Phenomenological Research

What makes a study phenomenological

Interviewing people about an experience and analysing what they said is not automatically phenomenology. Plenty of theses claim the label for what is a thematic analysis of interview data, and it is one of the easier claims to dismantle in a viva.

A phenomenological study is organised around getting to the experience itself. That has consequences for every stage:

• Participants are people who have actually lived the phenomenon, recruited for that reason alone. Homogeneity is a strength here, not a limitation to apologise for — you want people who share the experience, so that what varies is how it was lived rather than what was lived.
• Interviews stay with concrete experience. “Take me to the moment you realised” is a phenomenological question. “What do you think about the policy” is not — it collects opinion, and opinion is not experience.
• Analysis resists explaining. The pull towards causes and consequences is strong and has to be actively held off, because explanation belongs to a different kind of study.
• Findings describe the structure of the experience — what is essential to it, what varies — rather than listing topics that came up.

The test worth applying to your own transcripts: are participants describing what happened and what it was like, or are they offering views about it? A dataset made of views cannot support a phenomenological analysis however it is coded.

Descriptive or interpretative

The two traditions diverge on what you are meant to do with your own prior understanding, and everything downstream follows from that. You are expected to state which you are in.

Descriptive phenomenology

Rooted in Husserl and developed methodologically by Giorgi and by Colaizzi among others. The aim is to describe the essential structure of an experience — what makes it the experience it is, across the people who have had it.

It asks you to set aside your existing assumptions, theories and expectations so the phenomenon can appear on its own terms. That setting-aside is bracketing, and it is the practice most often claimed and least often evidenced. Doing it properly means writing down what you already believe about the phenomenon before you start, returning to that record during analysis, and describing in the thesis how it was managed. A single sentence asserting that assumptions were bracketed is not evidence of anything.

Interpretative and hermeneutic phenomenology

Rooted in Heidegger and developed by van Manen and others. It holds that setting your understanding aside is not possible: you are already inside the world you are studying, and your prior understanding is what makes interpretation possible in the first place.

So instead of bracketing, you make your position explicit and work with it reflexively. The analysis moves between the parts and the whole — individual passages and the developing overall account — in what is usually called the hermeneutic circle.

Choosing this tradition and then claiming to have bracketed your assumptions is a contradiction, and a visible one.

Interpretative phenomenological analysis (IPA)

Developed by Smith and colleagues, and widely used in psychology, health and education. It is interpretative rather than descriptive, and has three commitments worth knowing.

It is idiographic: each case is analysed thoroughly in its own right before any move to compare across participants, and the individual case remains visible in the write-up. It works with small, homogeneous samples — often single figures, with published guidance for student projects suggesting notably few participants, precisely because depth per case is the point. And it involves a double hermeneutic: you are making sense of the participant making sense of their experience, which is why the analyst’s interpretive role is acknowledged rather than minimised.

IPA is the most clearly specified of the phenomenological approaches, which makes it the easiest to follow correctly and the easiest to be caught departing from.

How the analysis proceeds

The specific procedures differ between traditions and you should follow and cite one rather than assembling your own. The shared shape, though, looks like this.

1. Read each transcript whole, more than once, before doing anything analytic to it.
2. Work through it closely, staying near the participant’s own language. Descriptive approaches identify meaning units; IPA makes exploratory notes on content, language and concepts.
3. Move from the participant’s terms towards analytic ones, without leaving the experience behind. This is the step where studies drift into ordinary thematic analysis.
4. Complete one case before starting the next, in idiographic approaches, so a later case does not silently determine what you notice in an earlier one.
5. Look across cases for shared structure and for meaningful divergence — divergence matters, and flattening it into consensus loses what the approach is for.
6. Write it as an account of the experience, using extracts as evidence and keeping individual voices identifiable rather than dissolved into aggregate themes.

Where phenomenological studies come apart

• Opinions instead of experience. The commonest failure, and it originates in the interview guide rather than the analysis. Fixable only before collection.
• Bracketing asserted, not evidenced. No pre-reflection record, no account of how it was managed.
• Tradition and practice contradicting each other. Claiming Heideggerian phenomenology and then bracketing; claiming descriptive phenomenology and then interpreting freely.
• Thematic analysis wearing the label. Themes that name topics, no attention to the structure of the experience, individual cases dissolved into aggregates.
• A sample too large and too varied. Thirty participants with heterogeneous experiences cannot be analysed at the depth the approach requires, and heterogeneity means you are no longer studying one phenomenon.
• Explaining rather than describing. Sliding into causes, barriers and recommendations, which belong to a different design.