It is a common misconception that the empirical chapters are the place for your analysis. Often this confuses the reader. In fact, you need to split the empirical facts and discussion of those facts into two distinct sections. 

In this post I want to explain why, more often that not, you need have separate empirical and discussion chapters and why you shouldn’t combine them. Then, I’ll talk you through practical steps to employ when you’re writing each up. 

Whilst they are closely related, they occupy two very different spaces within a PhD thesis.

We can understand a PhD as having four distinct sections:

1. Introduction – this is where you introduce and outline the entire study.

2. Background – this where you lay the groundwork for your thesis (in your literature review, theory framework and methods).

3. Core – this is where you present your findings.

4. Synthesis  – this is where you relate the core to the background.

The empirical chapter(s) is/are where you present the facts of your study. They occupy the core of the thesis. 

The discussion chapter though is where you interpret and discuss your findings in relation to the thesis and wider discipline. That is why is occupies the synthesis stage of the research.

Your job when writing your discussion then is to interrogate and critically engage with the findings and relate them to the research aims, objectives, research questions and gap. Most of your cutting-edge analysis and engagement with your findings will take place here.

The job of a discussion chapter is therefore to critically examine your findings with reference to the discussion in the background chapters of the thesis (introduction, literature review, theoretical framework and methods) and to make judgments as to what has been learnt in your work. In essence, the job of a discussion chapter is to tell the readers what your findings (may) mean.

The reason for this distinction between empirics and discussion is to make life easier for your examiner. They are looking at whether you are capable of both presenting observations in line with your methodology, and interpreting their significance in the context of the thesis as a whole.

However, as with everything in the PhD, there will be exceptions to this. Some theses, particularly those from within the liberal arts and social sciences, may not need a separate discussion section because the nature of their study may mean that empirics and discussion are intertwined. It is also worth pointing out that although not every thesis will have a discussion chapter, all theses will contain discussions of some sort, however short.

It’s important to note that the form your empirical chapter will take – or indeed if you have one at all – will depend heavily on the nature of your thesis and the discipline in which you’re working. Someone working in a more theoretical space – for example, philosophers or the theoretical mathematicians – might rely less on first-hand empirical data and more on proofs. Those working on quantitative studies will have more data-driven, empiric-heavy empirical chapters.

Broadly speaking then, the emphasis in the empirical section or chapter is on factual recount and summary. You’ll be categorising your findings into particular themes and using a variety of visual elements (tables, figures, charts, and so on) to present your results. You need to show the reader what your data ‘looks like’.

You need to do it well, too. If you present data in a messy way, your examiner might think that your thinking is messy.

By the time you have finished your empirical chapter, your reader should be able to answer six questions:

1. What are the results of your investigations?

2. How do the findings relate to previous studies?

3. Was there anything surprising or that didn’t work out as planned?

4. Are there any themes or categories that emerge from the data?

5. Have you explained to the reader why you have reached particular conclusions?

6. Have you explained the results?

You are providing sufficient detail that others can draw their own inferences and construct their own explanations. Think of it as presenting the case for a jury.

That means that an empirical discussion should:

1. Tell the reader how the data was collected, with reference to the methods chapter/section

2. Tell them how they can access it if they wanted to replicate your study

3. Discuss what the results look like (using visual aids, such as tables, diagrams, graphs and so on)

4. Provide rich summaries of the findings

5. Discuss the gaps in the findings and analysis

6. Analyse the results

7. Discuss the implications of your findings

8. Discuss the limitations of the findings

Many students struggle to write up their discussion chapters. The reason is that they lack the confidence needed to make the kinds of knowledge claims required. The discussion chapter is where you start to develop your scholarly authority, and where you start to make truth claims about your interpretation of what’s going on. By implication, that means it is where you start to agree or disagree with existing literature and theoretical ideas. 

Another reason why students struggle is that they fail to realise the significance of their findings or, put differently, they don’t think their findings are significant enough in their own right. By the time we come to write our discussion, we are so conditioned by the findings that we may not realise that they are significant and do, in fact, make a contribution. There’s often an expectation gap here; students expect their contribution to be big, whereas, more often than not, the contribution a PhD actually makes is small and specific. 

As with every stage of your thesis, you must relate your discussion section/chapter to the background. Specifically, you need to relate it to the empirical chapter, aims and objectives, research questions, the gap in the literature and, if relevant, your theoretical framework. There will, therefore, be a lot of signposting to other parts of the thesis; doing so is necessary if you want to show the examiner that you can relate your findings to the broader context of your thesis and discipline.

One of the biggest obstacles is synthesising your empirical data and being able to critically discuss it in relation to this broader context. The authors of How To Write A Better Thesis offer up an effective technique you can use if you’re struggling to do this. They refer to it as a ‘mud-map’:

1. You can start by writing a long list of everything you have found.

2. See if you can sort and organise this list. Categorise each finding based on whether it is speculative or based in empirical fact. This is important because your discussion will need to be somewhat (but not too) speculative.

3. Try to categorise your different findings into themes.

4. Now try to find linkages between these themes.

5. Organise these themes into different section headings for the discussion chapter, and try to come up with sub-headings. 

When it comes to writing your discussion chapter, you can start by writing a few sentences that summarise the most important results. 

One danger when writing discussion sections is that they can be too wordy, offer too much interpretation and lack a clear structure. To avoid this, you should make sure that every element of your discussion section addresses one of the following questions: 

• What are the relationships between observations? (The mud-map you developed earlier will help here.)
• Are there any trends and generalisations amongst the results? Are there any exceptions to these? 
• What are the causes of, or mechanisms behind, the underlying patterns you have uncovered?
• Do your results agree or disagree with previous work? 
• How do your findings relate to the theoretical framework you developed, if applicable? 
• How do the findings relate to the hypotheses you developed, if applicable? 
• What other explanations could there be for your results? This issue is more pertinent if you are engaging in theory creation/inductive reasoning.
• What do we now know as a result of your research that we didn’t know before? 
• What is the significance of these findings? 
• Why should we care about the findings?

When your discussion chapter is finished, your reader should be able to answer the following questions:

1. How do the findings relate to the theory and methods discussed previously?   

2. Why you have reached particular conclusions?   

3. How do your findings relate to the gaps in the literature you identified earlier?   

4. What implications do the findings have for the discipline and for existing understanding?

5. How do the findings relate to your research questions/aims and objectives?   

A particularly important theme that I want you to always bear in mind is that your interpretation and discussion of the findings needs to be done in such a way that it relates back to the aims, objectives, research questions, gap and any theory. Running through your thesis will be a central argument – your thesis statement – and it is in the empirical and, particularly, the discussion chapter, that you will present all of the evidence and logical argument necessary to support that argument.

Unsurprisingly then, these core sections of the thesis are the most important, as they are where you make your contribution. Of course, you have outlined what your contribution is in the introduction, so what you are arguing is no surprise. But, it is in the core of the thesis where you drill down into the detail and critically engage with that contribution, using your data to rigorously support your line of argument.