Information for Students Interested in pursuing a Ph.D. with Professor Clifton

If you are not yet a Purdue student, see also my page for students desiring admission to Purdue.

When you are granted a Ph.D. by Purdue, the outside world expects two very different things:

  1. You are capable of, and prepared to, lead an independent research program, and
  2. You are prepared to teach college-level computer science.

The purpose of a Doctoral program is to get you to that point. When admitted, the faculty believe you have the potential to do the above. Turning the potential into reality demands effort from both you and the faculty. To that end, I give below some things that I expect of my Ph.D. students, and what you can expect of me.

What I Will Provide

Rest to be written...

What I Expect of You

Rest to be written...

Suggestions for preparing a Preliminary Exam document if you want me on your committee

When I see a preliminary document, I am looking for several things:

  1. That you have studied your research area sufficiently, and established that you know the history and current state of that area,
  2. That you have the intellectual ability and inclination to produce original and exciting results in the area,
  3. That you have the ideas necessary to produce a solid, high-quality dissertation on your chosen topic, and
  4. That you have a plan that will result in your being prepared for and capable of landing a tenure-track position at a research university, a post-doctoral position with the leaders in your area, or a position at a top research laboratory.

Most preliminary documents do a good job of the first two points: demonstrate knowledge of related work, and through presenting already completed work show that you are capable of producing good results in your chosen area. The third point is often weaker - does your document really demonstrate what you plan to do to complete your dissertation? The fourth point is rarely covered well.

My suggestion is to write your preliminary document as your dissertation outline (possibly even using the dissertation format), to at least the Chapter/section level. Several parts should already be complete, a related work/background section and one or two chapters drawn from work that you have already completed and submitted or had published. The remaining parts should have a brief description of the work to be completed, possible risks/pitfalls and what you will do if you run into problems, and a timeline for completion - at least at the Chapter level, if not at the section level.

To establish point four, I recommend that you tie the timeline to submission target dates, e.g., The material in Chapter four Sections 4.1-4.3 will be completed in October 20??, serving as the basis a submission to SIGMOD. The remainder of the chapter will be complete in July 20??, and the combined material will be submitted as a Journal article at that time.

Recommended Classes


This page last modified 10 September, 2009

Valid XHTML 1.1