Making the most of your IPA data: How to beat some common errors

If you’re a student or academic conducting an IPA study, you’ll know that the data is where the magic happens. Interpretative phenomenological analysis is all about inductive work – that is, immersing ourselves fully in the data so that we can explore what’s going on for the participants and then communicate our interpretations of those experiences to readers.

 

This means that the quotes you choose to present in your findings section – along with the interpretations of those quotes – are perhaps the most important element of your dissertation, thesis or paper. Selecting the quotes that best illustrate the participants’ experiences and that allow you to demonstrate how you reached your findings will stand you in really good stead when you reach your viva or are facing peer review.

 

There are some mistakes I see frequently in findings sections, so I’ve written this blog to help IPA researchers identify the best quotes to share and present their interpretations in the optimum way.

 

1. Select the data that fully evidences the points you want to make

A common issue I see is quotes and interpretations that don’t match up. When you’re immersed in your data, it’s easy to forget what your readers don’t know. This means you might state interpretations as confident findings – but without the data to back it up, examiners and peer reviewers won’t know what led you to those ideas.

While it’s fine to communicate some contextual information without data – for example, you might explain that you’d asked a certain question just before this data was given to you, or you might set the scene for what the participant is talking about – any interpretive findings must be backed up by evidence. Therefore, once you’ve got a draft of your findings section, check all the claims you’re making about the participants’ experiences and ensure that the data that led you there is on the page.

 

2. Ensuring that the quotes you present are long enough to paint a picture

Another frequent problem in the theses I look at is that the quotes are uniformly too short. Of course, not every quote you include can or should be a full paragraph, but in IPA, if all your quotes are one-liners, this is a red flag to examiners.

 

IPA researchers tend to explore complex, emotive topics about which little is currently known. Subjects like these cannot be adequately illustrated through short quotes alone.

 

Additionally, IPA analysis is about digging deep into the data – exploring the messiness, the contradictions, the hidden meanings. Analysis like this is often more fruitful if it’s been applied to longer sections of data.

 

3. Being truly interpretive rather than descriptive

Building on the above, another common error is that even if great quotes have been presented, the analysis that accompanies them is too descriptive. IPA is a deeply linguistic form of qualitative analysis, so we should be taking the time to really dig into the language that participants are sharing with us and considering how it adds to our understanding of their lived experiences.

 

Let’s take this piece of made-up data, from a made-up study about junior doctors’ mental health, as an example:

 

Sarah: I was working 14 hour days, was battling sleep on my drive home every night. I was fighting to do well at work, but when I got home after every shift, I was destroyed. I felt completely laid to waste. Yet I had to plaster a smile on and do it again the next day.

 

A researcher might state that Sarah’s long working hours are harming her mental health. And that’s correct, of course. But for an IPA study, this interpretation doesn’t go far enough.

 

Let’s take a closer look at this data, which contains a few striking words and phrases – note that the participant was ‘battling’ and ‘fighting’, and yet despite this, she was ‘destroyed’ and ‘laid to waste’. These words conjure up the image of a bloody battle – a battle which Sarah is losing and which has left her ruined, a shell of her former self.

Her only defence against the war zone of work is to ‘plaster on a smile’, a phrase which shows that her cheerful work persona feels like a performance. Additionally, a sticking plaster feels completely inadequate for facing the battle of work she has described.

 

Ok, so I made this data up and deliberately loaded it with language that could be quickly unpicked in this way, but it’s amazing how often participants pepper their words with vivid imagery that, when analysed, can provide a resonant insight into their life worlds.

 

Your data will almost certainly be full of interesting language, so don’t miss the opportunity to analyse it deeply. Insights like this will advance your work greatly.

 

4. Show divergence as well as convergence

Any student of IPA can recite the idiom that we’re looking for divergence as well as convergence. However, it’s rare for me to see divergence mentioned in findings sections. Seeking and discussing convergence is easier and – perhaps – feels safer. But divergence is important too. Why? Because human experience is messy and varied, and cannot be reduced to yes or no experiences, much as we might wish it could be!

 

Divergence can take several forms. It could be that most participants report a certain emotional reaction to (say) a cancer treatment, and another reports the opposite emotional experience.

 

If you were conducting a study about bereavement, all participants may share the experience of anger, but for some, this takes the form of shouting at friends and family, while others take up boxing or direct their anger at themselves.

 

Or perhaps you’re doing a study about waiting for a liver transplant. Participants may attach a similar meaning to an experience while expressing it in very different ways. For example, most participants might describe putting their lives on hold because of uncertainty about the future. However, another might respond to the same uncertainty by embracing new experiences and living more intensely in the present. Although their behaviours differ, both accounts may reflect a shared awareness that the future can no longer be assumed.

 

All such examples of divergence should be included in your write-up, as they all demonstrate different aspects of the experiential elements you’re presenting through your analysis.

 

5. Structure your quotes and interpretations to tell a narrative

A final common error I see is a failure to structure subthemes and GETs so that they tell a coherent story. Instead, students sometimes just start with one participant, quoting all their relevant data, along with interpretations, before moving onto the next one.

 

No matter how insightful your analysis, this is not a good way to write up IPA. It leads to repetition and means you will miss out on showing how one element of a subtheme leads to another, which builds on the first and paves the way for the next.

 

As you create your PET and GET tables, think about how your findings can be structured to tell a story that best reflects the various experiences of the participants and has the potential to offer a resonant example of the meanings behind those experiences.

 

You will almost certainly have to keep tinkering with this structure as you write up, but that’s fine as this is part of IPA.

 

 

If you’re currently analysing IPA data and would value specialist input on theme development, interpretation, or analytic depth, my colleague Rachel Starr and I offer dedicated IPA mentoring via IPA Insights. We support researchers at all stages of the IPA process, from early coding through to thesis write-up and viva preparation.

You can contact IPA Insights here.

Alternatively, if you’re looking for proofreading or written feedback on your academic work, you can contact PGPR via this form.