Interpretive phenomenological analysis (IPA) explores lived experience and how people assign meanings to make sense of their experience (Smith 2004). IPA attempts to make sense of the participant’s subjective world through a process combining descriptive and interpretive activity (Willig 2001). IPA is descriptive in that it attempts to present an account of subjective experience; however, it is also acknowledged that such experience is never directly accessible. Thus, IPA is also interpretive as it acknowledges the researcher’s role in creating a thematic account (Smith et al 1997). IPA is not a hypothesis-driven approach, but rather answers broader research questions. IPA can be used to question or further explore knowledge gained from the existing literature (Smith 2004) by focusing primarily on participants’ accounts of their experiences. It is only at the end of an IPA that the findings are then integrated with existing research.

This chapter will explore the theoretical background of IPA and how it differs from other qualitative approaches. It will provide an outline of how to conduct IPA research and a practical illustration of how to analyse an interview transcript using IPA. The final sections of this chapter will focus on the use of IPA in health research, focusing in particular on how IPA has contributed to the field of dementia research.

The origins of IPA can be found in philosophical traditions of phenomenology, in theories developed within the social sciences, and in social cognitive psychology. From the social sciences, IPA has been influenced by concepts derived from phenomenology and symbolic interactionism (see Chapter 21 and Chapter 22 for additional information on these approaches). Moreover, both symbolic interactionism and phenomenology contribute to the foundations of IPA through their emphasis on the importance of how events are perceived and on the meanings that people attribute to events (Smith 1996). IPA also shares with social cognitive approaches a conviction that there is a link, albeit an imperfect one, between verbal accounts, underlying cognitive processes and behaviour (Smith et al 1997). The accounts people present in interviews are understood as relating to internal psychological meanings and experience. This is particularly relevant when considering the applicability of IPA in health research (Smith & Osborn 2004) and may help to account for the frequent focus in IPA findings on psychological issues relating to self and identity.

IPA aims to explore subjective experience (Smith 2004 and Willig 2001). IPA proposes that people will experience the same event differently because of the different meanings they attribute to the event (Willig 2001). IPA, although expressly ‘interpretive’, incorporates elements drawn from both descriptive and interpretive phenomenology. The descriptive phenomenological aspects of IPA are concerned with the aim of identifying the individual’s personal perceptions of the phenomenon under investigation (Smith et al 1999) and developing a credible account of these. However, IPA recognises that insight into the personal world is elicited only through a dynamic, interactive research process involving interpersonal engagement which brings with it an inevitable element of interpretation. Thus, an IPA is always an interpretation of the participant’s experience, and is dependent upon, and complicated by, the researcher’s own beliefs, assumptions and understandings (Smith 2004), which operate both within the interview context and subsequently as the researcher engages with the interview transcripts. This interpretive activity distinguishes IPA from the purely descriptive phenomenological approach. IPA researchers must therefore attempt to acknowledge pre-existing values, assumptions and beliefs that may affect the interpretation of data, and to reflect on factors that may influence their interactions with the participant or the text. Through acknowledging these influences and possible biases, the researcher attempts to reduce their impact on the analysis.

Some of the assumptions underlying IPA have been challenged. In particular, researchers have questioned the extent to which IPA can explain, rather than simply describe, the nature of individual experience (Willig 2001). Of course, description may itself be valuable, especially in the case of poorly understood phenomena, and this kind of information can contribute to enhancing clinical practice. Arguably, this criticism can be countered by the observation that many IPA studies do move beyond simply describing themes in order to build dynamic and interactive models – or process-based accounts – as part of the interpretive element within this approach (as an example, see Clare 2003). Unlike grounded theory, the main aim of IPA is to provide an account of participants’ experience with regard to a particular phenomenon, rather than developing a theory of that phenomenon. IPA in contrast to grounded theory is more concerned with answering specific research questions and providing an account of participants’ experiences (Brocki & Wearden 2006), and the data collection and analysis phases are usually conducted sequentially. The resulting account is necessarily a thematic one, presenting key elements that are shared across the accounts of a group of participants. It has been noted that focusing on identification of themes across accounts is likely to imply a lack of attention to narrative elements. IPA does not generally focus on narrative, story or plot and is suitable for use when the primary aim of the research endeavour is to find the key thematic components of a particular phenomenon, or experience.

From a social constructionist perspective, there has been some critique of the reliance in IPA on participants’ use of language to communicate their experiences. Willig (2001) commented that since people can describe an event in different ways, IPA may be examining how people talk about experiences in a given context rather than the nature of the experiences themselves. However, as previously noted, IPA is based on the assumption that what people have to say about their experience does reflect something of their ‘inner world’. While both IPA and discourse analysis recognise the importance of language in describing people’s experiences, these two approaches differ with regard to the link between behaviour and underlying cognitions (Smith et al 1999). Discourse analysis interprets verbal responses as behaviours and minimises the link between overt behaviour and underlying cognitions. IPA, on the other hand, with its links to social cognitive psychology, proposes that there is a link, be it direct or indirect, between people’s verbal responses and their underlying thoughts and feelings (Smith 1996). Therefore, IPA endeavours to study the cognitions underlying these verbal responses. Again, this may help to account for the widespread application of IPA in clinical and health research since it is typically through talk that individuals and practitioners negotiate the meaning of, and responses to, physical symptoms or psychological states.

The following section will explain some of the factors to consider when planning a research project using IPA, such as sample characteristics and data collection. This section will also explain, step-by-step, how to analyse interview transcripts using IPA.

This section will provide an outline of how to analyse an interview transcript using IPA. Analysis using IPA can be done either by hand with or without the assistance of a basic word-processing package, or through using computer-assisted qualitative data analysis software such as NVivo. The example presented in this chapter was analysed by hand in order to illustrate the various stages of IPA. When learning to use IPA we would suggest that it is advisable to begin analysing transcripts in this way, in order to become fully familiar with the method, before progressing to the use of data analysis software.