Methodology tutorial - structure of a master thesis

This is part of the methodology tutorial.

Introduction

Learning goals

• Understand that a master thesis is an argument
• Learn how to sequence a thesis
• Understand that you may have to respect certain standards

Prerequisites

• Methodology tutorial - the research plan
• Most other methodology tutorials in one way or another ...

Moving on

• Do other research :)

Level and target population

• Beginners - master thesis

Quality

• Should be ok, although important elements are missing regarding the principal chapters

A thesis is an “argument”

In other words: The organization of the written theses has nothing to do with the organization of the research plan or its little section on planning. In particular:

• A research plan (i.e. the research design) defines and organizes your work according to logical criteria.
• The research planning (i.e. the little section at the end of your research plan) organizes your time according to workpackages and deliverables.
• Research is just done and not told, i.e. you do not tell people your personal experience with this. The thesis is not a story. It presents the results of your research (including a literature review and and methodological explanation on how you did it). There may exist exceptions, e.g. in some forms of ethnography.

The structure of your thesis if defined by two main elements:

1. The research type/approach and related methodological criteria.
2. Some rhetorical principles, i.e. your thesis should be readable.

A reader must understand your objectives, the theorectical background, the questions, how you did it and your answers (results).

Presentation and typographic structure

Let's first a look at some superficial presentation issues.

The word processor

Start by admitting that you don't know how to use a word processor. Don't feel ashamed. Most people don't. E.g. I had to write a wiki entry about Microsoft Word before writing a larger text. I usually use FrameMaker which is very different because it was designed for people that write real text and I had to make a real effort to adapt to MS Word. - Daniel K. Schneider

Here is a list of "must know" things:

• Define styles (and make sure to configure MS Word that will inhibit modifications on fly them or addition of new styles)
• Automatically create tables of contents and figures
• Create indexes
• Automatically number titles
• Create stable numbered lists

Don't loose days with repetitive re-fomatting.

There exist two "formatting" strategies

• Either learn how to create a good list of styles (you may need between 15 and 30 for a master thesis depending on your research type
• Ignore my advice, but then only spend your last day with manual formatting.

Professionals do it this way

• Each type of paragraph has its own style
• Never use TABS or empty lines (e.g. paragraphs are separated by space, not an empty line, so add horizontal space to the paragraph style element definition).

Your list of styles

You need at least the following elements:

• Numbered Chapter, numbered section, numbered sub-section and unnumbered sub-section. If you use MS Word, just define styles for heading 1 to heading 4.
• List elements (bullet list items and enumerated items). You may, but usually don't have to define these at two levels
• Normal paragraphs
• Citation paragraphs (indented)
• A style for fixed formats if you plan to present code
• One ore more good table styles
• Figure captions

Tip: Read Microsoft Word for some advice.

Titles et sections

Here is some advice about titles and sections

The table of contents not only is a navigation tool but it indirectly defines your argumentation flow. This is why wording of titles and structure is important.

Do not use too many section levels (like 12.3.4.1.a). Your thesis is not a military or administrative operations manual, but a flow of connected ideas.

• Too many levels will make orientation difficult for the reader. He won't understand where he is.
• You may add unnumbered titles at the section or sub-section level or maybe use something like (a) .... (1) ....
• Each numbered sub-section represents an important topic
• Titles should summarize a topic (without being too long)

You have to find a compromise between:

• flow of argumentation (avoid sub-titles because the "cut" intro a text)
• structure (use sub-titles to separate topics)
• readability (use un-numbered sub-titles to structure contents that stretch over several pages)

Layout

There exist several schools of thought. Make sure to consult official guidelines too !

Page numbering

Either just number from 0 to n or use the more sophisticate following scheme:

• Roman numbers for preface, table of contents etc.
• Normal (arabic) numbers for the main part
• Something like A-1, etc. for annexes.

Headers and footers

• On top of left pages you should put the current chapter title
• On top of right pages you should put the current section title
• On bottom right (and if you use left/right pages, also on bottom left) include the page number. Make sure to do this for early drafts too. I hate students that present drafts without table of contents and numbered pages and your advisor will hate you too - Daniel K. Schneider.

Notice: In MS Word 2003 adding chapter headings is painstaking labour since you have to do this for each section, so you may skip this. But any real word processor can do this really easily.

Line length

• Don't write long lines ! Readers will get lost
• Simply use decent margins and don't use small fonts (10pt is too small)
• You may indent titles to the left (certainly not to the right as some Word default styles will do)

Some "modern additions"

• Use boxes to present "special information" like case summaries or important conclusions.
• Use side headers (also very difficult to do in Word 2003).

Figures an tables

• Label and number each of these. If you work with a real word processor, let them float to the top or the bottom of the page. If you use Word, you can do this manually (but only the day before you turn it in).

Fonts

• Use a font with serifs (e.g. Palatino or Times Roman)
• Non serif fonts may look prettier to you, but they diminuish readability since serif fonts add some "invisible line" that the reader's brain will use to keep track.

The organization of a thesis

Here are the most important parts of an academic piece:

Elements	Importance	Main functions
Foreword	*	Personal Context
Table of contents	**	Navigation
Abstract	*	Main objective, result and scope
Introduction	***	Objectives, global approach
Principal part	**	(depends on your research type)
Conclusion	***	Summary of results, further work and scope
List of sources	*	Data anchoring
Indexes	*	Navigation
Bibliography	**	Theoretical anchoring
Annexes	*	Presentation of detailed data, materials, etc.

Foreword

The foreword is not part of your thesis.

You may use it to:

• thank people and tell them how much you like your cat
• explain why you have chosen this subject
• (maybe) excuse yourself for things to didn't do
• announce some followup

Things that relate to your work belong to the introduction

Tip: Give some thanks to your advisor. He/she probably deserves it and even if he/she doesn't, it's good policy.

Table of contents of tables and figures

Mandatory

• Position: At start and after the foreword
• Must match titles in the text (this should be obvious). Even Word can generate this easily.
• You also should add tables for the figures and the tables. This will allow people finding synthetic information.

Do not forget that you don't just write a thesis for a jury. Other people may read it and they want to do this fast and maybe just find some bit of information as quickly as possible.

Abstract

• This is often mandatory
• If it is not, you also may summarize your thesis as a paragraph in the introduction

Introduction

The introduction (as well as the conclusion) is the most important chapter of your thesis. Some people will decide to read or not to read your thesis after looking at the first page.

• Even serious readers like the jury, will read the introduction first and they should clearly understand what you did. Also make sure that they find your subject interesting. You have to frame readers.

A reader must understand:

Elements	Details
The big question	.... summarizes your subject, i.e. what you wanted to find out.
	.... implicitly or explicitly defines a scope
The "language"	.... which major concepts you use, word definitions you use, etc.
The general approach	.... research type, global approach, principal methods used
	.... the structure of your thesis

In general, the introduction includes:

A description of your research subject (including the big question).

A short discussion of the interest of your work and its scope (including what you will not do).

A synthetic list of research questions and/or working hypothesis (if your research is rather theory-finding). Alternatively, they may appear after the literature review part.

A list of some important definitions, e.g. an explanation of the words you use in the title of thesis or the big question. You also can do this in the literature review.

A presentation/discussion of the global approach, unless you dedicate a special section to this. In the latter case you should just briefly describe the approach in a single short paragraph.

A short guide for the reader. It will help the reader finding things and also show that you can provide a rationale for the adopted structure.

An introduction of the object(s) you study. E.g. if you do some policy implementation research, you may present the context and the legal basis.

Notice: A working hypothesis is not a scientific hypothesis that can be tested. It's just a more aggressively formulated general research question. "Real" hypothesis exist in theory-testing approaches. They are grounded in theory and can be properly tested with data. Such hypothesis are always presented after the theory part and then even further operationalized after or in the methods chapter.

Principal chapters

It is difficult to give some useful advice about the principal chapters, since there exists a large variety.

Certain research types / approaches have strong guidelines for content structuring. However, we can try to formulate some general principles.

In all empirical studies, you should:

• present and discuss existing work
• present data and results
• discuss results
• link results with the questions and hypothesis
• confront results with existing knowledge

However, the order is not necessarily the same for a given research type...

Writing Style

Avoid including lots of statistical indices in your sentences, rather use tables for this (except in experimental psychology, where text is supposed to be unreadable)

The literature review chapter

Usually, the review is done in chapter two. However, there may exist exceptions (e.g. in history)

But in any case this chapter should be used. Research questions are presented in detail after the review and they must be grounded in the literature review. Research results have to be confronted to the theory later on (e.g. after or during presentation and discussion of the results). This is an often observed problem in master thesis. This principle also works the other way round. If you don't use theory, do not present it.

Conclusion

Recall the principal results of your research

Discuss the scope of your results and provide an outlook

• You may discuss the external validity (i.e. attempt some generalization)
• Discuss questions for which you don't have answers or things you didn't implement (and why)
• You could formulate a new theory that could be refined/tested in further work
• You can formulate new interesting research questions. Often the value of a master thesis is to generate new research ideas that your adviser could pick up or that you could turn into a Ph.D. thesis :)
• You could globally compare your work to other empirical studies (if not already done so)
• You may discuss the practical usefulness of your work (in particular if your thesis does not have a practical aim)
• If you produced an applied and/or design-oriented thesis, you should recall major recommendations (e.g. present a set of design rules for practitioners)

List of sources

If necessary, this chapter will include a list of all your primary sources, e.g.

• laws and other legal or paralegal documents
• historical sources

You also may include these in the bibliography.

Indexes

You may (but usually must not) produce an index of concepts and authors. It will help the quick reader to find interesting "spots" and also provide an idea about the way you tackled your research.

Most word processors can do this fairly easily (even Word)

Annexes

Annexes are important and include everything that is not strictly necessary for the presentation and the discussion of empirical results.

Typically, the annex(es) may include:

• all research artifacts (such experimental materials, survey questions, etc.)
• some raw data (e.g. stories told be individuals), most often just excerpts
• Intermediate analysis data like descriptive statistics, qualitative summary tables, etc.)

Bascially, the annex allows a critical reader to figure out if you did it right. It also will help other persons to replicate some or your empirical research, e.g. apply your questionnaire to a different population.

Bibliography

It must include each and every reference you directly or indirectly used.

You do not need to include any other references, i.e. texts you didn't use.

You must respect a certain standard (and be coherent)

Tip: Start doing the bibliography right from the start.

• As soon as you use a text, include it !
• You may learn how to use a reference manager, although for a typical master thesis you also can do it "by hand".

Citations

You will have to respect a given norm. In some institutions you can choose, in others you will have to comply. See the citation article for some links that might help you.

The presentation

Do not forget to prepare your presentation. A good presentation can make a little difference. Often the jury is divided and has to decide between a somewhat lower and a somewhat higher grade.

Time is usually very limited, so stick to the essential !

(1) Make sure to present the essential things only

• Your introduction should be as short as possible (no mumbling about how you found your subject and how your thesis changed your life and the one of your dog or cat). Just state the objective (main goal and big question) of your thesis
• Then present the research questions and the results. Make sure to arrange this part according to standards observed for your research type. E.g. in experimental psychology you'd present hypothesis, materials, results, discussion.
• Discuss some outlook
• End with a stunning conclusion, do not finish mumbling...

(2) Plan a test run with some friends

Ask them to tell you what wasn't clear

(3) Prepare the delivery

Once you feel that the contents are ok, you will have to deliver in time and with a minimum of style. The only way to get this right is to repeat the presentation using your voice at least 3 times (looking at the slides and mumbling will not do).

Here is a typical "talk menu". It's in french since French food has a good reputation (I am not going to comment about typical French talk delivery quality here ...).

After you are done

Suggest to your advisor to present your work as a conference paper with him as second co-author. This is fairly standard procedure in the US but not necessarily in other countries. Both will profit if he/she does his/her share ...