Methodology Guide
Interpretative Phenomenological Analysis
Smith, Flowers and Larkin's idiographic approach to lived experience research, and how QualIntel OS supports it.
What is Interpretative Phenomenological Analysis?
Interpretative Phenomenological Analysis (IPA) was developed by Jonathan Smith and colleagues (Smith, Flowers & Larkin, 2009). IPA investigates how individuals make sense of their lived experience — their thoughts, feelings, perceptions, and understanding of events. It uses a process called the double hermeneutic: the researcher is interpreting the participant's own interpretation of their experience.
IPA typically uses small, purposive samples of 3–8 participants. Analysis involves close, careful reading of individual cases before moving to cross-participant patterns. The emphasis is on idiographic depth — understanding each person's account in its own right — before making broader claims about the group.
Commonly used in: Psychology, health sciences, counselling, education, nursing, organizational behaviour
Why rigour documentation matters
Examiners of IPA work expect close textual engagement — evidence the researcher read each transcript carefully, not skimming for themes. They expect convergence and divergence between participants to be documented, the double hermeneutic to be acknowledged, and interpretive claims grounded in specific extracts. Because IPA uses small samples, any over-reliance on a single participant's voice must be noted and addressed.
How QualIntel OS supports Interpretative Phenomenological Analysis
- IPA mode recalibrates quality checks for small samples — the over-reliance threshold is adjusted for 3–8 participant sets rather than triggering false positives at large-sample norms
- Evidence review supports close, document-by-document engagement with individual transcripts
- Convergence and divergence across participants visible in the codebook evidence panel
- IPA synthesis prompts foreground idiographic analysis of individual accounts before cross-case patterns
- Complete audit trail of researcher's interpretive decisions at every evidence and theme level
- AI Disclosure Statement documents AI's role (candidate evidence surfacing only) vs researcher's role (interpretation, confirmation, and analysis)
Frequently asked questions
Is IPA compatible with AI-assisted analysis?
IPA's emphasis on the double hermeneutic and researcher interpretation makes AI assistance appropriate only when the researcher retains full interpretive control. QualIntel OS does not interpret participant accounts or generate interpretive claims — it surfaces candidate transcript segments for you to review. All interpretive decisions remain with the researcher, and the AI Disclosure Statement documents this clearly for examiner review.
How does QualIntel OS handle IPA's small sample size?
Quality checks in QualIntel OS are recalibrated in IPA mode for the typical 3–8 participant range. The over-reliance warning thresholds (which flag when too much evidence comes from a single participant) are adjusted so they suit small purposive samples. The quality check report shows evidence distribution across participants so you can address convergence and divergence in your write-up.
Does QualIntel OS support the double hermeneutic in IPA?
QualIntel OS does not interpret participant accounts — that is the researcher's role in the double hermeneutic. It surfaces candidate transcript extracts that may be relevant to a code, supporting your close reading. The synthesis editor in IPA mode prompts you to articulate the interpretive moves you are making, and the reflexivity template supports you in documenting your positionality as an interpreting researcher.