Descriptive Phenomenology — Giorgi's Descriptive Phenomenological Method

Both methods analyse first-person accounts of lived experience, but they differ fundamentally in their epistemological stance. Giorgi's method is strictly descriptive: it aims to articulate invariant psychological structures across participants and suspends interpretation. IPA (Smith, Flowers, and Larkin) is explicitly interpretive and idiographic — it foregrounds the researcher's interpretive activity and typically attends closely to the particular before moving to shared themes. Giorgi produces a general structure; IPA produces an interpretive account of individual cases.

Descriptive Phenomenology's goal is not statistical representation but the articulation of invariant psychological constituents. Giorgi and other phenomenologists argue that if data collection continues until no new constituents emerge, the sample is adequate regardless of its size. Three participants can suffice for a tightly bounded, well-defined phenomenon; broader or more variable phenomena typically require eight to fifteen. The criterion is constitutive saturation, not numerical adequacy.

Free imaginative variation is the technique used during the transformation step: the researcher mentally varies features of a meaning unit — imagining the experience without this feature, or with it altered — to test whether that feature is essential or merely contingent. If the experience would still be recognisable without a particular feature, that feature is contingent; if removing it would make the experience a different kind of experience altogether, the feature is invariant and belongs in the essential structure. This is the core mechanism by which Giorgi's method achieves a claim to describe essences rather than mere regularities.

With caution. The method requires verbatim, first-person descriptive accounts in which participants describe their experience in their own words at length. Interviews designed for other purposes (e.g., structured surveys or theory-testing protocols) typically contain too many researcher-directed questions and too few open participant descriptions to yield adequate meaning units. Secondary analysis of phenomenological data is defensible if the original interviews were open-ended and rich; otherwise, new data collection is recommended.

Yes. Unlike thematic analysis, which typically yields multiple themes, Descriptive Phenomenology produces one general essential structure: a coherent statement of the necessary psychological constituents of the experience, presented as an integrated whole. Individual constituents are listed within that structure, but the structure itself is the primary finding. Variation across participants is acknowledged in the write-up, but the aim is the invariant core, not a catalogue of differences.