Writing a Dissertation

Many students really would like to know what a dissertation should be composed of. I will try to clarify what is expect here.

In our research group, most of us feel that the dissertation is the end of a process that has been ongoing for the last many years. This means that the dissertation is not written in the last 6 months of the student’s life at the university – it is written over the years that the student is conducting the research. Why do we think this way? Well, no one except the committee and the student actually cares about the dissertation – no one outside of this select group will ever actually read it.

What that really means is that the student should be writing papers and submitting them to journals over the course of at least 2-3 years before they graduate. These papers should obviously be the meat of the dissertation, since, if they are published, they will be peer-reviewed, high quality research reports. This idea also accomplishes a few other things: (1) the student stays focused on getting research done, because they know that they have to complete a certain number of papers before they can graduate; (2) it spreads the writing out over a few years, instead of in a massive crush at the end; and (3) it guarantees that the research will be published, instead of relying on the student to transition hastily written material from a dissertation into a publishable quality paper after graduation. Further, if the student is constantly writing, they will get a feeling for what life as a research is really like!

In my research ground, I typically look for three or more papers completed before I allow them to write a dissertation. This is almost an arbitrary number, since some of the papers could be short and some could be long. I have had many students with three papers, and I have had at least one student who will graduate with at least five.

After the student completes three or more papers, then they should start crafting the dissertation. There are obvious hurtles that have to be passed through from the university, while others on the committee may have strong feelings about certain aspects of the dissertation. These basically boil down to:

1. The introduction. I, and others that I have known, have been all over the map on what to expect for the intro: (a) super thorough; (b) super skimpy; and (c) middle of the road. I think that I have come to grips with the fact that students who graduate are not going to be knowledgeable on every aspect of the general topic that they are studying. Systems are quite large, and students are highly focused. They can’t be expected to know everything. Also, when I have expected that, they tend to just include random facts on the topics that they don’t understand, and the questioning of them on these subjects is often unsatisfying and down right depressing. In some ways it distracts from the true purpose of the dissertation and defense – to determine whether the student knows how to conduct research and convey that research to the general community. Quizzing them on something that is tangentially related to their field should be done at qualification time. Once they are qualified, they have proven that they can memorize random facts. So, it is my opinion that the introduction should be easily read by a general audience (i.e., parents), and should give a good introduction to why their specific problem is important in the grand scheme of things. I would say that 20 double spaced pages is what I would aim for.

1b. Technique. Sometimes, if the student has used some data set or some model that not everyone in the universe understands intimately, then need to include a techniques or methodology section to the these. In some ways, this is introductory material, but is treated as a separate chapter. Some advisers want gory details on the model/technique, while others just want to make sure that the student has a basic understanding of it. The student should discuss this will the different committee members to make sure they know how much detail is desired of them.

2. The meat. This part can be quite controversial, since it is how to turn three to five papers into a thesis. In my opinion, this should be done as unceremoniously as using a stapler. In other people’s opinion, the papers should be rewritten in the style of the thesis. For example, the introduction and conclusions should be surgically removed and included in the main intro and conclusions. Further, all references to personal work should be changed to other chapters instead of student, 2010. I, personally, think that this is hogwash. But, there are others on the committee, typically. Be aware that some people feel strongly about this topic.

3. The conclusions. This should be a relatively short chapter (less than 10 pages) summarizing the student’s main contributions to the field. They should further discuss, briefly, what should be done in the future to further this line of research. It is quite simple to write the conclusions.

4. The abstract. While this is the shortest piece of writing in the thesis, it summarizes everything. It is quite difficult to write a good abstract. I would say that the abstract should, in some ways, be a list of the main conclusions. So, I would take a few introductory sentences, a few meat sentences, and then boil the conclusion down to a few sentences. Ta-da!

5. Grammar. There are a lot of people out there who are extremely picky. I would recommend having someone (professional) read the thesis to make sure periods, commas, and other things are in the right place and that all of the words are spelled correctly.

The student should also be aware that faculty appreciate having a lot of time to read the dissertation. Getting it in many parts is actually good. For example, published papers (which should be clean of grammar and spelling errors already!) should be passed out to the committee weeks before other sections. Then, as chapters are finished, and checked (i.e., finalized), they should be passed out. I can pretty much guarantee that committee members do not want to see drafts- no one wants to read 100+ pages of material more than once, except the student’s adviser.

Finally, it should be noted that not all research that the student conducted needs to actually be included in the thesis. If they did a single paper on a subject that is tangential to the other research papers that they are including, then they may want to consider not including it. There is no winner in the “who can write the longest thesis” contest. Also, faculty members would probably appreciate a more focused, shorter thesis, than one that is just chocked full of random studies.

So, with this recipe, students should be able to achieve dissertation nirvana.


About aaronridley

Professor at the University of Michigan, Department of Climate and Space Science and Engineering.
This entry was posted in Uncategorized and tagged , . Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s