How to write a dissertation
A dissertation represents the crowning achievement of your academic journey. It's a unique opportunity for original research and contributing new scientific knowledge to the world.
Choosing a topic in the domain of operational research is choosing not to take the path of least resistance. Rather, you will be expected to challenge yourself and go beyond what most people achieve in their dissertations. The upside being that this is likely to teach you skills on which you are likely to build for the rest of your professional or academic career. This does not mean that top grades are an absolute requirement to deliver an excellent dissertation, but motivation and perseverance are essential.
Your responsibilities
- Work independently: You are in control of your dissertation. This is not a task to complete where someone else is listing the sub-tasks. A dissertation is a chance to prove that you are capable of doing original research. Your only real responsibilities are:
- A progress report after the first year
- Delivering your dissertation on time
- Being present for your defense
- Ask for guidance: Your supervisor will be available for feedback. To make optimal use of this help, make sure that you give input. When setting up meetings, prepare questions and any preliminary results in advance. Send these materials at least three days before the meeting.
- Willingness to learn: You will be required to learn new techniques, as well as programming skills. Expect to spend considerable time doing this.
- Do not procrastinate: Start immediately. Quality research takes time to mature and can't be rushed.
Properties of good research
- Novelty: Your dissertation answers a previously unanswered question, and does more than simply apply a technique or formula. Examples include a comparison of the performance different techniques, using an existing technique on a novel problem,...
- Research Questions: Formulate clear, measurable research questions. Explain why these questions are significant and how the answers can be applied.
Style and Language
- A clear structure: An academic article and a dissertation have a predefined structure to make them easy to read. Typically this means that you at least include the following:
- Introduction: Provides background, states the problem, and outlines the scope and objectives of the research.
- Literature Review: Reviews existing research related to your topic, identifies gaps, and sets the stage for your study.
- Methodology: Explains how you conducted your research, including the methods and materials used.
- Results: Presents the findings of your research in a clear and systematic manner.
- Discussion: Interprets the results, discusses implications, and places them in the context of the literature reviewed.
- Conclusion: Summarizes the study, reiterates its contributions, and suggests avenues for future research.
- Clear and Academic Language: Keep your sentences short, to the point, but maintain an academic tone.
- Chapter Previews: Mention what each chapter will discuss. Do this in the introduction of each chapter.
- Use visuals: Figures clarify the text, so in this case, "more is more". Figures can be taken from existing literature, but if so, the source must be cited in the caption below the figure.
- Co-authoring Guidelines: If you're writing the master's thesis with a partner, you're welcome to divide the different sections between yourselves. However, consistency is key. Agree in advance on elements like the abbreviations you'll use, the way you'll write specific terms, and so forth.
- Summary Tables for Reader Convenience: It's useful for the reader if you create summary tables of the above text. This can be organized either by chapter or by section. By doing so, you offer a clear overview of the material discussed. This approach also makes it easier for the reader to pick up where they left off if they've temporarily lost track.
- Formatting: Your dissertation should be formatted using LaTeX and adhere to academic standards.
Literature review
A literature review is required. Your literature review should show related research, and indicate what is novel about your research.
- Consistency Between Literature Review and Research: There should be a degree of consistency between your literature review and the research. These two components shouldn't stand as separate sections; rather, they should complement each other.
- Be Concise: Given the extensive literature on operational research focus only on what is relevant for your research. There is no need to copy large parts of a handbook, you can assume that a person who reads your dissertation has some specialized knowledge.
- Critical Comparison: It's essential not just to report the literature but to compare, weigh pros and cons, etc. For example, if you discuss two techniques that address the same issue, compare their advantages and disadvantages.
- Citation Importance: Proper citation is crucial. When in doubt, it's better to over-cite than under-cite.
Assorted tips
- Start small: Often you will have big ideas, which automatically translate to very complex programming challenges. It is essential to try and make your problem definition as small as possible, and to add elements incrementally.
- Google scholar: Use google scholar's "Cited by" functionality to find the most recent research on a topic.
- Zotero: Use Zotero to manage your literature research and your citations, as well as notes. As you read, jot down what you want to retain for the literature review or further research. This will aid in subsequent processing. You'll never use an entire paper, so marking it while reading will save you time in locating important information later.
- LaTeX: Start writing in LaTeX immediately, do not try to convert things from word back to LaTeX, this rarely results in an attractive looking document.
Collecting Company Data
Data collection is not something to leave until the last minute. Plan ahead and be proactive.
- Talk to Family and Friends: Use your personal network to make collecting data easier. Family, friends, neighbors, and acquaintances can sometimes provide connections in the companies you're interested in.
- Direct Contact: Find someone within the company who can directly assist with your research. Search for their contact details online. Avoid sending emails to generic addresses, as they may not reach the intended recipient.
- Clear Objective: State up-front what issue you're addressing with your research. Try to convince the company that your study could benefit them as well.
- Time Estimation: Provide a specific time frame for any requested meetings, e.g., "The meeting will last approximately one hour." Also indicate your general availability but remain flexible if the company suggests another time.
- Cite Your Advisor: Mention your academic supervisor's name and indicate that the company can contact them for further clarification. This can instill trust.
- Career Fair Contacts: Companies at job fairs are usually looking for talent. Mention that you came to know about the company at a specific job fair. They might be more willing to help.
- Send a Reminder: If a company hasn't responded within a week, it's advisable to send a reminder email. While this may seem pushy, it often prompts a response. However, limit reminders to just one to avoid appearing overly aggressive.
Feedback template
Your input for a feedback meeting is ideally structured as follows:
Dissertation Information
Topic:
[Provide a brief description of your dissertation topic]
Research Objectives:
[List the main objectives you aim to achieve through this dissertation]
Current Progress:
[Give an overview of which chapters or sections you've completed or are working on]
Specific Questions or Concerns
- [Question 1: Could be about methodology, structure, or particular challenges you're facing]
- [Question 2: Could be about relevance of literature, clarity of argument, etc.]
- [Question 3: Any other questions you may have]
Reading materials
Describe the materials that you send over, this can be datasets, code, and drafts of your dissertation. When sending over preliminary versions of your dissertation please highlight novel work or areas where you would most like to get feedback on.