What to expect from a “prospectus”

There is no sense in trying to provide a precise definition of what a prospectus should look like in terms of format or contents. Prospecti inevitably will vary significantly, depending on the nature of the dissertation project and the individuals involved. Nevertheless, it is useful to have some consensus about roughly what is expected from a prospectus.

The attached memos by Carol Heim, Sam Bowles and Steve Resnick and Rick Wolff -- with their similarities and differences of emphasis -- may serve to illustrate the rough parameters of a prospectus. As a key element, all three memos agree that a prospectus is not an almost-complete dissertation or an extensive literature survey. It is a relatively brief but clear description of the research project that is being proposed. The main purpose of the prospectus is to ensure that the dissertation project is “viable” in the sense that (i) it addresses interesting and well-defined question(s), (ii) the proposed methods are appropriate, and (iii) the student has the background and skills to carry out the project within a reasonable time frame.

I. What is a Prospectus?

A prospectus is a 20-30 page proposal for your dissertation. It describes the problem you have chosen to study and why it is important, as well as the specific questions you will pose and the method you will use to answer them. A prospectus is not a condensed version of your entire dissertation. It is primarily a plan for your work, rather than a report on your results.

There are many advantages to writing and defending a prospectus early in your graduate program. An agreed-upon framework for your dissertation is established with several faculty members who become knowledgeable about your project. Valuable feedback can be obtained and necessary modifications in your plans can be made before you proceed too far down blind alleys. Actually putting some ideas down on paper is usually extremely helpful in clarifying them.

An approved prospectus provides you with reasonable security that if you carry out the plans outlined, your dissertation will be accepted. It is not a binding commitment on your part to every detail of the prospectus. Many if not most dissertations end up looking significantly different from the prospectus. But getting the prospectus done allows you to get on with your research, and it is only through doing the research itself that you will discover how you wish to modify your original plans.

II. What Should a Prospectus Include?
A prospectus should include the following elements:

A. Statement of the Problem (including literature survey) B. Method
C. Discussion of Data (if relevant)
D. One-page Chapter Outline

E. Select Bibliography
A. Statement of the Problem

Indicate the problem you will study and why it is important. Include a brief discussion of relevant literature (though not the complete literature survey that your dissertation will contain). You should indicate both links to existing work, and the originality of your particular contribution. Is there an important debate in the literature upon which your study will shed further light? Has anyone worked on this problem before? If not, why not? If so, why does more work need to be done? Were there problems with the methodology employed by previous studies, or did they leave important questions unanswered? What will we know after reading your dissertation that we didn't know before? One good approach for this section is to move from a general discussion of the issues and literature to a specific statement of your question or

1

working hypothesis (or hypotheses). You can then go on smoothly to the section on Method. B. Method

Explain how you will answer your central question or test your main hypotheses. Will you be proving theorems, using econometric modeling, doing archival research, distributing survey questionnaires, and/or conducting interviews for in-depth case studies? Provide a description of your method.

Describe how your method will allow you to rule out possible alternative explanations. What specific results or evidence will enable you to establish your claims? Discuss briefly any potential shortcomings of the model or method, and corrections or adjustments you will make if possible.

C. Data

If your dissertation has an empirical component, discussion of the data is necessary to establish that your project is feasible. A prospectus should not present and analyze your data, but it should show that you have located specific data that will allow you to carry out the method you describe in B above. You should identify and discuss the data sources and their location, if not on campus; any processing of the data you will need to do to get it into an appropriate form for your analysis; and the ways in which the data you will use differ from the "ideal" data that you would use if it were available to answer your question or test your hypotheses.

D. Chapter Outline

Include at the end of the narrative a one-page chapter outline, with three or four subheadings under each chapter. The number of chapters will vary, of course, but a usual number will be five to eight, including an introduction and a concluding chapter.

E. Select Bibliography

The bibliography indicates some of the most important sources you will draw upon in writing the dissertation, and indicates that you are familiar with the literature. It is not only a bibliography of sources used or cited in the prospectus itself, as if the prospectus were a paper. Include all works you have cited specifically in your literature survey as well as other important related works, and be sure to include citations for all of the main data sources you will be using. The bibliography does not need to be longer than three to five pages.

III. When and How to Start Writing a Prospectus

2

Now!

While taking field courses, jot down separately from your notes on the readings any ideas that seem interesting to you, or that might be the basis of a research project. (Save all of these, including the ones you do not address in your dissertation. They will be valuable for future work.) Field courses cover the literature in a field--what holes does that literature contain that you might fill?

Ask visiting seminar speakers whether anyone they know is doing research on the specific questions that interest you. Look at journals in your field to identify current debates, and the kind of data or method that scholars in your field consider appropriate for answering certain kinds of questions. Call or e-mail them about their work. Talk to faculty members and other students.

Don't hesitate to speak to people at conferences or seminars. Most will be pleased that you are interested in talking with them. If possible, do some background work--have specific questions to ask them, and be familiar with some of their work if they have published in the field. But don't assume that you need to have a fully articulated and polished proposal in order to speak with them. Ask people with whom you speak "can you think of anyone else with whom I should talk?"

After an initial conversation with a faculty member at UMass, it often is most productive to come back with a few things written down. Perhaps show it to a friend first, then drop it off for the faculty member to look over before the conversation. This need not be a prospectus or even a draft of one. It could be titled, if anything, "Some Preliminary Thoughts on . . ." But many people find even rough notes a helpful basis for discussion.

Write down chapter outlines of the dissertation, or outlines of individual essays if you are thinking of doing a three-essay dissertation, early and often. It's an excellent way to get a sense of the project or essay as a whole, and how its various parts will fit together.

Most important, however, is seeing your advisor on a regular basis and discussing your ideas with him or her. Each of these meetings will be most productive if you can provide working notes or a progress report prior to the meeting. Expect input from your advisor and make a genuine effort to respond to it. Your advisor may encourage you to move in directions you had not considered and you should take these suggestions very seriously. With your advisor, set specific dates for completion of particular tasks.

Good luck!

Carol E. Heim
March 2006 (revised)

3

From: Sam Bowles
Concerning: some idiosyncratic ideas about the prospectus Date: 21 July, 1999

I’m often asked what a good prospectus should do. I guess that every faculty member has a slightly different view on this, which is why I stress the idiosyncracy of these comments. The best way to clarify what a prospectus should be is to look at a number of successfully defended prospectuses

The purpose of the prospectus is to determine if the argument of the proposed thesis is viable in the sense that with a reasonable amount of work (given the capabilities of the candidate) it seems likely to yield a valuable contribution to knowledge in a form acceptable as a doctoral dissertation. The prospectus and the defense are thus a way of insuring candidates against the risk of embarking on a major commitment of time without some consensus on the committee that the project is promising and the candidate prepared adequately to undertake it.

A good prospectus is one which will allow an intelligent judgement to be made on the viability of the project. This means that it will reveal not only the strengths of the project but also its shortcomings and unresolved problems. Thus it should include (prominently, in the first few pages) an argument which is coherent and would be interesting, if true. Often this argument will be of a negative nature, namely to show that a widely accepted claim by reputable economists is false, and that a more convincing claim is supportable. If there is any doubt about it, the importance of this demonstration (its implications for policy, or for other important claims, for example) should be made clear. (This may justify including some review of the relevant literature, but literature reviews should be addressed precisely to the questions to be raised, not an aimless meandering through all the stuff you had to read to get to where you are.)

Then the prospectus should carefully develop reasons for believing that the questions you wish to ask can be answered satisfactorily by some reasonable standard. This part will normally include research of an exploratory and preliminary kind, some initial regression equations or data analysis, for example. If a new data set is to be used you must demonstrate that you have access to it and know enough about it to have confidence that you can use it for the purposes you propose (one way to do this convincingly is to do some initial data analysis with the data set.)

If you are going to collect new data you should indicate why you think it will allow you to test the hypotheses you are interested in . It helps if you are quite specific

1

in stating exactly which hypothesis will be tested for example by which estimated coefficients, in the cases of econometric work.

In thinking about how to write your prospectus, imagine that you are writing it for a very well read specialist in your field who holds a viewpoint different from your own, but who is not hostile to your argument. Think of all the objections that a knowledgeable critic might raise. You will presumably not be able to answer all of these objections, and your prospectus should be candid about where you have doubts (otherwise the committee may mistakenly approve the document and send you off on a wasted two year project resulting in an indefensible dissertation.)

When I review a prospectus I often ask three questions: is there an argument here which might eventually be published in a leading academic journal in the relevant field? does the candidate have the skills (language, mathematical, econometric, country knowledge) to carry out the project in a scholarly way? and...does the argument connect (logically, metaphorically) to matters of importance?

Remember, the dissertation should do a job for you; its not your life. It should help you find the position you want and to publish a couple of first rate articles. It is not the first draft of your collected works. A dissertation of 60 pages is more than enough if the pages are good. It is to be -- in the old guild usage -- your masterpiece: that which you use to demonstrate your capacity to become a master in your trade. But like the medieval sense of a masterpiece, it should not be confused with your life's work, or your best work. And contrary to the suggestion of the more modern usage of the term masterpiece, it does not have to be perfect.

Being thoroughly familiar with the literature in your field, particularly the new work, not sufficient, but it is necessary. So you should do two things. Spend some time reading the relevant articles over the past five years or so in the major journals in your field as well as the top journals in economics (the Journal of Economic Perspectives is a good place to start, along with the Journal of Economic Literature). As important, carefully check over the lists of available working papers from the main institutions contributing to your field. Working papers are leading indicators of what will be in the journals two years hence. Get any and all working papers which seem possibly related, read them and follow up their references.

Finally, when you have an idea that interests you, don’t be too hard on it. New ideas, like infant industries, need a little protection. If it excites you, give it a chance, and don’t shoot it down with “somebody must have already done this” or “this is obvious,” or finding fault with it because you haven’t had time to develop the idea

2

fully.

It takes some courage to try to write something new about something you care about, and to expose your ideas to well informed critics that do not share your priors. But that' s what you have to do. That's what science is all about.

3

Thoughts on producing a prospectus Rick Wolff and Stephen Resnick

Although we have supervised many dissertations over many years, we doubt that any one set of rules would work in some general way to yield a defendable prospectus. So we try in what follows simply to provide some insights derived from our dissertation supervising in the hope that they may facilitate the prospectus process. We assume that these insights will need all sorts of modifications in particular cases.

Two aspects of prospectus writing are obvious and yet also merit comment. A graduate student must start to write the prospectus and then must finish. Both tasks can prove to be difficult and many students find one to be much more difficult than the other. Our experience suggests the following kind of writing/research strategies for graduate students to deal with these two aspects.

1. At an earliest stage, be open to different broadly defined dissertation ideas, be willing to play with them (i.e. to consider, discard, and modify them in your mind), and discuss them with fellow graduate students and interested faculty. Pay special attention to those ideas/topics that seem to hold your interest and hopefully excite you. We say this because the writing of a prospectus and dissertation is hard, often lonely work with very little if any emotional and/or financial support in and by this society. Too often others – distant and even close family members, friends from your other life, neighbors, and so forth - will have very little if any understanding of what it is you are doing. Hence to deal with all of this, you will need more than anything else an idea/topic that has real meaning and importance for you. Believing in your project’s larger importance can sustain you through to completion when the inevitable problems of the dissertation and life in general recur.

2. Next, meet with your prospective chair or co-chairs and discuss these ideas with the initial goal of gradually narrowing the choice of topic to a very few or if possible only to one. Whether the graduate student makes the choice of topic with others’ inputs or on her/his own, once the choice is made, we have found it useful to ask the graduate student for two or three pages summarizing (a) the chosen topic, and (b) the rationale whereby the student chose it. The student’s response has usually been revealing in important ways. For example, it has proven very significant if these pages prove very difficult to write or cannot be produced. As soon as possible such a situation should become the object of discussions between the student and his/her dissertation committee. The goal of those discussions is either to arrive at a different topic choice or else at new way to resume the student’s work on producing a few pages on the old topic. We have found the latter to be what most often happens. What our experience has taught us is that the worst possible outcome is lack of contact between the student and the collaborating faculty; that can extend a temporary problem into a long delay in getting the prospectus writing underway.

3. Once these initial few pages are completed (and regardless of how many revisions that may require so that student and faculty advisors are satisfied), the next step is to expand

1

them into a full prospectus. We have found a twenty page prospectus, not counting footnotes, appendixes, or references, to be sufficient. What has served best as a guide for students is to conceive of the prospectus as a plan of operation: an outline showing how the proposed dissertation will structure and develop the central arguments in the student’s chosen topic. It is important not to confuse the prospectus with the dissertation itself. The prospectus does not make the arguments; it simply outlines their structure and basic logic of development.

4. We have found that certain problems constantly confront students working on prospectuses. For example, it is not easy to identify and clearly specify the proposed dissertation’s central idea or question. It is likewise difficult to distinguish between central and sub-themes. Yet, reaching clarity on these central dimensions of the prospectus has, in our experience, repeatedly been a key moment in the evolution of a successful dissertation. Not struggling through to a prospectus that is clear on this basic logical structure produces larger and more intractable problems later during the research and writing of the dissertation itself. To take another example, students understandably find it difficulty to give up on relatively more marginal aspects of their dissertation topic that are interesting, on which they have spent research time and energy, etc, yet are aspects not central to what has finally emerged as the dissertation’s core contributions. Tough decisions need to be made to delimit the scope of the dissertation – and to express these limits in the prospectus – in order to keep the dissertation under control and within boundaries of what the student can produce in a reasonable amount of time. Dropping certain subthemes can be a difficult but crucial moment in the crafting of a successful prospectus. To take a final example, once a student has defined and written into the prospectus both the central theme(s) and the sub-themes to be treated in the dissertation, it is crucial for the prospectus to specify as well how theme and sub-theme will be connected in the dissertation. Clarity on this at the prospectus stage will save great distress later in the dissertation-writing stage. Hence it is a pre-requisite for us to agree to a defense of the prospectus.

5. The prospectus must state at its outset why the writer thinks the chosen topic is interesting and important. The point here is to persuade the reader (and the writer too) of the worth of the project. If done well, this task will evolve into the writer’s eventual prospective job seminar. In explaining the worth of the research, it is helpful to show how its prospective results differ from and/or provide an interesting extension of what already exists in the relevant literature. Doing this may well take five to six pages of a twenty page prospectus and perhaps an entire chapter of the proposed dissertation. (Of course, in a history of thought dissertation, it will take that many more prospectus pages and dissertation chapters.) Additionally, it may be useful for the prospectus to include in a concluding section a brief yet informed speculation in regard to consequences of the dissertation work for others’ research.

6. Based on the above points, we offer some specific suggestions as to the contents and page breakdowns of a prospectus.

2

- The opening two or three pages should be devoted to stating what the central theme or problem in the dissertation is and how and in what ways the related sub-themes connect to produce a logical and clear study.

- The remaining pages offer a plan of what will be done to explore concretely that identified theme/problem. We think it useful to break this plan down into proposed dissertation chapters. Perhaps five pages of explanation for each proposed chapter. Our experience suggests that three core dissertation chapters - amounting to 15 pages of prospectus explanation - to be sufficient. Again, the main aim of the writer over these pages is to inform the committee what is planned. Nonetheless, we also are aware that it is next to impossible to avoid actually doing some of the expected analysis. To control the prospectus writing - meaning to get it finished in a reasonable time period – we suggest that it present only a glimpse into what the analysis might look like. We cannot give any exact rule here, but in the past we have suggested to students that they offer a bit of analysis to feel confident that the project is viable. Finally, once the 15 pages are completed, a couple of pages summarizing the research including its importance suffice.

- When empirical work is part of a dissertation, we ask the student to add to the prospectus an appendix of a few pages outlining the methodology to be employed. For example, some students have wanted to conduct samples of a population - to provide empirical information on the class nature of, say, an agricultural region or some particular population grouping. We ask them to include an appendix attached to the prospectus outlining their plan of empirical research. Other times proposed empirical work can be folded into the prospectus itself without any appendix. For example, there may be a dissertation chapter that offers an empirical story with supporting data that elaborates the central theme of the dissertation. In these cases, we ask the student to include in the prospectus a brief discussion of how that empirical research will be carried out and how it will elaborate the theoretical argument made in other chapters.

Gerald Epstein
Revised, April 30, 2007

1. What is the question?
Note: There may be more than one.

If there are "stylized facts" that prompt/relate to the question(s), then lay out the key "stylized facts".

2. Why is it important?

In principle there are several dimensions to this (i.e., theoretical, empirical, policy) but focus on the most relevant ones to your prospectus.

3. What are the standard answers to the question (s)? 4. What is wrong with them?

Maybe they are not completely wrong; maybe there are some lacunae in them that need filling.

5. What is/are your alternative answer(s) to the question(s)?

Another way to put this, is what is/are the main hypotheses you are proposing to explain the phenomena at issue?

6. What prima facie (i.e., initial) reasons to have to believe that your answer(s) might be correct? 7. How have previous analysts studied the phenomena (answered the question(s)?

8. How are you going to answer the question(s) and how will it differ be similar to other work (eg. Econometric modeling, surveys?, simulations? What?)

9. Show that these methods are suited to answering your questions. (investigating/testing your hypotheses).

9. What data sources will you use to do your work. What holes still need to be filled. 10. What do you expect the contribution of your work to be?
11. How will your work be divided into chapters for the dissertation?