Perhaps the most important aspect of writing a thesis is its structure. Structure is everywhere in the thesis and it is sometimes difficult to identify or define. In essence, it is the organizing principle of the text. You can compare it with the way you organize your books. If you are a very badly organized person, your books will most probably be lying on the floor, stacked randomly in irregular piles. In your quest to become more organized, you might want to buy a set of bookshelves. This will tidy up your room, and you can see all books and their titles from one viewing point.
Yet, with many books, buying bookshelves may not be enough. You will probably add more structure to your library, for instance by dedicating sections of your bookshelves to your science fiction books, study books, reference works, etc.. When your library is large enough, you may even want to order them in alphabetical order, add an index system with keywords linked to specific titles, etcc. These are more hidden way to structure your text. The more aesthetically oriented book owner may even sort his or her books based color, but this is not very functional if you actually want to find a book.
In a way, structuring a thesis works similar to organizating your book collection. The structure in the book collection makes it easier to find books, while the structure in the thesis makes the text easier to read. The reason that the text becomes easier to read is because your text becomes predictable: You don’t start talking about the details of data collection in the introduction, or introduce a complete new explanation in the conclusion. The reader needs to be served as if he or she is in a restaurant: start with an aperitif, then a starter, next the main course, don’t forget about the dessert and of course the coffee at last.
The first obvious way to structure a thesis is with sections and section titles. If you write a thesis with less than twenty pages, it is usually sufficient to have one level of section numbering. The typical titless will depend on the nature of your research field, the methodology iused, etc.. For a typical Finance thesis, reporting on an empirical research project, the following sections are likely to be sufficient for a thesis up to 25-30 pages:
- Introduction
- Literature review
- Data
- Methodology
- Results and discussion
- Conclusion
While this structure is somewhat boring, the reader is looking for novelty and surprises in your ideas and results. Too much structure is also a killer. For instance, it may be acceptable to have two levels in the section number (2.1 Early Literature, 2.2 Literature), but three levels in a document of 25-30 pages is really too much. What is the argument against so many subsections? Let me start about the positive side of using subsections: it makes it much easier to find particular elements back. However, the downside is that each time you add in a section, subsection, or sub-sub section, you disrupt the flow of the reader. Compare it to a restaurant: rather than serving the main course in one go, it is like serving the meat, potatoes, vegetables, salt., pepeer, and gravy seperately. By serving all these items seperately, the waiter not only interrupts you on a minute-by-minute basis, the composition of the meal also gets lost.
The introduction
There are many ways to write an introduction, in particular for an experienced writer. My recommendation here is focused on the beginner, the research or student who writes his or her first or second thesis. A good begin is to plan the subsections that you want in your introduction:
- Start with introducing and describing the topic. This can be done in many ways, for instance by defining the topic, describing a relevant case, a news article, presenting statistics on the topic’s relevance, etc. You can reserve one, two or three paragrahps for this.
- Next you start describing the problem that you want to address in relation to your topic. It is usefull to include references here. The problem statement is the core of your introduction. It can take several paragraphs.
- In the end, your problem statement should be followed by a research question or a research objective. While the problem statement is a discussion of a broader problem, the research question is specific and presents your contribution to the solution of the problem. One paragraph may be enough.
- A preview of your research. Briefly mention the data and method that you are going to use.
- Relevance of your work. A discussion of the relevance of your work and its contribution to the academic research and/or the practical relevance. One or two paragraphs.
For instance, your topic could be about mergers and acquisitions. So your first two paragraphs could define the concept of a merger and/or acquisition, what the size of the market is, etc.. Then you could continue with a discussion of the relevance of a merger for shareholder value: it might be destructive becaus it create unmanageable conglomorates or it might create shareholder value. Don’t forget to add in references here. Perhaps one of the reseasons that people disagree about this is because some mergers are done with the idea of cost-cutting, whilst others are directed towards an expansion of markets. You research question may then become: Do cost-cutting mergers increase shareholder value relative to mergers focused on expansion?
The example here is nothing more than a sketch, so in the end it needs to become convincing because of your (strong) arguments. The introduction is really important, it needs to be carefully written because it contains the motivation of your study.
Auke Plantinga 