Structure 

For a general overview of how scientific writing could be structured, see the visualization on the right (or the large version here). We also recommend reading “So what is a (diploma) thesis?” by A. Bernstein (from IFI) and “How to write a technical paper or a research paper” by M. Ernst.

Note: The following structure applies to most theses. Especially, if you conceptualize and implement an approach throughout your thesis work, but it is mainly meant as a guideline (if you want to deviate a lot, talk to us). We also recommend to consult previously completed theses (e.g. TaskSnap, AmbientTeams, or Boosting Collaboration).

1. Abstract  

• Summarizes (and sells) the key motivation, gap in research, the solution to the gap (what did you do to address the gap), evaluation, main result(s), and possibly main discussion point within 200-300 words.  

2. Introduction 

• “Establishing your territory”: explain your problem and show why it is relevant.
• Subsequently, discuss attempts of other researchers in that area (summarize related work, and state what they have done and what not [see next point]). 
• “Establishing a niche”: Explain to the reader what is still missing: What do we still need to understand? What do we still need to solve the problem more effectively? What is the niche that research so far failed to address? 
• “Occupying the niche”
  • Introduce your approach in a conceptual manner (i.e., without technical details).  State how you are addressing the niche.
    
  • Introduce your research questions and enumerate them in the introduction. Don’t just provide a list but embed them within the argumentation. Why are these questions relevant, novel, and interesting?
  • Introduce your methodology to answer these research questions; that’s usually a summary of the study design. 
  • Introduce main findings in 1-3 sentences. 
  • Announce what contributions your work provides (e.g., a prototype, a prototype, an interesting methodology, a study contributing to the body of knowledge, other artifacts, etc.). 

3. Related Work (Literature Review) 

• Think of the main themes and larger narrative! Don’t enumerate one related work after the other (see below), but rather think of discussing a whole group of related work (by theme): what do they have in common and how do they differ? For this, it helps to take notes on each paper and think about the space of related work and which “dimensions”/themes divide that space and then explain that. The idea is to give an overview of the space and use some examples, rather than just providing a list of examples. 
• You can use the related work section to strengthen and motivate your research (reinforce the introduction). 
• For related approaches, always state how your work differentiates, especially for the ones closest related to your work. 
• Level of detail: Most of the time, you should focus on high-level differences in your approach. It makes sense to go into a bit more detail (but not too much) for work that is very related to yours. 
• Also see our Guidance for performing a literature review.

4. Approach 

• Not everyone might have this section since that depends on the type of thesis; similarly for prototype section.
  
• Discuss your 3-5 key concepts (higher-level ideas that characterize your work – basically important design decisions that you made and that differentiate your work from others [these are the items that others might port/replicate in other contexts, on other platforms or the like]). 
• Don’t talk about the implementation details of your prototype.  

5. Prototype Implementation 

• This can also be integrated with the approach section, e.g. for each higher level design concept/decision you can go into more details. You can also mainly stay on the conceptual level and present parts of the specific implementation. Always think about what is more a conceptual idea that’s independent of the specific implementation and what’s just an implementation artifact.
• First 1-2 sentences: Introduce your prototype (“we built xyz, a Chrome Extension, to …”) 
• Structure your write-up in terms of high-level components. These don’t necessarily have to correspond with your software/implementation components (you might, as stated above, also integrate it with the approach).
  
• Don’t write a user manual or technical documentation (“there is a button on the bottom left. If you click it, it renders the workday-visualization screen”). Instead, use a user-centric perspective that outlines their benefits (“users can inspect their work patterns using the workday-visualization by using the UI controls.”). In most cases, adding the software architecture or database schemas is unnecessary, but a data or user flow diagram might be valuable in some cases. 
• Motivate, motivate, motivate: Why did you decide to implement a certain feature, and why did you omit another? Ideally, motivate it by citing relevant related work. 
• Your tool/approach implements the concepts outlined under “Approach” – always relate implementation details with these concepts. 

6. Method 

• Procedure: what was the procedure of the study? Usually starts with “we asked participants to read and sign the consent form, install our approach on their machine and we provided a tutorial. It may end with “after the final semi-structured interview, we collected the anonymized user data…”. Write down everything else that happened in between. (this is very similar to what you have in your consent form)
• If you have a survey or interview, state what it was about and provide some sample questions as well.
• Participants: Mention the number of participants, their age range, and relevant backgrounds/experiences (e.g., job roles). You can also explain how you recruited them. 
• Data Collection and Analysis
  • How and what quantitative and qualitative data was collected? How was it analyzed (e.g., for qualitative data this could be “Thematic Analysis”)?
  • Always motivate your method. Why did you do an interview? Why did you do a field experiment?
  • Here you can also state if you had to exclude certain data from the analysis due to some reasons (make sure to justify it).
    
• Also see our Guidance for Qualitative/Thematic Analysis.

7. Results 

• Think about what your main insights are, not all, just the ones that are relevant to your thesis topic / main point (also see “Know your message and stay on message”).
• Answer your research questions. Sometimes, it makes sense to write a section for each RQ, whereas most of the time, the results can be written up around the main themes you’ve identified through a thematic analysis. 
• Identify and think of 3-4 high-level findings (=sub-sections). For each, write a paragraph or subsection (for sub-themes). Use a descriptive heading summarizing your finding. In the paragraph, provide compelling quantitative and qualitative evidence supporting your finding.
• For important results/observations, also use figures/tables that highlight your findings and provide further evidence. Think about how to best communicate your main findings, some figures / tables can really help, but ensure to only include the ones that help with your main points and cut the ones that are not relevant since they will distract the reader.
• Don’t just enumerate (see below). 
• Make it a story and think about what a good story line of your results is, also with respect to the niche that you are addressing and the research questions. 

8. Discussion 

• Take a step back and think about the gap that you’ve identified and attempted to fill with your research. How well could you answer the research questions? How do the results generalize and/or apply in practice? How do the results relate to existing related work (e.g. do they support/replicate findings from earlier work, do they complement or extend them)? What could be concrete next steps for future work? Do you have any “expert” advice or insights that may explain your findings? 

9. Threats to Validity 

• Consider your work’s internal, external, and construct validity limitations. Discuss these threats. Also, discuss how you tried to mitigate these threats.

10. Conclusion 

11. References

12. Appendix

Writing Style and Tips 

• Think about the structure and what you want to write first. It’s a good idea to first write detailed bullet points before crafting full sentences. 
• Use topic sentences. The first sentence of each paragraph should summarize the whole paragraph. (sometimes you will write these sentences explicitly in the document, sometimes you might just have them for you as a comment in your latex file; but these will help you to focus.) 
• Hierarchical Structuring: The first sentence of a paragraph should summarize the whole paragraph or section (topic sentence). And the first paragraph/sentence of each section/chapter should summarize its substructure. 
• One key message per paragraph. 
• Use of Abbreviations: Write out the full name/definition the first time you use it and put the abbreviation in brackets behind it. 
• Don’t enumerate: Watch out for certain adverbs (“first”, “second”, “another”, “finally”). These hint at sub-optimal writing patterns where you list one statement after another. Typically, these are found in related work and the results section. Try to rewrite by grouping statements and discussing these groups in terms of similarities or differences. 
• Consult other research papers and see how they wrote up their work. Use them as orientation. When in doubt, discuss it with your supervisor. 
• Use the structure “Tell them what you are going to tell them, tell them, and tell them what you told them!” (from Bernstein 2005) 

Getting Feedback 

• Generally, your supervisor provides one (full) round of feedback on one of your drafts. It might be be possible to get feedback section by section, but feedback is usually more coherent with a (mostly) complete draft.
• Make sure to send your draft early, ideally around 2 weeks before your deadline. Coordinate with your supervisor. Don’t expect feedback when you send your draft only a few days before the deadline. 

Writing Template 

Bachelor or Master Theses

• Use the HASEL Template
• There is no page limitation; many theses are around 25-50 pages counting from intro to conclusion (including the empty pages after each chapter and any figures, excluding Appendices). 

Master Project

• Use the CHI ACM Template (single-column)
• Your project report should be at most 9 pages (excluding figures and references, including tables).

Presentation 

• After submitting your thesis or project report, inquire with your supervisor about setting a date for the presentation (aka defense).
• The presentation consists of a 15 minute talk, followed by a Q&A session by members of HASEL and others from the audience.

Grading 

The thesis will be graded using the following scale:

• 6.0 Thesis quality significantly exceeds expectations: The quality of the thesis is excellent and close to or good enough for submission to a scientific venue. The approach and the evaluation are well-motivated, sound, and thorough and present a significant contribution. The write-up is clear, concise, and correct.
• 5.5 Thesis quality surpasses expectations: The thesis work represents a well-designed, -thought-out, and -evaluated piece of research that makes a profound scientific contribution. The write-up is clear, well-presented, and does not neglect any key points (literature review, sufficient evaluation). To be submitted as a publication to a major scientific venue, the work would still require significant modifications, theoretical justification, or additional validation.
• 5.0 Thesis quality meets expectations: The student implemented an approach that fulfills the given requirements. An evaluation was performed. The thesis represents a significant contribution or study or a complete analysis that was carried out with great care, but one of these aspects was neglected. The thesis represents a good effort, but there are obvious areas where improvements could be made.
• 4.5 Thesis quality partially meets expectations, minor deficits: The research work does not produce convincing results. The evaluation is weak, although a sufficient analysis of the shortcomings is present. The quality of the writing is sufficient.
• 4.0 Thesis quality meets minimal quality requirements; it has major deficits and it is clearly below expectations: The thesis lacks independent contributions and only combines existing approaches in a trivial way. The effort was barely sufficient and the student failed to provide any interesting insight on the topic.
• <4.0 Thesis quality does not meet minimal quality requirements, has major deficits, and is below expectations.

[1]: Burrows, T. (2011). Writing research articles for publication. Unpublished manuscript, the Asian Institute of Technology Language Center, Khlong Luang, Thailand.

Note: This guide aims to guide students in writing good theses/project reports with/for HASEL. It is subject to change. Not all the suggestions might always apply. Ask your supervisor in case of questions.