03 · What You Need to Know
How to Define the Unit of Analysis in Narrative Inquiry
What Does Unit of Analysis Mean in Narrative Research?
The unit of analysis identifies the primary entity or phenomenon about which the study makes interpretive claims. It establishes the level at which evidence is organized and conclusions are developed.
In narrative research, this decision requires particular care because stories can be embedded within other stories, told by several people, revised across encounters, or circulated through institutions.
For example, a researcher may collect interviews from teachers but analyze how each teacher constructs a professional identity through a narrative account. Alternatively, the researcher may examine particular stories about academic integrity or investigate how an institutional narrative changes across documents and meetings.
These studies may use similar materials while making claims about different analytical objects.
Narrative research does not have one universally prescribed unit of analysis. Its methodological traditions differ in whether they emphasize experience, narrative configuration, storytelling practices, identity, or socially circulating accounts.
Why Is the Participant Not Automatically the Unit of Analysis?
A participant is a person who contributes evidence to the research. That person may tell one extended story, several distinct stories, or multiple versions of an account across different encounters.
For instance, one teacher might recount a story about first encountering AI, another about negotiating a difficult assessment decision, and a third about professional disagreement with colleagues.
If the research investigates how the teacher constructs an overall professional trajectory, these accounts may be examined together as part of one developing narrative.
If the research investigates how teachers narrate moments of ethical conflict, each relevant narrative episode may become an analytical unit, while its relationship to the broader account remains important.
The distinction matters because counting participants does not establish what was analyzed. Six participants might contribute six extended narratives, several dozen episodes, or a collaboratively constructed account of one institutional event.
Participant
The person contributing interviews, stories, documents, or other narrative material.
Unit of observation
The source or encounter through which evidence is collected, such as an interview, conversation, diary entry, or meeting.
Unit of analysis
The principal narrative entity or phenomenon about which the study develops interpretive claims.
These categories can coincide, but they should not be assumed to be identical. An interview may contain several stories, and one narrative may be constructed across multiple interviews or documents.
When Is an Individual's Narrative Account the Unit of Analysis?
An individual's narrative account may be the appropriate unit when the research seeks to understand how a person connects experiences, constructs identity, or develops a meaningful account of change.
For example, a researcher might investigate how an early-career teacher narrates the transition from university preparation to classroom teaching.
The analysis may examine how the participant organizes events, identifies significant relationships, constructs continuity, and interprets professional development.
In this situation, isolated statements about challenges or successes should be interpreted in relation to the broader narrative.
The analytical focus is not the person as a complete psychological or social entity. It is the person's situated narrative account and what that account makes intelligible about the phenomenon.
This distinction helps avoid treating a narrated identity as an independently verified description of everything the participant is or has experienced.
When Is a Complete Story the Unit of Analysis?
A complete story may be the analytical unit when the research focuses on how a particular sequence of events is narrated and given meaning.
For example, a teacher might tell a story about discovering that students had used generative AI in an assessment. The account could include an initial incident, an interpretation of what happened, subsequent conversations, and a resolution or continuing uncertainty.
The researcher may examine how these elements are connected, what the narrator emphasizes, and how the story constructs professional responsibility.
Labov and Waletzky's influential work on oral narratives provides one approach to examining how narrative components contribute to the organization and evaluation of experience.
However, not every narrative follows a conventional beginning, middle, and ending. Researchers should avoid imposing a fixed story structure when the selected narrative tradition recognizes fragmented, nonlinear, or unresolved accounts.
When Is a Narrative Episode the Unit of Analysis?
A narrative episode is a relatively bounded segment of storytelling concerning a particular incident, interaction, or experience. It may be embedded within a longer interview or narrative account.
Episode-level analysis can be appropriate when the research concerns how particular situations are narrated rather than participants' complete life trajectories.
For example, a researcher might examine episodes in which teachers describe confronting suspected AI-related academic misconduct.
The analysis could investigate how narrators establish responsibility, portray institutional authority, or explain their decisions within these episodes.
However, extracting an episode from a longer account can remove important context. Researchers should preserve sufficient information about the surrounding narrative and storytelling circumstances to interpret the episode responsibly.
Is an Event the Same as a Narrative About an Event?
No. An event refers to something that occurred or is represented as having occurred. A narrative is an account through which events are selected, connected, and interpreted.
Several participants may describe the same institutional policy change differently. Their narratives may emphasize different causes, turning points, responsibilities, or consequences.
If the research seeks to reconstruct the event itself, additional historical or documentary evidence may be necessary. If the research seeks to understand how participants narrate the event, the differing accounts become central analytical material.
The distinction is particularly important when participants disagree about what happened.
Narrative research should not automatically treat every account as an independently verified historical record. It can investigate how events are represented without assuming that all narrated details are factually established.
Can a Collective or Institutional Narrative Be the Unit of Analysis?
Yes. Narrative research may investigate stories that circulate across organizations, professional communities, or other social settings.
For example, a university may construct an institutional narrative presenting generative AI as an opportunity for educational innovation. Faculty members may reproduce, modify, or challenge that account in meetings and professional discussions.
The analytical unit might be the institutional narrative itself, including how it develops across official communications and participant accounts.
Alternatively, the researcher might examine competing narratives about the same institutional change.
Such studies require careful attention to where stories are produced, how they circulate, who has authority to tell them, and whose accounts receive recognition.
The narrative should not be treated as a single shared belief merely because it appears in several institutional materials.
How Does the Narrative Tradition Affect the Analytical Unit?
Different narrative approaches direct attention toward different aspects of storytelling.
| Narrative approach |
Possible analytical focus |
Important consideration |
| Experience-centered narrative inquiry |
Narrated experience within temporal, social, and place-based contexts. |
Preserve relationships among experience, context, and the developing inquiry. |
| Thematic narrative analysis |
Meaning and content within individual or collective stories. |
Maintain narrative context rather than reducing accounts to disconnected topics. |
| Structural narrative analysis |
How a story is organized and how its organization contributes to meaning. |
Attend to narrative form, sequencing, and evaluative elements. |
| Dialogic or performance-oriented analysis |
Storytelling as situated interaction involving narrators and audiences. |
Examine how the circumstances of telling shape the account. |
| Collective or institutional narrative analysis |
Stories circulating within groups and organizations. |
Investigate production, circulation, contestation, and social context. |
Riessman's methodological work provides an influential distinction among thematic, structural, and dialogic or performance approaches. Clandinin and Connelly's narrative inquiry emphasizes the situated and relational character of experience.
These approaches should not be treated as rigidly exclusive. Researchers may examine several narrative dimensions, but the analytical unit and procedures should remain coherent with the selected methodological orientation.
Can One Study Have More Than One Level of Narrative Analysis?
Yes. A study may analyze individual narratives and subsequently compare relationships among them. It may also examine narrative episodes within longer accounts.
For example, a researcher could first interpret each teacher's story of professional change, then compare how turning points are constructed across participants.
This does not necessarily mean that the study has several unrelated units of analysis. The researcher can distinguish the primary analytical unit from embedded or comparative levels.
Problems arise when researchers shift between levels without explanation. A finding about recurring narrative episodes should not automatically become a claim about the complete identities of participants or the culture of an entire institution.
Researchers should explain how conclusions at one analytical level are supported by evidence developed at another.
How Does the Unit of Analysis Shape Participant Selection?
The analytical unit influences what kinds of participants and narrative materials are needed.
If the unit is an individual's professional narrative, the researcher may select participants capable of providing relevant accounts of a particular transition.
If the unit is a narrated event, participants may be selected because they experienced or interpreted that event from different positions.
If the unit is an institutional narrative, the researcher may need documents, meetings, public statements, and accounts from participants involved in producing or contesting that narrative.
Consequently, selecting participants and stories should follow the analytical purpose rather than a generic sample-size convention.
The number of participants and the number of narratives should be reported separately when they differ.
How Does the Unit of Analysis Affect Data Collection?
A study focused on complete professional narratives may require extended interviews that allow participants to connect events and reflect on their significance.
An episode-focused study may use prompts about particular incidents, while a study of collective narratives may examine interactions and documents across several settings.
Repeated interviews may be useful when the analytical unit is a developing narrative that changes over time.
However, a single interview can contain substantial narrative material, and multiple interviews do not automatically produce a more appropriate analytical unit.
The decision about collecting narrative data in one interview or over time should reflect the narrative phenomenon and the evidence required.
How Does the Unit of Analysis Affect Coding and Interpretation?
Qualitative coding can support narrative analysis, but the unit of analysis determines how coded material should be interpreted.
If the study examines complete stories, researchers should avoid treating isolated quotations as independent evidence detached from their narrative trajectories.
If the study examines narrative episodes, researchers may compare how similar incidents are constructed while preserving the circumstances of each telling.
Polkinghorne distinguishes analysis of narratives, which may identify patterns across stories, from narrative analysis, which configures events into a meaningful narrative account.
This distinction illustrates why narrative research cannot be reduced to one uniform coding procedure.
Researchers should select analytical techniques that support the particular claims they intend to make about their narrative units.
What Happens When Stories Are Reduced to Themes?
Breaking stories into thematic fragments can obscure chronology, narrative identity, and the relationships among events.
For example, two teachers may describe professional uncertainty about AI. One may present uncertainty as the beginning of a successful transition, while another may narrate it as a continuing threat to professional autonomy.
A cross-participant theme called uncertainty about AI captures a shared topic but may conceal the different meanings uncertainty acquires within each story.
The methodological challenge of preserving individual stories without reducing them to themes is especially consequential when the analytical unit is a complete narrative or narrative trajectory.
Thematic comparison can still be appropriate within some narrative approaches, provided that relevant narrative relationships remain visible.
How Should You Report the Unit of Analysis?
A methodology section should identify the principal analytical unit and explain how it relates to participants, data sources, and analytical procedures.
For example, a researcher might state that the primary unit of analysis was each participant's narrative account of professional identity development, constructed through interviews and relevant reflective materials.
Another study might identify narrative episodes concerning AI-related assessment disputes as its units, while retaining the broader interviews as contextual evidence.
Researchers should also specify whether findings concern individual stories, relationships across narratives, or collective narrative practices.
Clear reporting prevents the analytical focus from shifting unnoticed between people, events, and stories.
Watch Out
Do not automatically equate one participant, one interview, and one narrative unit. A participant may contribute several stories, one story may develop across multiple encounters, and collective narratives may involve many participants. Define the unit according to the research question and the claims the analysis will make.
