Six steps to the perfect discussion section

You’ve passed ethics, recruited participants, collected your data, analysed and written up your findings – phew! What a marathon. The hard work is basically over, right?

Well, sadly not, because there is one more hurdle to jump; the dreaded discussion section.

This is where you compare your work to the existing literature. Sounds simple, but this can be the hardest chapter to write. I have a theory (entirely untested, I should point out) that these chapters are especially hard for women, who have generally been socialised not to brag about their achievements, something you definitely need to do here.

Read on for six tips on how to leap over this final hurdle and write a devastating discussion.

1. Read around your findings

You will already have done plenty of reading for your lit review, but before you start writing, do some more, especially if you are doing qualitative work. In quantitative research, you shouldn’t discuss any papers you haven’t already bought up in your intro, but this is not the case for qualitative explorations, where the researcher will have been expecting the unexpected. Now that you have your unexpected findings, search for similar papers and make notes on the relevant points.

I find it helpful to start a new document listing each of my key findings, and to make notes of any existing findings which confirm, contradict or add to my own, along with a note of which paper the new findings have come from. Include the key findings from the key papers from your intro or lit review on this list as well. You might want to highlight findings which back yours up in one colour and those which don’t in another. This list document comes in really handy once you start to write.

Immerse yourself in the relevant reading

Photo by Siora Photography on Unsplash

Look for work from the same or similar methodologies to you, as well as work from other areas. Find and read papers that are cited in useful studies.

Top tip: if your university doesn’t have access to a paper you need, email the author, or look for them on ResearchGate or Twitter. They will probably be happy to share their work with you. (I am always happy to share my papers, just get in touch via the form below to ask.)

2. Think about format

The format of your discussion section should mirror that of your findings. This helps your reader to logically follow your train of thought; especially vital if your reader is a PhD examiner, for whom you want to make life as easy as possible.

Start with your first finding. Briefly recap it. A common problem I see with PGPR clients is that they spend too much time reminding the reader of findings. The findings chapter is the one before the discussion; try to trust that your work is interesting enough that the reader won’t have forgotten it already.

Think carefully about your format before you start writing
Photo by The Creative Exchange on Unsplash

Here is an example of a recap from a discussion section in my PhD, which was about the experience of living with and being treated for renal failure:

Seven of the participants talked about the impact that ill health made on their lives. Some found that ill health was a wearing intrusion, dragging them down and restricting them. Others had found ways to come to terms with living with their chronic conditions.

Following this, you might want write a brief summary of the existing literature, after which you can start comparing your work to what’s out there already.

3. Look for areas where your research confirms other findings

The next step is to demonstrate how your findings concur with existing work (if they do). This is where the list document you made earlier will come in useful. Look for findings that are similar to yours and tell the reader about those similarities. If you don’t have anything on your document which backs up your findings, have another look, just to be sure. However, don’t force similarities if they’re not there.


How similar does your work look to existing work?

Photo by Chan on Unsplash

4. Look for areas where your research builds on existing work

This step is similar to the previous one but can be trickier. Your work might appear to contradict existing work. Novice researchers may panic that this means their findings are ‘wrong’. However, is there a reason your findings are different from Professor Big-Brain’s? Did you speak to different participants? Has the political climate shifted? Perhaps participants reacted differently to you – an Asian female – than to Professor Big-Brain – a white male. If further research would be beneficial, point that out.

Perhaps your findings add a new dimension to a model or set of guidelines. If so, clearly demonstrate this and give yourself a gold star.

You might have an entirely novel finding – something no-one has found before. Again, check the literature carefully so that you can be confident you’ve not missed anything, but if so, use clear language to tell your reader that you have found something new and important. Don’t be shy about this! These are the kinds of findings you might include in bullet points about ‘what this adds to existing research’ when you’re submitting papers for publication.


Don’t let your brilliant new ideas sink without trace  

Photo by Kristopher Roller on Unsplash

5. Think carefully about what your reader needs to know

PGPR clients’ discussion sections are often overly long. We don’t need every detail of the papers you’re comparing your work to. Consider your examiners or the reviewers of your paper as you write. These people tend to be overworked as it is – and reviewing/thesis examination is extra work which they have to fit into their busy day. Do they really need to know exactly how many people Dr Finickity interviewed, or do they just need to know what those participants experienced?

Here is another example from my PhD, demonstrating a succinct comparison:

Several participants talked about a loss of freedom via the restrictions placed on them by their illness, either in terms of being too fatigued to live life to the full, or in terms of more practical concerns such as diet restrictions. Authors of previous qualitative work on ill-health have found similar themes. The restricting impact of ill health in terms of both social life and diet spoken about by Charlotte is reflected in findings by King et al. (2002), in which dietary restrictions and ill health were both found to have a major impact on the diabetic renal patients.

6. Ask PGPR for feedback

Discussion sections are difficult – but the PGPR team has plenty of experience writing, marking and examining these chapters. We are happy to offer feedback on how to get your discussion chapter just right, so fill in the form below if you would like some extra support.

Top tips for formal academic writing

The world of academic writing can be a terrifying place, full of tricky rules and customs. If you’re a student working on an essay, thesis or paper, you might have been told that the writing you’ve poured your heart and soul into is too informal. This vague bit of feedback isn’t much use on its own. But fear not, the Post-Graduate Proof-Reader is here to remove the mystery with some tips which will allow your words to rub shoulders with the greats.

You don’t have to be scared of academic writing any more!

Photo by Photo Boards on Unsplash
  • No abbreviations

You may find you’re using more shorthand than you realise. You’re sadly unlikely to be encouraging your readers to BYOB to your essay – but you may well be using eg, ie or etc. These are not good academic language, so always make the following substitutions:

eg = for example

ie = such as

etc = and so on

The only exception to this is acronyms. If you’ve defined a term with an acronym the first time you use it, use that acronym each subsequent time. For example: In this thesis, I have used interpretative phenomenological analysis (IPA). Smith devised IPA in the mid-90s…

  • Hedging your bets

Research is a tentative world. Experiments are flawed. Some results are replicated, others are disproven. When you’re analysing qualitative data, there is often room for another interpretation. As such, it can be hard to know when you can confidently state that such-and-such a finding is bona fide or when you should be more cautious.

Think about your subject matter, as this will help you decide on your tone of voice. Let’s look at an example from some hypothetical qualitative findings. If a participant you’ve named Emma says:

I finally had to accept that I needed dialysis, and that made me really depressed

you can afford to be bold in how you phrase your interpretation. You don’t need to say ‘It appeared that the thought of dialysis distressed Emma’; she’s telling you in clear and unambiguous language that she was depressed, so it’s fine to state that as a fact.

However, Emma might go on to say:

Although the ward was chaotic, the only company I had was the bleeping of the machines

This is more ambiguous, so reflect that by saying ‘it seems Emma felt isolated.’ You can apply the same logic to the rest of your paper; if an existing finding or theory is uncontroversial, report it as so. Hedge your bets with anything more ambiguous.

Be cautious when describing ambiguous findings – or you could fall into dangerous territory!

Photo by Nicolas Cool on Unsplash
  • ‘The model states…’ – or does it?

Remember that models, papers, theories and chapters are not sentient and cannot suggest, confirm or deny anything themselves. It is always the people who wrote or created those ideas who have the agency.

For example, ‘a recent paper agrees’ is not right, whereas ‘the authors of a recent paper agree’ is. ‘The ANOVA demonstrated’ is not right, whereas ‘our use of ANOVA demonstrated’ is.

This can be a hard one to get right but keep trying and it will become second nature. This article on the issue from Walden University is helpful if you want to read more about this.

*Note that in APA 7, the rules on anthropomorphism, as this rule is known, have relaxed. Check out our blog to read more about this.

  • Humanising language

It’s important for all academics – and especially psychologists – to use language in a way that is respectful of people and mindful of diversity. For example, rather than the unwieldy ‘he or she’, use ‘they’, which is not only neater but also makes space for people who identify as non-binary.

Always state a person’s humanity before other identifying factors, especially factors which might be stigmatising. So rather than ‘HIV patient’, you should say ‘person living with HIV’. ‘Participants’ is a better word than either ‘subjects’ or ‘patients’ as it implies an active rather than a passive stance.

Not only is language like this better for humanity, it is also in line with British Psychological Society (BPS) guidelines and the standards of most journals where you might be sending your work.


Share the love by using language which puts people first

Photo by Hian Oliveira on Unsplash
  • Hire PGPR

If all this feels like far too much like hard work, don’t worry – just get in touch with us via the form below, and we can do all the heavy lifting for you.