Developed over the last fifteen years interpretative phenomenological analysis (IPA) has established itself as an increasingly popular qualitative research method for psychologists, particularly in the fields of clinical and health psychology. An accessible and flexible approach, the clarity and rigour of IPA's analytic procedure, along with its inclusion in the curricula of courses in organizational psychology, leave it poised to make a significant contribution to the field of management research.
IPA takes its place in the broad and diverse tradition of phenomenological (q.v.) approaches to inquiry which has its roots in the transcendental phenomenology of Husserl. Such approaches tend to be concerned with the ways particular individuals experience the world in their particular contexts rather than with abstract generalizations about the objective nature of the world (Giorgi and Giorgi, 2003) [existential phenomenology].
IPA is phenomenological in this sense, dealing as it does with individuals’ personal perceptions or accounts of phenomena rather than striving to arrive at objective statements regarding these phenomena. IPA is also an interpretative endeavour, the researcher attempting to get close to the participants personal world, to take an ‘insider perspective’, while acknowledging the necessary role played by the researcher's own perceptions and concepts in making sense of other peoples’ accounts of their experience [field studies].
The idiographic character of IPA should also be noted [individualism]. Individual cases provide the starting point and general categorizations are only gradually developed from these. The individual voices of participants are privileged even in those IPA studies which present their findings in more general terms.
Theory development is not a necessary aim of IPA, which values richness and depth of description of focal phenomena over explanation. Such a description may of course lead to the development of explanatory or theoretical constructs.
A further distinctive aspect of IPA is its acceptance of a connective chain between the accounts individuals give and those individuals’ underlying cognitions [causal cognitive mapping; cognitive mapping]. The recognition of cognitive entities such as beliefs and attitudes provides a bridge between IPA and the social cognitive approach in psychology.
A typical IPA project might proceed along the following lines. Participants are selected purposively to provide a sample homogeneous with regard to their experience of a particular [Page 116]phenomenon (e.g. becoming a mother, being promoted, undergoing training). They are then asked to describe this experience. Semi-structured interviews employing open-ended and non-directive questions are the most usual way of eliciting the description and the interviews are recorded and transcribed. Accounts using diaries, journals or other means may also be considered. IPA is a way of engaging with and making sense of such participant-generated texts. These texts are analyzed one at a time. To begin with the researcher will read through the transcript a number of times, noting the initial responses and interpretations prompted by the account. The researcher will then attempt to methodically identify and record themes which seem to capture the gist of what is being said by the participant [conversation analysis]. The next stage involves looking for connections and similarities between themes and grouping them into a more manageable number of superordinate themes [matrices analysis]. This may occur in a number of stages. Eventually, a summary table of overarching themes is produced. This aims to encapsulate the essence of the researcher's reading of the participant's account.
Analysis of the other participants’ data usually proceeds in one of two ways. Each participant's account may be treated in the same manner as the first so that a collection of individual master themes is gradually accumulated which can then be integrated into a set of overarching group themes. Alternatively, the table of themes from the first participant may be used as a template (q.v) to code the material from further participants, the template developing and undergoing revision as each participant's account is analyzed. The final integrated list of group themes should aim to capture the quality of the participants’ shared experience of the focal phenomenon and to reveal something about the nature of that phenomenon.
IPA projects are often written up in a fairly conventional manner, with introduction, method, results and discussion sections. The most distinctive aspect of IPA reports is the analysis/results section, organized as it is around the themes that emerge from the analysis, aiming to provide a coherent account of the participants’ experience, using quotation to illustrate that account and distinguishing between participant report and researcher interpretation. A detailed description of an IPA project, with illustrative examples of each stage in the research process, may be found in Smith and Dunworth (2003).
IPA has only recently begun to receive critical attention. For example, Willig (2001) highlights IPA's reliance on the representational validity of language [representations; semiotics], notes the possible constraints on the applicability of an approach which requires participants to be able to reflect upon their experience in interesting ways, and suggests that IPA's concern with the how rather than the why of experience may constitute a further limitation.
IPA is a particularly useful approach when examining process (q.v) and change. It provides an accessible, flexible, researcher- and participant-friendly method for exploring the experiences of individuals and groups. As with all phenomenological approaches, however, the emphasis on subjective accounts exposes the method and subsequent findings to concerns of generalizability, reliability (q.v.) and replicability. Moreover, theory development is only ever an incidental outcome rather than a defining purpose, making this a practically mannered, rather than overtly explanatory, technique.