How to choose a topic for a thesis or dissertation is an important issue. This post answers a question from a student (you know who you are!). It’s mostly a compilation of experience about visualization, proposal and creation of your thesis or dissertation. Some students fear (or panic) when the time of thesis arrives. I’ve even met people with the EBT (Everything But the Thesis) syndrome, but that is a very complex problem and I’m no psychologist. Please, keep reading if you are not afraid of the document we refer to as “thesis”, and you are determined to succeed. Effectively, the thesis or dissertation represents a written expression of your specialized knowledge, with an scope scarcely above that of the toughest work you fulfilled during your studies (sometimes not even that). Nothing out of the ordinary. It is a document in which you express, typically as the last requirement for the degree, your command over a concrete subject. Often, the thesis is a document of regular transcendence.

The problem lies in the huge load of stress students have to endure, stemming from the academic and social environments: you have to deliver a work with superlative quality (and has to distinguish you from your peers), the haste to innovation, the won (or lost) prestige, the opportunities, the castles in the air, the afterward, failure poking its head out of the window, the defense, among other factors. But sometimes, to tell the truth, it’s simply that the student has acquired no significant skill in the career’s subjects (not so unlikely as it may seem), or suffers from a traditional chronic laziness. Nevertheless, for the time being we will focus on the essence of the document. Upcoming posts (perhaps) will touch on those surrounding topics.

1. How to choose a topic for a thesis?

Let’s try to unravel the document’s mistery. Best dissertations are done by those researchers very knowledgeable about their topic and about research techniques and methodologies. They know that every study should start by investigating the studied object. In our case, we must first establish what a thesis is, and what its purpose is. Here the average student answers, with praiseworthy self-confidence, that a thesis is a document whose exclusive purpose is to achieve, finally, the degree. Bravo, it’s impossible to be clearer, but we must never forget that, in life, milestones or stages are not as important as the paths and transits between such stages. Yet, something stands out in such answer:

You have to realize that your goal is to fulfill an academic requirement

However, this by no means allows you to take a lot of liberties, or to face up to your research half-heartedly. It’s but a reason to act with humility. There are three key words for thesis’ success:

Humility, Communication and Balance

Irrespective of the importance that you confer to your investigation, at no time you should work thinking of drawing international attention to you, or dreaming of that incredibly remunerated job you will obtain, or about the avalanche of trophies and medals. If you distract your mind with this, at some time you are going to lose your concentration, and the project will just slip out of your hands. Please, remember that the thesis represents, simultaneously, discipline and learning exercises, and you must assume it like so. The composure and integrity you show while walking your paths is more important than the path itself, and than the destination. Finally, realize that a thesis is an exercise of communication. And, never give too much, never ask for too much.

2. Preliminaries

The word thesis derives from Greek θέσις (position). The first meaning of the American Heritage Dictionary is, for our intentions, the best one:

“A proposition that is maintained by argument.”

Implicit in this definition is the logical linking of ideas (reasoning) to arrive to a valid conclusion. In other words, the thesis presents/explains a series of inferences, from premises or facts, to finalize with a conclusion or set of conclusions. As people overly concentrate their anxiety in the thesis, they forget something even more important: the hypothesis. If you want to know the cornerstone for how to choose a topic for a thesis, it’s this: the hypothesis. Hypothesis, literally, is what it is located below the thesis, and it is not indeed the writing-desk of the tutor or the jury :-). The AHD provides three equally useful entries:

1. “A tentative explanation for an observation, phenomenon, or scientific problem that can be tested by further investigation.”
2. “Something taken to be true for the purpose of argument or investigation; an assumption.”
3. “The antecedent of a conditional statement.”

Interpreting, what underlies every thesis is the hypothesis, your stem research idea, the guess you want to prove with your work.

3. Genesis
As said, the base of the investigation is the hypothesis:

Hypothesis is anticipation

We will choose a path that, we anticipate, will lead us to a predefined point. Here the student must show confidence. Students cannot let the poorly understood academic rigor to strangle them: at this time, imagination and intuition establish the success. Step forward, you can doubt the establishment, weave:

…My friend, I shall be pedagogic,
And say you ought to start with Logic…
…
The philosopher comes with analysis
and proves it had to be like this;
The first was so, the second so,
And hence the third and fourth was so,
And were not the first and second here,
Then the third and fourth could never appear.
That is what all the students believe,
But they have never learned to weave.

J.W.V. Goethe, Faust, Translation to English by Walter Kaufmann, 1962.

Basically, the hypothesis must contain your preconceived ideas on a subject. It means the point to begin with. It represents what you want to prove with your investigation. In a more personal tone, the formulation of the hypothesis should answer the following question: what is “that” I want to demonstrate with my work? Some caveats apply:

• The hypothesis should not lean in other hypotheses.
• Humility: All hypothesis must be verifiable (the thesis will verify or deny it) and in agreement with your resources. Be realistic. You have limited time and access to sources.
• Communication: Other researchers in your topic must understand your statements clearly.
• Balance: Neither very ambitious nor very simple.

4. The Topic, at last

Let’s be practical. A hypothesis is located within some subject or specific matter. To firmly discover and confirm the preference for a thesis subject may require a long time. It depends on the student. I know people who even before entering the university already had idea on their topic of investigation. Excellent. But some, regrettably, arrive to the last year of studies, to the last term, without a clear idea on what to research. If the cause of such ignorance is they have an abundant list of subjects that they would like to investigate, then the rest of this post does not apply. Select the subject you like the most, and start digging.

I’ll be brutally frank.

If you are finishing your courses and you are absolutely clueless about your thesis’ topic… besides being a bad student, you are a survivor. Period. 🙂

In any case, I do strongly recommend preparation of a document (physical or electronic), where you must write down those subjects that you find interesting. But don’t think of them as tentative subjects for thesis (yet). They only are interesting subjects, in which you would like to work, topics that you are at ease with, or in which you have identified some opportunity or deficiency in the established methods. The sooner you begin, the better. Ideally, you should not leave this for the last year. And please, write down the subjects. Don’t trust your memory so much. Don’t be like those students of Algebra who think that only reading proofs, without writing proofs, suffices to approve the final exam. It is necessary to write. Nevertheless, we have a bit of plastic surgery (mends) for those threatened by time:

• Saving Private Organization: Or public, it does not mind. Locate some company (it can be the company where you work, if you work, or look for an internship), and try to identify some deficient process which you can improve by applying the knowledge gained through your studies. This approach has been the salvation of thousands in the computer science/systems area: you look for a company where something is done following an outdated way (preferably some very specific activity of that type of company) and as thesis, you build a system that automates such process in distress. Done. It is very likely that many systems solving that problem already exist, but nevertheless universities keep accepting this kind of work. Why? Attention, please. First, it is a way for bringing closer the university and the industry. Second, although, as I said, it is likely that some system solves the problem, the trick is to highlight some peculiarities of the company as an obstacle to using the existing systems, and at the same time, as a justification to build a new one. Third, it is a great opportunity for companies to obtain a system that, in other way, could cost them a lot of money. And finally, you have a pretty good chance of getting into the industry, after graduated. Win-win.
• Salvation’s Lab: Universities offer programs allowing their students to go to laboratories in corporations or other universities. Normally, such are environments with a lot of things-to-do, which could become great opportunities for thesis. At heart, it constitutes a variant of the previous technique.
• Standing on the shoulders of giants: Alternatively, you could start by reading some specialized journals in your field. Restrict your attention to updated journals (issues not being older than 1 year). Not necessarily you have to access to printed versions. At this point, Internet may prove a valuable tool. For example, if you are interested in the Computational Intelligence field, take a look at the most recent issues of its specialized journals and conference proceedings. Most do not allow free access to all their articles, but it does not matter. Hardly you would read them all, anyway. Pay attention to the contents and to the titles, to the abstracts and keywords of the newer articles. That will be enough. Analyze and detect the current trends in your field. What are the topics being actively investigated? Any research got your attention? Perfect. Write down a short note about the subject, and the link to the page. Did you find a very clear or complete article? Print or photocopy it. And continue your search. You will decide when to stop, but I suggest you should have at least 5 entries (tentative subjects for thesis) in your list. And please do not stop although you run into a subject that suddenly you are delirious about, because it is always a good idea to have “backup” subjects.
• In Direct: Attend to as many conferences as you can, and apply a scheme similar to the one of the specialized journals. The advantage of the conferences is that a greater possibility exists of detecting which are the current trends and hypes, what topics are lacking (this may mean research), and the possibility of contacting other (experienced) researchers in your field. Sometimes, your contacts establish an important difference, and can help you selecting a research subject.
• The more, the better: Modern departments and universities do not hide their predilection by interdisciplinary studies. If you have studied artificial or computational intelligence you surely met crowd of examples about this. Nevertheless, you have to be cautious with the amount of knowledge that you will have to handle during your research, because it stems from different fields. You must have sufficient skills in the involved subjects, or to be sure about the underlying knowledge not being out of reach for you. However, an advantage of this approach is that you can relate your work to a subject that indeed fascinates you (for example, you may link music with computing). The harder part is to strongly tie the topics with a good hypothesis. If you succeed, researching will be a joy, as your motivation will hardly tremble.
• The dog ate my homework: Finally, you can always follow the popular route: choose an advisor who can provide you with a thesis topic. It should not be that difficult to you to locate some teacher with several research topics. I have to say, I don’t like this approach, but a lot of people indeed prefer this style. Naturally, ask beforehand about the time that other students have required in order to complete their theses with that advisor, and if theses or projects supervised by that professor have suffered considerable revisions after defense. Don’t worry that much about the advisor’s resume. Instead, try to know about the style that the professor uses for advising. We do not want the advisor to become a bother rather than an aid. If possible, take a course with the potential advisor. Attention to what I said before:
  • how quickly will the professor review your drafts?
  • how experienced is the professor dealing with academic bureaucracy?

You shouldn’t choose advisors which are willing to let you to research as you wish. In the long run, they won’t guide you, and will forget about you when a new student arrives with a neater topic.

If now the problem is that you have a lot of topics, simply select the one in which you are more skilled and feel relaxed. But don’t forget about other potential topics. You may need to change your topic later, for any reason. Remember, choose the the topic you like the most. That’s the most important factor when you’re thinking how to choose a topic for a thesis or dissertation. Your thesis is a step toward achieving personal goals. And remember that life is the power to decide.