Thesis Proposal Guidelines: a work in progress
Created by Tracy Hammond, Texas A&M University, hammond at cs dot tamu dot edu

---

Note that this is meant to complement http://www.cs.tamu.edu/academics/graduate/phd_requirements, so read that first.

LENGTH: There is no set length for a thesis proposal, but if I had to give an ideal length, I would say it should be 10-15 pages. But, say what you need to say. Mine was 86 pages. That is not to be a goal, shoot for 10-15 pages, but if it goes over, it is okay.

CONTENT: A thesis should state what you plan to do, why if you do it is worthy of a PhD, and evidence that it is possible.

CLARITY

· Should be clear and understandable to any educated person

· There should be no spelling mistakes

· There should be no grammar mistakes

· Writing should be clear and cogent

· Writing should be as brief as possible, i.e., not rambling

· Uses images when necessary to clarify ideas

· Sentences flow from one to the next

· Overall flow is good

· Does not use 'I'

FORMAT: You must have the proposal title page printed out and signed by all of you members. Document should be single-column, double-spaced to allow for easy review and commenting by your committee members.

THESIS STATEMENT: You must have a thesis statement that can be summed up in one to two sentences. What problem are you solving? What will the broader population learn or gain from your work?

CONTENT:

Abstract

· Place the abstract on its own page.

· This can be 1-3 paragraphs. It must be less than one page.

· Format must be single-column double-spaced.

· Motivate the problem

· Explain your thesis statement

· Explain what will be gained from the completion of your thesis

· Do not make this a story

· People should be able to *only* read your abstract and know exactly what you plan to do and why its important

· This is the most important part of your document

Introduction/Motivation:

· Introduce the area

· Describe the problem you are trying to solve

· State your thesis statement again

· Why is this problem important?

· Why is this problem worthy of a PhD?

· Provide any background information necessary to understand the problem

· Any intelligent person should be able to understand, and be motivated by, your problem

Previous work:

· List enough work so that your reviewers are confident that you know what related work has been done in the field

· Describe how your work differs from theirs (i.e. why their work does not solve the problem you are trying to solve)

· Does your previous work appropriately set the stage for your work? - Does it tell a story?

· Describe how the previous work leads to the conclusion that your thesis statement is necessary of attention

· Describe any pilot work you have done that shows that this is possible

Implementation/Methodology:

· What do you plan to do.

· You need to convince your reviewers that what you plan to do is possible of completion in a few years.

Evaluation:

· What does it mean for you to be successful at your thesis statement?

· How will you evaluate your work?

(Optional and will be removed from the thesis) Qualifications

· Do you have some very bizarre expertise that would not be held by a normal person in your position (i.e., a CS PhD student) that is necessary for the completion of your dissertation?

· This section is meant only to mitigate some risks defined below

· This section should be used sparingly, and ideally, all necessary info should be made obvious in the previous work

· This section should be short, a paragraph only

· Expect that if you do include this section, there is a possibility that I will ask you to remove this section and/or instead beef up your previous work.

· I did not include such a section in my proposal, however, I can provide two examples when it might have been appropriate

· If I were doing an ethnography on Africa on how people use technology, I might include here that I have a MS in Anthropology

· If I were building an algorithm to revolutionize investment banking, I might include that I worked for four years at Goldman Sachs and thus have a large network base of possible users.

Risks:

· What could go wrong?

· There should be some risks, otherwise, someone would have probably done it already

Backup Plan:

· What if you do everything you planned, but things go horribly wrong, is there an out plan that would still be worthy of a thesis?

· What are the contingencies for the risks defined above?

· Just saying I tried this and it didn’t work is not a thesis.

· In order to get a thesis you must teach something to the research community

· What important thing will you teach the world if things go wrong?

Time Table:

· Define the steps that you plan to take to solve your thesis and provide a time table for them

· The timing, while imperfect, should be appropriate

· The time table will be used by your committee to just if what you plan to do is possible by you, and if you truly have an idea of the difficulties of what you are undertaking

Broader Impacts:

· If you complete your work how will this affect the broader research community?

· This is slightly different than your motivation section, but you will probably feel some overlap, and it will be difficult to tease out the differences

· Think of this section as, after you have provided the community with the benefit described above, what next?

· For instance, you may want to talk about what future research will be possible after you accomplish your goal.

Conclusion:

· Summarize what you told them

· What were the key findings

· Similar to abstract, but you can assume people have read the paper

· What did you want people to get out of the paper?

· What should they walk away remembering?

Bibliography:

· References are properly cited

· References consist primarily of archival publications, as opposed to websites

· Expect to have 20+ references. Your final dissertation will have 100+ references.