Part 4: Finishing and the Job Hunt

Choosing a dissertation topic

Here is a link to an article on choosing a dissertation topic. It includes ideas for where to find dissertation topics and some guidelines for doing replication studies.

And here is the classic animation of what your contribution to knowledge as a dissertation writer:

Dissertation Support Group

Here is a link about forming a dissertation support group to help you get through the arduous process of producing your dissertation.

Fainstein's guidelines for PhD thesis proposals and dissertations

[I am in the process of organizing my files and found this useful guide for preparing proposals and dissertations prepared by Susan Fainstein. I found them useful. Perhaps you will, too.]


The doctoral dissertation requirement is intended both to prepare graduate students for the role of professional scholar and to assess their prospects for successful performance in such a role. Many students experience considerable difficulty in expeditiously completing the thesis. These guidelines are intended to assist the process.

Some Guidelines for Selection of Dissertation Problem

A frequent cause of delay in getting started on a dissertation, and a major source of frustration thereafter in many cases, is the choice of a dissertation topic. The prospects for such difficulties will be diminished if the following suggestions are borne in mind.

  1. An area of study is not a topic: do not propose simply to write a thesis on regional planning, environmental constraints, or health policy. To confuse a broad area of study with a thesis problem is to confuse a dissertation project with a career.
  2. Select a problem within an area of study with which you are already familiar, rather than an area with which you would like to become familiar. (As you take courses, you should try to write term papers within the area in which you expect to write your thesis.) A dissertation is supposed to advance the frontier of a field of knowledge, and it usually takes one or more years to become familiar enough with the literature in any area to know its frontiers moderately well. The thesis problem should be formulated as an argument, or set of propositions, though it need not be stated as formal hypotheses. It should be of general interest rather than a narrow test of a trivial point.
  3. Build on prior research and theory. Determine what the relevant literature contains before you formulate your problem in final form. Consider the conclusions of writings that interest you and compare your approach and findings with them. You should indicate the implications of your topic to advancement of understanding in your subject area. You should also indicate its policy implications. Extending the generality or establishing the limits of previous conclusions may involve viewing them from a different theoretical perspective from that which guided their author, using different indicators or procedures, or simply sampling a different universe. The objective, in any case, is to pick up the task where previous investigators concluded it, and carry their work distinctly further.
  4. Do not insist on a magnum opus, something that will completely discredit a major work or be a new classic, or forge a wholly new area of debate. Remember that thesis topics are usually relatively narrow. The point is to demonstrate your ability to do in-depth research and to use it to support an argument. Excessive ambition at the thesis stage is probably the tragic cause of more unfinished graduate degree pursuits than anything else. Frequently when students realize that their thesis is not so great after they have spent much time on it, they neurotically set still higher standards for themselves, feeling more compelled than ever to write a thesis that will make up for lost time by propelling them immediately from the level of overdue degree candidate to distinguished member of the academic profession. One good piece of advice: no topic or question is a bad one. What matters is what you do with it. Modesty and maximum care in the preparation of the thesis proposal are the best guarantees of rapid mobility to an employment status that provides opportunities and resources for moving as quickly toward distinction as your qualities merit.
  5. Theses often develop out of research assistantships, where a graduate student is working for a professor or research organization and is allowed to use part of this work for her/his thesis. This has many advantages for the student and the faculty member, but it also poses important ethical problems for both. The advantages, of course, are that the student gets advice, resources, and pay for dissertation preparation that s/he might not otherwise be able to acquire, while the employer gets extra motivation and effort from the student. The problem for the student and faculty member is to assure that the student's thesis represents a distinct creative and independent effort. Complete independence may be an impossible goal. Even without employment, a student may be lucky enough to have a specific thesis problem suggested by a faculty member or someone else. The student should demonstrate an appreciable amount of initiative, creativity, and responsibility in all major steps of carrying out the research, relating the problem to prior theory and research, planning research procedures, carrying out the research operations or supervising those who do, and writing it up. What constitutes an "appreciable amount" will have to be assessed by the student and faculty member separately in each project.

Preparing the Dissertation Proposal

The dissertation is supervised by a committee of five or more faculty members, three of whom should be members of the PhD Committee in Urban Planning and two of whom should represent another discipline. (These "outside" members need not be Columbia faculty members.) The student should discuss the proposal with the faculty sponsor at an early stage. At this time, or soon thereafter, the student ought to give the sponsor a preliminary abstract or outline of the proposal to serve as a basis of discussion. The additional members will be invited to join the Thesis Committee after consultation with the sponsor.

After the thesis committee has been formed and its chair has approved the proposal, the student presents the finished proposal to students and faculty in the Program.

The proposal, not to exceed 20 pages in length, should have the following sections (which are modeled on a research grant proposal format):

  • Abstract A 2-3 page summary.
  • Title: This should be brief--less than 10 words, in most cases—yet indicative of both aims and operations.
  • Introduction: This should state succinctly exactly what you wish to investigate, why, and how. For example, it can state the major hypotheses and indicate the theory and/or prior research from which they are derived, or it can describe the dimensions of the phenomena to be explored and indicate the contribution you expect to make to the development of theory.
  • Theoretical Foundations and Significance: This should summarize and cite relevant literature, it should describe and assess all research which tests or nadds to theory in the field, and it should indicate how the proposed research will provide a contribution. This section should be concise, yet have as much length and thoroughness as is necessary to establish that the proposed research will advance existing knowledge. While length and style may vary, precision, clarity, and tightly logical organization will enhance this section.
  • Methods or Procedures: The more carefully and completely these are set forth, the easier and more significant the thesis is likely to be. This is the section of the thesis that the committee will examine the most closely, and it is important to spell it out in detail and to show how the findings that will be produced relate to the research questions. Moreover, detailed planning permits you to anticipate and avoid difficulties which otherwise might only become manifest after they render a large amount of prior time and effort worthless. Also, this section should set forth a realistic schedule for completion of each stage of the procedures. Supportive material on details of procedure, instrumentation and sampling, and any highly detailed summation of relevant literature should be attached as appendices.
  • Resources Available: This should be a brief statement of how you expect to get access to necessary subjects, data for secondary analysis, sponsorship, funds, computer services, or whatever other resources will be needed. These needs, of course, vary greatly from one topic to the next.
  • Results: State your anticipated findings. Indicate how the results of your specific research will line up to the general questions you stated earlier. Every graduate thesis should be oriented to academic publication, either as one or more articles or as a book, and publication is encouraged. This section of the proposal should also set forth a chapter outline of the thesis.

NOTE: Dissertation proposals should be no longer than 20 pages (not including the abstract). You should limit the bibliography to major works, the list not to exceed three pages.


Standards of style are set forth in Kate L. Turabian, A Manual for Writers of Dissertations (University of Chicago Press). The student should be alert to taking into account all University requirements in addition to these departmental suggestions. These are available on the University's webpage. An abstract not exceeding 600 words must be submitted with the thesis. Often drafting the abstract early helps in organizing the dissertation writing, as it makes it easier "to see the forest from the trees."

Some Further Suggestions

A few more bits of advice may be helpful. One is that the secret of most great writing is rewriting. Also, most great research probably involved revisions of procedures while the research was in progress. The more you are willing to analyze your plans and your preliminary findings as you go along, the more you are likely to anticipate pitfalls or augment the quality of your final project. Seek consultation from your adviser and others you deem appropriate, especially before you make a major revision. Keep all of your committee members posted on your progress and allow them to choose whether they wish to comment chapter by chapter or after receiving the entire first draft. Indeed, drastic deviation from the proposal without advance approval from your sponsor at least, and possibly the entire committee, may be grounds for rejecting the dissertation. Finally, in writing your thesis, you should include reports on problems encountered and suggestions as to how they might be avoided. The dissertation is supposed to be a learning experience, so you should report all the important things that you learn, either in the text or in appendices.

The actual work of stating a set of findings is a long and arduous process. It is good to keep a log of research operations and problems. Set up a filing system for the materials you collect right at the start. Be sure to maintain a bibliographical data base as you go along and to key your notes to their anticipated textual use. Allow a long time for analysis and interpretation of the data once it is in hand, rather than deluding yourself that the thesis will "write itself" once the data are in. A long piece of writing differs in many respects from shorter efforts. One frequently forgets what one has included earlier and loses track of how one intended to use something. The more careful you are about making detailed outlines, using topic headings and summary paragraphs, and devising monitoring systems, the easier your work will be.

A final warning is not to let your thesis grow appreciably after it is approved. There is a great temptation to add more questions, sample another population, or otherwise move what was originally manageable and satisfactory to something unmanageable. Diffuseness is usually the result of initial failure to think carefully through the kinds of findings one is seeking. Length is not a measure of quality. Many good theses have been presented in 150 pages or less. Just because you collected a piece of information or did an interview is not reason enough to include it. Much of the material you accumulate should never be used.

It is a good idea to look at completed theses to get an idea of what a thesis looks like.

If you are seeking funding, you must plan on at least a 6-month lead time between when you apply and when you actually receive support. If your thesis involves field work, you must take this into account.

It is crucial that you give your dissertation priority over other things. Particularly if you have family responsibilities, you must decide what is truly important ("quality time" with your children), and what is not (baking cupcakes for your child's birthday). Any service that you can buy, do so if you can afford to or do without it if you possibly can. Even if you are good at painting and like to do it, defer doing your own decorating. Try not to be in a position where you have to do your data analysis and/or your writing while on your first full-time job. Usually the job proves demanding, and this is another source of unfinished or excessively delayed theses. Thereby, it is often a source of delayed or permanently limited upward mobility on the job.

Perhaps the most important advice one can give, both for the dissertation and for subsequent research and writing, is that you should not feel that your entire identity depends on the reaction that any particular piece of work elicits. This is the major cause of non-productivity in both graduate student and academic faculty life. Set yourself a deadline and present the best report you can muster at that time, regardless. The pressure will help move you to closure where you have been indecisive or hypercritical. Often committing yourself to give an oral presentation on your research is a useful way to impose deadlines on yourself.

The best way to approach a thesis is to regard it matter-of-factly as a job to be completed. Avoid being overly emotional about it. Do not look on your thesis adviser as a surrogate parent. Try to respond to interim criticisms constructively rather than defensively. If you can translate your research scheme into a mechanical set of tasks rather than an ordeal by fire, you will have a minimum of difficulty.

GSIS requirements

Dissertation proposal defense

  • Documents to submit generally include:
    1. Outline
    2. Roughly ten page (double-spaced) summary. The breakdown is probably something along these lines: Introduction and question (one page), Lit review (five pages), Methodology (three pages), and expected findings (one page).
    3. References
  • Jury members
    1. At least three people are required.
    2. Proposal defense members need not be members of the dissertation defense.
    3. Physical presence is not mandatory.
    4. Defense is open to other students and professors.