Narrative analysis is a family of qualitative methods for analysing the stories and accounts people tell in order to understand how they make meaning of their experiences. Rather than breaking data into decontextualised codes, the narrative researcher treats each account as a whole — with a beginning, middle and end — and asks how it is structured, what work it does for the teller, and what it reveals about identity, culture and circumstance. In short, you study the story, not just the topics within it.
Use narrative analysis when your research question is about lived experience, identity, sense-making or process over time, and your data take the form of extended first-person accounts (interviews, diaries, life stories, social-media posts). It is most powerful when how someone tells their story matters as much as the facts they report.
What is narrative analysis?
Narrative analysis is an umbrella term for a set of approaches that take the story as the unit of analysis. People do not experience their lives as a list of variables; they organise experience into narratives — sequenced, evaluated accounts that connect events, characters and consequences into a meaningful whole. Catherine Kohler Riessman, whose Narrative Methods for the Human Sciences (2008) is a standard reference, distinguishes several analytic strategies precisely because “narrative” can be attended to in different ways: what is told, how it is structured, how it is performed for an audience, and what visual or material forms it takes.
The common thread is a refusal to fragment. Where many qualitative methods chop transcripts into thematic codes and re-sort fragments across cases, narrative analysis preserves the sequence and context of each account. The order in which someone tells you things, what they linger on, what they justify, and how they resolve (or fail to resolve) the tension in their story are all treated as data, not noise. This makes narrative analysis a strong choice within qualitative data analysis when meaning lives in the shape of the account, not only its content.
Narrative researchers also take seriously that stories are co-constructed. A life-story told to a sympathetic interviewer differs from the same events recounted to a sceptical employer. The teller, the listener, the setting and the cultural stock of available story-lines all shape what gets said. Good narrative analysis therefore attends to the conditions of telling, not just the text.
When should you use narrative analysis?
Narrative analysis suits questions about identity, experience, transition and meaning-making over time. Choose it when:
- Your question is about how people understand and account for their experiences (“How do first-generation students narrate belonging at university?”), not how many hold a view.
- Your data are extended, sequenced accounts — life histories, illness narratives, career stories — rather than short answers.
- Process and temporality matter: turning points, before-and-after, trajectories.
- You want to keep the person whole and study within-case coherence, rather than pooling fragments across many participants.
- You are interested in identity construction — how tellers position themselves as, say, a survivor, a victim, an expert or an outsider.
It is a poorer fit when you need prevalence, comparison across many short responses, or a tightly bounded coding of manifest content — there, content analysis or thematic analysis will serve you better.
The main approaches to narrative analysis
There is no single “narrative method”. Riessman’s typology, widely taught, organises the field into four broad approaches. You may combine them — reading a story structurally and thematically is common — but each foregrounds something different.
1. Thematic narrative analysis
Here the focus is on what is told — the content of the story — but, crucially, themes are derived from and kept within whole narratives rather than pooled across fragmented codes. You ask: what is this story about, and what recurring meanings run through it? Unlike standard thematic analysis, you resist clipping a quotation out of its narrative context; the theme must hold for the case as a story.
2. Structural narrative analysis
Structural analysis attends to how a story is built. The classic toolkit is William Labov and Joshua Waletzky’s model of the “fully formed” oral narrative, which identifies six functional elements. Mapping a story onto these elements shows how a teller orients the listener, builds tension, and signals what the story means to them (the evaluation is often where the analytic gold lies).
| Labov element | Question it answers | Example cue |
|---|---|---|
| Abstract | What, in a nutshell, is this story about? | “I want to tell you about the day I nearly quit my PhD.” |
| Orientation | Who, when, where, what was the situation? | “It was second year, my supervisor had just left…” |
| Complicating action | What happened? What was the turning point? | “…and then my data got corrupted the night before the panel.” |
| Evaluation | So what? Why does this matter to the teller? | “I realised then how alone I’d let myself become.” |
| Resolution | How did it end? What finally happened? | “I emailed my second supervisor and we rebuilt it together.” |
| Coda | How does the teller return us to the present / draw a moral? | “That’s why I never work without a backup now.” |
Not every story contains all six, and a missing element is itself meaningful — an account with no resolution may signal an experience the teller has not yet made sense of.
3. Dialogic / performative narrative analysis
This approach treats storytelling as a social performance produced for an audience. The questions shift to: who is the story being told to, and to what end? How does the teller position themselves and others? What identity is being performed, and how do the interviewer’s prompts shape the telling? Here the interview is analysed as interaction, not as a transparent window onto facts — which is why narrative scholars pay close attention to the dynamics of interviews in research and to their own role in the conversation.
4. Visual narrative analysis
Stories are not only verbal. Visual narrative analysis examines images — photographs participants take or choose, drawings, comics, video diaries, social-media imagery — as narrative texts, usually in combination with talk about them (as in photo-elicitation). The analyst reads composition, sequence and the participant’s commentary together to reconstruct the story the images tell.
What data suit narrative analysis?
Narrative analysis works on any data that carry extended, first-person accounts. Common sources include:
- Interviews — especially life-story, narrative or biographical interviews that invite people to tell, at length, rather than answer in fragments.
- Diaries and journals — solicited (kept for the study) or naturally occurring, capturing experience as it unfolds.
- Life stories and oral histories — whole-of-life accounts, often used in sociology, gerontology and history.
- Social-media and digital narratives — blog series, Reddit threads, Instagram captions and video diaries, where people narrate identity and experience publicly over time.
- Written accounts and letters — reflective essays, illness blogs, testimony.
Whatever the source, you need rich, sequenced material. A dataset of one-line survey answers cannot be analysed narratively, however numerous.
How to do narrative analysis: a step-by-step process
Narrative analysis is interpretive and iterative, but a defensible study typically moves through these stages:
- Frame a narrative question and generate rich accounts. Design data collection to elicit stories — open prompts (“Tell me the story of how you came to…”), minimal interruption, and follow-ups that invite elaboration.
- Transcribe with care. Because how things are said matters, transcribe pauses, repetitions, emphasis and false starts, not just the gist. Keep each participant’s account intact.
- Immerse: read each story as a whole. Read and re-read the full account before fragmenting anything, noting your first impressions of plot, tone and turning points.
- Identify the boundaries of narratives. Decide where a story begins and ends within the transcript — participants embed stories inside wider talk.
- Choose and apply your analytic lens. Map structure (e.g. Labov’s elements), trace themes within the case, examine the performance, or read the visuals — or a combination, justified by your question.
- Interpret meaning and identity work. Ask what the story does for the teller: how they position themselves, what they evaluate, what cultural story-lines they draw on.
- Compare across cases (carefully). Look for patterns and instructive contrasts across participants without dissolving individual narratives into a generic theme list.
- Write up, anchored in whole stories. Present extended extracts so the reader can follow the narrative and judge your interpretation against the evidence.
Structural reading: all six Labov elements are present, and the heaviest evaluation falls not on physical loss but on identity loss (“the strong one”), telling us where the meaning sits for Margaret. Thematic reading (within the case): a clear theme of identity reconstruction through dependence runs across abstract, evaluation, resolution and coda — the story is not really about a stroke, it is about renegotiating who she is. Dialogic note: she frames the moral for an audience of students and, here, the interviewer (“I tell my students…”), performing a redemptive, teacherly identity. Holding the account whole reveals an arc that scattered codes would miss.
How narrative analysis differs from thematic, content and discourse analysis
These methods are easy to confuse because they all work with qualitative text. The decisive difference is the unit of analysis and what gets preserved.
| Method | Unit of analysis | Core question | Keeps the story whole? |
|---|---|---|---|
| Narrative analysis | The whole story / account | How is this account structured and what meaning does the teller make? | Yes — sequence and context preserved |
| Thematic analysis | Codes & themes across the dataset | What patterns of meaning recur across participants? | No — data fragmented and re-sorted |
| Content analysis | Coded units, often counted | What categories appear, and how often? | No — often quantified |
| Discourse analysis | Language in use / social action | How does language construct reality, power and subject positions? | Partly — focus on talk-as-action, not plot |
In practice: thematic analysis would pull “loss of identity” quotations from many transcripts and group them; narrative analysis keeps Margaret’s loss inside her arc from fitness to renegotiated strength. Content analysis might count how many participants mention “help”. Discourse analysis would zoom in on how the word “strong” does identity-work and draws on cultural discourses of stoicism. None of these is superior — they answer different questions.
“Nature and the world do not tell stories, individuals do. Interpretation is inevitable because narratives are representations.” (Source: Riessman, 2008)
Strengths and limitations
Narrative analysis earns its place when meaning is bound up in the telling, but it asks a great deal of the researcher. Weigh the trade-offs honestly in your methodology chapter.
Strengths
- Holistic: keeps each person’s experience whole, capturing arcs, turning points and contradictions that fragmenting methods erase.
- Preserves context: sequence, situation and the relationship between events are retained, so interpretation stays grounded.
- Rich on identity and meaning: superbly suited to how people construct selves and make sense of change over time.
- Gives voice: foregrounds participants’ own terms and framings, valuable for marginalised or seldom-heard groups.
- Reflexive and transparent: treats the interview as co-constructed, encouraging honesty about the researcher’s role.
Limitations
- Subjectivity: interpretation is central and unavoidable, so different analysts may read the same story differently; you must argue for your reading from the text.
- Time-intensive: rich transcription and whole-case immersion make it slow; samples are necessarily small.
- Limited generalisability: small, in-depth samples support analytic insight and transferability, not statistical generalisation.
- Demanding to do well: requires skill in interviewing, theory and reflexive interpretation; it is easy to do superficially.
- Representation problems: turning a living, performed story into static text always loses something.
Common mistakes to avoid
- Fragmenting the story. The cardinal error: chopping accounts into coded snippets and re-sorting them — that is thematic analysis wearing a narrative label. If your write-up is a list of themes with quotes ripped from context, you have lost the narrative.
- Treating stories as transparent fact. Narratives are representations, not unmediated truth. Analyse the telling; do not just mine it for “what really happened”.
- Ignoring the listener and the occasion. Forgetting that the story was performed for you, shaped by your prompts and presence.
- No clear analytic lens. Vaguely “reading for stories” without committing to structural, thematic, dialogic or visual analysis — and justifying the choice.
- Thin extracts in the write-up. Presenting one-line quotations so the reader cannot follow the arc or check your interpretation.
- Over-claiming generalisation. Implying that a handful of life stories speaks for a whole population.
Doing narrative analysis well
The best narrative studies are interpretively bold but evidentially disciplined. Anchor every claim in extended extracts; be explicit about which analytic approach you used and why; stay reflexive about your own part in the telling; and resist the gravitational pull toward generic theming. Build narrative thinking in from the design stage — the way you elicit stories shapes everything you can later analyse — and treat your methodology chapter as the place to defend these choices against the alternatives. If you want to see how the method connects to the wider toolkit, set it alongside other forms of qualitative data analysis and the design of your interviews in research.
Turn your stories into a strong dissertation chapter
Our subject specialists can help you design narrative interviews, analyse your accounts rigorously and write up findings that hold together.
