‘The person becomes the universe of exploration’

Astrid Coxon meets Jonathan A. Smith, recipient of an Honorary Fellowship of the BPS, best known for his development of Interpretative Phenomenological Analysis.

You were recently made an Honorary Fellow of the British Psychological Society: what was your reaction to this news?

I was taken aback in a sense… it’s an award of such high standing. Previous winners include some of the people who I was most drawn to when I first came into psychology… Jerome Bruner and Ulric Neisser, for example, were pretty influential on me in terms of the way they did psychology. What I've done since is very different, but it's nice to see that line of continuity. It's a mark of recognition for the importance of qualitative psychology within that long psychology tradition.

So where did your personal journey into psychology start?

My first degree was English literature. I soon realised that a lot of what I was drawn to in English literature was actually the psychological constructs. I wrote dissertations on perception in the poetry of William Wordsworth, and social interaction in the plays of Harold Pinter, for example. When I finished, that didn't go away. I still had that yearning. I worked in publishing and journalism, and then went to Sussex to do a conversion course in psychology. It was an exhilarating time. 

I think there's a surprising number of people in psychology who didn't start in psychology. They come from all over the place… that's quite a hallmark of psychology. There's something in common at the heart of it, but it's also a very broad church. Psychology can usefully draw on skill sets from a wide range of other places.

Yes, my own background is in philosophy, and I hear this story from quite a lot of colleagues – they started in different fields and transferred over. 

Tell us a little bit about IPA, Interpretative Phenomenological Analysis. What was your initial inspiration to develop this approach to qualitative research?

It’s a strange process, looking back, a bit scary as well… you realise how long you've been doing this! 

I started my DPhil in Oxford. I felt I had the freedom to explore a range of different ideas… in some ways psychology was less compartmentalised in those days. 

While IPA has developed a life of its own, it started in a much more modest way. I didn’t set out to develop a methodology that was going to be very successful in psychology; I set out to solve a problem, which was doing my PhD. I was interested in personal experience, and the meaning that experiences hold for participants, the way identities transform across the life course. When I looked at most of the methodologies that were available for psychology at the time, they did important things, but didn't really work for the question that I was interested in. The prior experience that I had, academically and professionally, gave me confidence to do what I wanted to do. I knew I had important issues that I wanted to look at: I’d given up a career, in effect, to do that. I had a heavy investment in it. I was going to make sure that I actually did what I wanted to do. 

I looked around at what was available, and drew on ideas from phenomenology and symbolic interactionism. I was fortunate to come across Rom Harré, who became my supervisor at the time, and he was in philosophy. He introduced a whole other ballgame, both fairly classic philosophy in the philosophy of identity and people like Locke and Hume. But also I was very drawn to his writing around personal accounts, the meaning of these things to participants and the importance of the single case. So I mixed a rich brew of interesting ideas and brought them together into something that worked. 

What IPA was becoming was something theoretically grounded, conceptually rich, but also that would enable us to do psychology, empirical in the broadest sense, collecting data from participants, analysing and presenting that material in a systematic way. 

What are the particular hallmarks of IPA as a qualitative approach?

There’s now a complex nexus of different qualitative approaches… a fuzzy set of similarities and differences. Essentially, a key facet of IPA is our commitment to looking at experience – the sense-making that participants are doing around important things that are happening to them. We're not looking at some notion of a pure experience, whatever that would be… it’s what the things mean to participants. I invoke the term ‘double hermeneutic’, which is a bit of jargon, I guess, but to speak to the fact that this is a heavily interpreted process, it’s interpretative at both ends: the researcher is trying to make sense of the participant trying to make sense of X.

The most distinctive feature of IPA is idiography – a strong commitment to looking at each case in turn. The person becomes the universe of exploration at the outset. It's a matter of staying with this person, or this family or this small institution. You do a detailed analysis of what's going on, the connections between the important resonances for that person. It's only after you've achieved some sort of partial closure on the first case that you move to the next one, and then the next one. This is painstaking, it's slow. But if it's done well, it’s an analysis that speaks to some sort of convergence, some sort of patterning across individuals, but at the same time, the idiosyncrasies they make manifest. If you look at that constellation, the different sets of ingredients and the way they're marked… that’s what sets IPA apart from other approaches.

What do you think gets overlooked when people start doing IPA?

Quite often with qualitative research it can seem deceptively straightforward. It's a bit of a cliché now, but the story used to be that students are drawn to qualitative methods because they don't like doing statistics. Qualitative research is just as hard as quantitative research, it's just hard in a different way. The personal demands on the researcher as agent in doing qualitative research – any qualitative research, but IPA in particular – are very great. 

The biggest part of doing qualitative research, and IPA in particular, is the analysis. That’s what takes the time. That is what enables you to present an account to an audience. However, the quality of the data you get sets a cap for the analysis that you can do. You may talk to somebody for an hour. But it's the quality of what happens in that hour that determines the days of analysis you do later, how rich and how far that can go. Getting good data from participants is a skilful activity, but it's not magic. It's something that people can acquire. Some people take to it more readily than others, and I spend a lot of time on that with my students. 

So the analysis can make or break the IPA, but you need good data to start with.

Let's talk about your pain research, what makes that well-suited to IPA?

Pain is fascinating, it’s elusive. It's affected by so many different individual and contextual factors, and each quantitative instrument only gets at one aspect. For a long time, a lot of papers would finish by saying, ‘well, pain’s personal, and we need to know more about that personal experience’. That's where the research finished rather than started. And what is good is that more recently we are finding research that starts there, which begins by asking people for their own personal accounts of the experience of pain.

Pain is an experience that knocks people over, but that's often not recognised. That takes a lot of getting at. If you do sit with somebody, and you get them to start talking about their pain, then rich accounts come out that weren't necessarily anticipated. We got extraordinary data and some of that still powerfully affects me now… for example, paraphrasing a participant, ‘if it was just the pain, it would be okay. It's actually what it's doing to my head. That is the worst part’. It gives an inkling of a whole other story… it’s not just about physical symptoms, it's more the way it's impacting the person, an unsteadiness, the way they think they've changed from the person they were before, an unreliability, sense of shame and stigma that comes with it. There’s a whole series of other very psychological things that are happening when you start to unpick pain.

So much more than can be measured on a scale… 

Could you tell me about the more recent research you've been involved in?

We have a whole series of different topics we are investigating. I am working with Roz Shafran (UCL) on a mixed methods study examining the effectiveness of a modified CBT intervention for young people with epilepsy and psychological difficulties. I am collaborating with researchers in India on how COVID-19 is affecting participants’ experience of work. A key role in my career has been articulating a way of thinking about psychology. I've applied that in so many different areas. I didn't know how conscious that was… probably at some point, it did become conscious, I wanted to show this was a universal approach, rather than it being pigeon-holed as ‘IPA is for research on X’. You can keep expanding the areas it applies to. For example I’ve got PhD students doing projects on: art perception; the impact of divorce on the sense of self; being an actor; and the different mood states in bipolar disorder. 

There are developments I'm particularly pleased about… multimodal things, expanding IPA, for example using it alongside pictorial representations, so that participants create a visual record at the same time as taking part in the interview. Visual methodology is quite a big thing in the social sciences now. In the pain research, starting with work with Jamie Kirkham and now with my PhD student Isabella Nizza, who asks people who are taking part in a pain management programme to draw their pain, and then talks to them about their drawing. These are not artists, but the images they come up with are incredibly striking. And the drawings trigger other thoughts: people tell a very interesting verbal story of their pain, prompted by the drawing. It’s a longitudinal approach too: we're watching the change in their pictorial representation, and in their account of pain as they go through the programme. 

You sound particularly proud of that work… what would you say has been the most challenging research you've been involved in?

The challenges are both particular and constant. Every interview you do, every person you work with, is a new event. To each of those encounters, you bring the experience you have, and some growing confidence that things will work. However, one's always surprised by the person who talks in a different way, who needs particular help in order to begin to feel comfortable to disclose their particular experience. 

In a way, maybe it’s surprising that I can still be surprised! These things are so fluid, so liminal. A lot of my research is on heavy duty things, major life transitions, threats, projects that have existential importance. Then we might research something that seems much less challenging for the person – we’ll encourage students to do something on a positive experience. But then you realise even something that seems relatively straightforward, for a particular person it triggers something… they can get quite upset by something you wouldn't have expected. 

One of the biggest challenges in the early days was getting qualitative research and IPA published… dealing with brusque statements about everything that was wrong with this sort of research… but persisting. That has changed a lot. There’s an increasing number of journals with a track record for publishing qualitative research, but there's always others that still need to be persuaded. 

Do you have any advice for early career researchers approaching journals with qualitative papers, or responding to reviewer comments? 

The process of peer review stays constant throughout one's career. I'm a well-established psychologist, I still have to go through this process, I still get the bruising rejection slips. It's devastating when it happens to you the first time, but remember that's the academic process. 

You've done your research, you're proud of it. But it doesn't stand in isolation. If you want people to know about it, you've got to work out a way of making that available. Look at relevant journals and do searches in terms of what qualitative research they've published before. Journals aren't static, editors change, and that's both frustrating but exciting that a journal could have been going for a long time and been quite conservative, and suddenly there's a change of regime and they're open. Take advice from people, in terms of supervisors and other people, and look at look at where journals are currently, in terms of their receptivity to qualitative research. 

For me, quality not quantity is important. We might have collected data from a range of participants and got a series of phenomenologically-informed experiential themes. If we present all of that in the 5000 words the journal makes available, it'll be superficial… just a taster of each of those. It's quite legitimate to say, ‘we're going to present this part of the data for this journal. We're going to show you at full length what it's like for these participants, we're going to give you enough extracts from enough participants to show you the convergence and the divergence. I've got another paper, which is drawing things together, and maybe that fits in a different place.’ So you're seeing IPA at its best. I don't like journals which say ‘you can publish your qualitative research, but you have to have the five extracts in a table separate from the narrative where you present your interpretation’. That doesn't make any sense to me. 

In terms of dealing with reviewers, I think be straightforward and honest. There's not an expectation that you have to do everything. You can have this whole litany of points made by reviewers, but what journal editors are looking for is integrity and effort. If you’ve made changes, and specifically articulated what those are, be patient and be robust. Over time, a lot of papers will find a home. Somebody will bite, and the paper will be improved through the reviewing process.

Perhaps we should feel reassured that even at this stage in your career, you feel that wounding of being rejected?

Every so often I get a stinging review that says ‘you haven't done this as Jonathan Smith says it should be done’, and I have to go and read up on how to do IPA properly, according to Jonathan Smith! That always gives me a wry smile.

Where you would like to see IPA going: how it might develop, future applications?

There’s been this fairly recent development of more complicated designs, for example multimodal where one collects data in different forms… the visual representation I talked about, and Johanna Spiers did some work looking at interview data alongside poetry. Then there’s a growing amount of longitudinal research… the irony there is that my PhD was longitudinal, then I just got caught up in a whole raft of other things, and now it's being picked up again, and other people are taking it forward. 

At the same time, it's important we don't just follow fashion. I would not want everybody to be doing multimodal longitudinal studies. People should still do pure and simple IPA, just talking to a relatively small number of people about their experience and recording it and analysing it. We want to see both the simple and the more complex working alongside each other.

In the early days of being an academic, I felt I had a handle on all the qualitative research that was being done… the latest take on discourse analysis, I could be aware of that alongside IPA and other things. Now, I can't even keep on top of the IPA. In one sense, that’s great. There's a lot of IPA being published and a proportion of that is very good. But there's quite a lot that's only okay, mediocre. I have mixed feelings about that. Partly I feel frustrated: I look at papers that have the potential to be much better, and I think ‘why didn't they do this?’. The other part of me is more charitable. Many people are doing IPA and qualitative methods on their own. Qualitative research is personally demanding, you need guidance and mentoring. Recently I’ve tried to help that process through, for example, writing papers showing what good work is, how to evaluate IPA and making useful material available on the IPA website. So one of my hopes for the future is that gradually, the proportion of IPA that is good and excellent increases. I'm sure it will. 

The other thing is mixed methods research. Historically, qualitative research has often needed to establish itself in its own terms, rather than being the younger sibling of the quantitative research. Now it's more established and more mature, the next phase for IPA and qualitative research in general is working alongside quantitative researchers as equals. 

When I look at mixed methods research, on the whole I don't think it is yet exploiting the full potential of what the qualitative can bring to it. Human phenomena are complex. We will understand those phenomena better if we are thinking about designs which go across, which link different qualitative approaches together, and are linking qualitative and quantitative methods in more powerful and integrated ways.

-       Jonathan A. Smith is a Professor of Psychology at Birkbeck, University of London.

-       Dr Astrid Coxon is a research associate and teaching fellow, working part-time in the Education Support Team, IoPPN.

BPS Members can discuss this article

Already a member? Or Create an account

Not a member? Find out about becoming a member or subscriber