Qualitative research methods: a glossary of terms

A priori code

Also called: a priori codes, predetermined code, preset code, a priori theme

An a priori code is a code defined before analysis begins, drawn from the research question, a theory or earlier studies, and then applied to the data. Template analysis and directed content analysis often start from a few, and in template analysis they may be modified or dropped when they prove unhelpful, with emergent codes added alongside.

A priori thematic saturation

A priori thematic saturation is saturation judged against codes or themes fixed before data collection, so that data are gathered until they illustrate each predetermined category. Saunders and colleagues describe it as one of four models of saturation, used when data are collected to exemplify an existing theory rather than to build a new one.

Adjacency pair

Also called: adjacency pairs

An adjacency pair, in conversation analysis, is a two-part sequence in which one speaker's turn, such as a question, a request or a criticism, makes a particular kind of reply from the next speaker the expected next action. Analysts study how the second part is given, delayed or displaced, for instance by a request for clarification first.

Analytic induction

Also called: analytical induction

Analytic induction is a strategy of qualitative analysis in which the researcher forms a working explanation from early cases and then tests it against each new case, revising the explanation, or the definition of what is being explained, whenever a case does not fit. Introduced by Znaniecki in 1934, it makes the deviant case its main tool.

Analytical framework

Also called: working analytical framework

An analytical framework, in the framework method, is the agreed set of codes, grouped into defined categories, that a research team applies to every transcript once the first few have been coded. It usually goes through several versions before no new codes appear, and it differs from a theoretical or conceptual framework, which sets out ideas rather than codes.

Audit trail

Also called: audit trails, decision trail

An audit trail is the full record a qualitative researcher keeps of how a study was done: decisions on sampling and coding, meeting notes, memos, reflexive notes and the steps from raw data to findings. It lets an outside reader or auditor follow the path of the research and is the main strategy for showing dependability and confirmability.

Authenticity

Also called: authenticity criteria

Authenticity is a set of quality criteria for constructivist qualitative research, set out by Guba and Lincoln alongside trustworthiness, that asks whether a study was fair to every group of stakeholders and whether it changed how participants understand and act on their situation. It comprises fairness and ontological, educative, catalytic and tactical authenticity.

Autoethnography

Also called: auto-ethnography, autoethnographic research

Autoethnography is an approach to research and writing in which the researcher describes and systematically analyses their own personal experience in order to understand a wider cultural experience. It draws on both autobiography and ethnography, and its results are often written in evocative, literary forms. Critics call it self-absorbed or insufficiently rigorous, a charge its proponents reject.

Axial coding

Also called: intermediate coding

Axial coding is the second stage of coding in Straussian grounded theory, in which the researcher puts data fractured by open coding back together by linking categories with their subcategories. Strauss and Corbin guided it with a coding paradigm of conditions, actions and consequences. Glaser criticised this as forcing a preconceived frame onto the data.

Bracketing

Also called: epoché, epoche, phenomenological bracketing

Bracketing is the practice, from Husserl's descriptive phenomenology, of deliberately setting aside one's own assumptions, knowledge and theories about a phenomenon so that it can be described as participants experience it. Whether complete bracketing is possible is disputed, and interpretive phenomenology rejects it, asking researchers to acknowledge and reflect on their preconceptions instead.

CAQDAS

Also called: computer-assisted qualitative data analysis software, computer assisted qualitative data analysis, qualitative data analysis software, QDA software, qualitative data management software

CAQDAS, computer-assisted qualitative data analysis software, is software that helps researchers store, code, annotate, search and retrieve qualitative data, draw diagrams and extract quotations. NVivo, ATLAS.ti and MAXQDA are examples. It manages the analysis but does not do it: deciding on codes, categories and themes remains the researcher's work.

Catalytic authenticity

Catalytic authenticity is the authenticity criterion asking whether a qualitative study prompted action, giving participants the insight or impetus to change their situation. It is one of the five authenticity criteria of constructivist inquiry, and it differs from tactical authenticity, which asks whether participants were actually enabled to act.

Category

Also called: categories, analytic category, conceptual category

A category, in qualitative analysis, is a group of codes that share a meaning, gathered under a label more abstract than any single code. Categories can hold subcategories and are usually a step towards themes or, in grounded theory, towards the concepts of a theory. In qualitative content analysis a set of categories is often the main finding.

Charting

Also called: charting data

Charting, in the framework method, is the stage at which the researcher summarises the coded data from each transcript and enters the summaries into the cells of a framework matrix, one row per participant and one column per code. Good charting reduces the data while keeping the participants' own meanings and pointing to useful quotations.

Co-coding

Also called: collaborative coding, team coding

Co-coding is coding of the same qualitative data by two or more researchers, who compare their codes and agree how each should be applied. It is most useful early on, while the list of codes is being built. Some approaches then measure the agreement between coders, while reflexive approaches use the discussion to deepen interpretation rather than to prove accuracy.

Code

Also called: codes, code label

A code, in qualitative analysis, is a short label that a researcher attaches to a segment of data, such as a phrase, a sentence or a passage, to capture a single idea in it. Codes can be descriptive or interpretive, and they are the building blocks that are later grouped into categories and themes.

Code saturation

Code saturation is the point in a qualitative study at which new interviews stop producing new codes, so the range of issues has been identified. Hennink, Kaiser and Marconi found it after about nine interviews in one study, and argued that it shows when researchers have heard all the issues but not yet when they understand them.

Codebook

Also called: code book, coding frame, coding scheme, coding manual, coding list

A codebook is a document listing every code in a qualitative analysis with its name, a description and examples of when it applies. It keeps coding consistent, especially across a team, and forms part of the audit trail. Codebook and coding reliability forms of thematic analysis depend on one, while reflexive thematic analysis avoids a fixed codebook.

Codebook thematic analysis

Also called: codebook TA, codebook approaches to thematic analysis

Codebook thematic analysis is Braun and Clarke's name for the thematic approaches, such as template analysis and framework analysis, that use a structured codebook or matrix to guide coding and chart the developing analysis, often to make team work easier. It sits between coding reliability thematic analysis and reflexive thematic analysis.

Coding

Also called: qualitative coding, data coding

Coding is the process of labelling segments of qualitative data, such as lines of a transcript, field notes or documents, with codes that say what each segment is about. It makes the data sortable, so that every passage on one topic can be gathered and compared, and it is the first step towards categories, themes or theory.

Coding paradigm

The coding paradigm is a device in Straussian grounded theory for relating categories during axial coding: the researcher asks about the conditions surrounding a phenomenon, the actions, interactions and emotions it prompts, and their consequences. Critics, Glaser among them, argue that applying it forces a preconceived structure onto the data.

Coding reliability thematic analysis

Also called: coding reliability TA

Coding reliability thematic analysis is the form of thematic analysis that treats accurate, reliable coding as the aim, using a structured codebook, several independent coders and a statistic such as Cohen's kappa to measure their agreement. It assumes there is a reality in the data that coding can capture, a view Braun and Clarke contrast with reflexive thematic analysis.

Coding template

A coding template is the hierarchical list of themes at the centre of template analysis, with broad themes containing narrower and more specific ones. The researcher builds a first version from a subset of the data, often seeded with a priori themes, then revises it against the rest of the data until it fits the whole data set.

Colaizzi's method

Also called: Colaizzi method, Colaizzi's descriptive phenomenological method, Colaizzi's seven-step method, Colaizzi's framework

Colaizzi's method is a seven-step procedure for descriptive phenomenological analysis, published by Colaizzi in 1978 and widely used in the health sciences. The researcher reads every account, extracts significant statements, formulates their meanings, clusters them into themes, writes an exhaustive description, condenses it to the phenomenon's fundamental structure and asks participants to verify it.

Condensation

Also called: condensed meaning unit, condensing

Condensation, in qualitative content analysis, is the shortening of a meaning unit to its essential content while keeping its original meaning, for instance by removing filler words or simplifying a long sentence. It adds no interpretation, and it comes before the condensed unit is given a code.

Conditional matrix

Also called: conditional/consequential matrix, conditional consequential matrix

The conditional matrix is a coding device in Straussian grounded theory that helps the researcher connect a phenomenon with the conditions shaping it at both micro and macro levels, from immediate interaction to wider organisational and social conditions. It is used during axial or selective coding and is also called the conditional/consequential matrix.

Confirmability

Confirmability is the trustworthiness criterion concerned with neutrality: whether a qualitative study's findings are clearly derived from the data rather than from the researcher's preferences or imagination, so that others could trace and corroborate them. It is Lincoln and Guba's counterpart to objectivity, shown mainly through an audit trail, reflexivity and triangulation.

Constant comparison

Also called: constant comparative method, constant comparative analysis

Constant comparison is the analytic method, developed by Glaser and Strauss for grounded theory, of continually comparing incident with incident, code with code and category with category, within and across data sources, to find similarities and differences. It moves analysis from description towards concepts, and many other qualitative approaches now borrow it.

Constructivist grounded theory

Also called: Charmaz grounded theory, constructivist GT

Constructivist grounded theory is the version of grounded theory developed by Kathy Charmaz, which holds that data and theory are co-constructed by researcher and participants rather than discovered in the data. It uses initial and focused coding, stresses reflexivity and abductive reasoning, and aims at an interpretive understanding of a process rather than a causal explanation.

Conventional content analysis

Conventional content analysis is the approach to qualitative content analysis in which coding categories are derived directly from the data, without preconceived categories, usually because existing theory on the topic is limited. It is one of the three approaches Hsieh and Shannon described, alongside directed and summative content analysis, and resembles inductive content analysis.

Conversation analysis

Also called: CA, conversation analytic research

Conversation analysis is the detailed study of how people organise social action through talk, using recordings of naturally occurring interaction transcribed in fine detail. Founded in American sociology by Harvey Sacks and Emanuel Schegloff, with Gail Jefferson, it examines turn-taking and sequences such as adjacency pairs, and now studies talk in clinics, classrooms and interviews as well as conversation.

Core category

Also called: central category

The core category, in grounded theory, is the central category around which a theory is built, the one that unites the other categories and to which they are all related. Finding it is a turning point in the analysis, after which coding in classic and Straussian grounded theory is directed at the core category and its links.

Credibility

Credibility is the trustworthiness criterion concerned with truth value: whether a qualitative study's findings are a plausible and accurate interpretation of what participants said and did. It is Lincoln and Guba's counterpart to internal validity, strengthened through prolonged engagement, persistent observation, triangulation, peer debriefing, negative case analysis and member checking.

Critical discourse analysis

Also called: CDA

Critical discourse analysis is a form of discourse analysis that examines how language and texts help build and maintain unequal power relations, linking the detail of a text to its wider social, cultural and political setting. It is associated above all with Fairclough, who treats discourse as the use of language as a form of social practice.

Critical ethnography

Critical ethnography is ethnography that links a detailed account of a group's meanings and practices to the wider social structures and power relations that shape them. It starts, like conventional ethnography, from the participants' point of view, but treats their meanings as mediated by structure and uses the ethnography to question those structures.

Data saturation

Also called: informational redundancy, information redundancy

Data saturation is the point at which new data repeat what has already been collected, so further data collection adds no new information. It is the most common sense of saturation and is often given as the reason for stopping recruitment, though methodologists disagree about how it can be recognised and whether it suits every qualitative approach.

Data triangulation

Also called: data source triangulation, source triangulation

Data triangulation is the use of several sources of data on the same phenomenon, collected at different times, in different places or from different kinds of people, such as patients, their families and clinicians. Agreement across sources strengthens credibility, and differences between them can themselves be revealing. It is one of the kinds of triangulation in qualitative research.

Deductive coding

Also called: theory-driven coding, concept-driven coding, top-down coding

Deductive coding is coding directed by concepts the researcher brings to the data, such as a theory, a framework or earlier findings, often through a set of a priori codes. It suits questions that test or extend what is already known, and many studies combine it with inductive coding so that unexpected ideas are not lost.

Deductive content analysis

Deductive content analysis is qualitative content analysis that starts from a theory, framework or model and codes the data against it, often through a coding matrix built from that model. It tests or extends existing knowledge, in contrast to inductive content analysis, and is close to what Hsieh and Shannon call directed content analysis.

Dependability

Dependability is the trustworthiness criterion concerned with consistency: whether a qualitative study's process was logical, documented and in line with accepted standards for its design, so that others can see how its findings were reached. It is Lincoln and Guba's counterpart to reliability and is usually shown through an audit trail or an inquiry audit.

Descriptive phenomenology

Also called: transcendental phenomenology, Husserlian phenomenology, descriptive phenomenological approach

Descriptive phenomenology is the approach to phenomenology rooted in Edmund Husserl, which aims to describe the essential structure of an experience as participants live it, with the researcher bracketing their own assumptions. The methods of Giorgi, Colaizzi and Moustakas belong to it. It differs from interpretive phenomenology, which holds that description free of interpretation is impossible.

Directed content analysis

Directed content analysis is the approach to qualitative content analysis that takes an existing theory or earlier findings as the source of its initial codes, then uses the data to confirm, extend or refine that theory. It is one of the three approaches Hsieh and Shannon described and works largely deductively.

Discourse analysis

Also called: DA

Discourse analysis is a family of approaches that study language in use, in talk and in texts, to show how it constructs social reality, identities and relationships rather than simply reflecting them. Its forms range from close linguistic study of single texts to broad studies of power and history, and include critical and Foucauldian discourse analysis and discursive psychology.

Discursive psychology

Discursive psychology is an approach to discourse analysis, developed within critical psychology as a reaction against mainstream social psychology, that studies how people use the discourses available to them in everyday talk to build accounts of the world, of themselves and of their thoughts and emotions. It treats psychological matters as things done in interaction rather than as hidden inner states.

Double hermeneutic

The double hermeneutic is the two-layered interpretation at the core of interpretative phenomenological analysis: participants make sense of their own experience, and the researcher in turn makes sense of the participants' sense-making. The idea explains why IPA findings are always the researcher's interpretation and why reflexivity matters so much in the method.

Educative authenticity

Educative authenticity is the authenticity criterion asking whether taking part in a qualitative study helped participants understand the experiences and views of other stakeholders better. It is one of the five authenticity criteria of constructivist inquiry, and it differs from ontological authenticity, which concerns participants' understanding of their own situation.

Emergent code

Also called: emergent codes, data-driven code, inductive code

An emergent code is a code created during analysis because something in the data called for it, rather than one decided in advance. Emergent codes drive inductive analysis and are often added to a priori codes that do not capture everything of interest. Braun and Clarke object to saying that themes emerge, since the researcher makes them.

Emic

Also called: emic perspective, emic approach, insider perspective

Emic describes a perspective from inside a culture or group: understanding behaviour and meaning in the terms that members themselves use and recognise. Ethnography and much qualitative research aim for an emic account, and insider researchers often start from one. It is paired with etic, the outsider's perspective, and many studies combine the two.

Essence (phenomenology)

Also called: essential structure, essences

An essence, in phenomenology, is the set of features that make an experience what it is rather than something else, shared by those who have lived it. Descriptive phenomenology aims to reveal the essence of a phenomenon through bracketing and imaginative variation, while interpretive approaches stress that any account of it is shaped by context and interpretation.

Ethnography

Also called: ethnographic research, ethnographic study, ethnographic fieldwork

Ethnography is a qualitative approach in which the researcher studies the culture of a group, its shared practices, beliefs and meanings, by spending extended time in its natural setting, observing, taking part and interviewing. Its outcome is a detailed, usually narrative, description of that culture, and it suits questions about why a group behaves and believes as it does.

Ethnomethodology

Ethnomethodology is an approach in sociology, founded by Harold Garfinkel, that studies the methods ordinary people use to make sense of everyday situations and to make their actions recognisable and accountable to others. It treats social order as something members accomplish moment by moment in practice, and it prepared the ground for conversation analysis.

Etic

Also called: etic perspective, etic approach, outsider perspective

Etic describes a perspective from outside a culture or group: studying behaviour in the researcher's own terms and categories, often in search of patterns that hold across cultures. A standard questionnaire used unchanged in a new country takes an etic approach. It is paired with emic, the insider's perspective, and the two are often combined.

Fairness (authenticity criterion)

Also called: fairness

Fairness is the authenticity criterion asking whether a qualitative study sought out and represented the views of all its stakeholder groups even-handedly rather than privileging some voices. It involves identifying every group with a stake, eliciting their perspectives and negotiating openly with them about conclusions and recommendations.

Familiarisation

Also called: familiarization, data familiarisation, data familiarization, immersion in the data, data immersion

Familiarisation is the stage of qualitative analysis in which the researcher reads and rereads transcripts, listens to recordings and notes first impressions until they know the data set closely. It opens reflexive thematic analysis and follows transcription in the framework method, and transcribing one's own interviews is a common way to begin it.

Focused coding

Also called: focussed coding

Focused coding is the second main stage of coding in constructivist grounded theory, in which the researcher takes the most frequent and most significant initial codes and uses them to sort and code larger segments of data. Focused codes are more conceptual than initial codes, and some are raised to become categories of the developing theory.

Focused ethnography

Also called: focussed ethnography

Focused ethnography is a form of ethnography, common in applied research such as nursing, that studies a specific activity, problem or setting through short, intensive field visits rather than long immersion in a whole culture. It makes up for its brevity with intensive data collection, often using audio or video recordings, and close, often collaborative, analysis.

Foucauldian discourse analysis

Also called: Foucaultian discourse analysis

Foucauldian discourse analysis is a form of discourse analysis based on the work of Michel Foucault, which examines how discourses, as historically specific systems of knowledge, constitute objects, subjects and power relations. Here discourse means institutionalised patterns of knowledge rather than the formal features of language, and the analysis asks what a discourse makes it possible to think, say and be.

Framework analysis

Also called: framework method, framework approach

Framework analysis is a structured method of qualitative analysis, developed by Jane Ritchie and Liz Spencer at the National Centre for Social Research in the late 1980s for policy research, in which coded data are summarised in a matrix of cases by codes. It suits teams and large data sets and can be used inductively or deductively.

Framework matrix

A framework matrix is the spreadsheet at the heart of the framework method, with one row for each participant or case and one column for each code, into which summarised data are charted. Reading down a column compares cases on one issue, and reading along a row keeps each person's account together in context.

Fusion of horizons

The fusion of horizons is Gadamer's term for what happens when understanding takes place: the interpreter's horizon, the outlook shaped by tradition and language, meets that of a text or another person and is widened by the encounter with difference. It is central to hermeneutic phenomenology, which draws on Gadamer's philosophy.

Glaserian grounded theory

Also called: classic grounded theory, classical grounded theory, classic Glaserian grounded theory, traditional grounded theory, Glaser's grounded theory

Glaserian grounded theory, or classic grounded theory, is the version of grounded theory kept by Barney Glaser after his split with Anselm Strauss in the early 1990s. The researcher enters the field with as few preconceptions as possible and lets a theory emerge through substantive coding, theoretical coding and constant comparison. Critics see positivist leanings in its claim to objectivity.

Grounded theory

Also called: grounded theory methodology, grounded theory method, GT, GTM

Grounded theory is a qualitative methodology for building a theory from data, set out by Barney Glaser and Anselm Strauss in The Discovery of Grounded Theory (1967). Data collection and analysis run together, guided by coding, constant comparison, memo writing and theoretical sampling, until the categories are saturated. It later divided into Glaserian, Straussian and constructivist versions.

Grounded theory lite

Also called: grounded theory-lite

Grounded theory lite is an informal name for studies that borrow some grounded theory techniques, such as open coding and constant comparison, without following the whole methodology or setting out to produce a theory. Braun and Clarke note that it is probably the most common form of grounded theory in use, and that it closely resembles thematic analysis.

Hermeneutic circle

The hermeneutic circle is the idea that a whole can only be understood through its parts and the parts only through the whole, so understanding grows by moving back and forth between them. In interpretive phenomenology and IPA it describes analysis that shifts between single passages and the whole account, revising the understanding of each in light of the other.

Hermeneutics

Also called: hermeneutic approach, hermeneutic inquiry

Hermeneutics is the study of interpretation: how the meaning of texts, actions and experience is understood. It grew out of the interpretation of texts in theology and law, was developed in the nineteenth century by Schleiermacher and Dilthey as a basis for the human sciences, and was turned by Heidegger and Gadamer into a philosophy of understanding itself.

Imaginative variation

Also called: free variation

Imaginative variation is a step in descriptive phenomenology in which the researcher imagines the phenomenon altered in many ways to find which features cannot be removed without it ceasing to be what it is. Those features are its essence. In Moustakas's method it yields the structural description, how the experience happened, beside the textural description of what happened.

In vivo coding

Also called: in vivo code, in vivo codes, in-vivo coding

In vivo coding is coding that uses the participants' own words, often a striking phrase from a transcript, as the label for a code. It keeps the analysis close to how people talk about their experience and is common in grounded theory, where in vivo codes help keep developing concepts grounded in the participants' language.

Indexing (framework method)

Indexing, in the framework method, is the systematic application of the agreed analytical framework to every transcript, marking each passage with the number or abbreviation of the codes that apply to it. It follows the development of the framework and comes before charting, and software helps most at this stage because indexed passages can be retrieved quickly.

Inductive coding

Also called: data-driven coding, bottom-up coding

Inductive coding is coding directed by the content of the data rather than by a theory or a list of codes brought to it, so that codes are built from what participants actually say. It suits exploratory questions, is the usual starting point in grounded theory and conventional content analysis, and can be combined with deductive coding.

Inductive content analysis

Inductive content analysis is qualitative content analysis that builds its categories from the data: the text is broken into small units, each is coded according to what it says, and the codes are grouped into subcategories, categories and main categories. It is close to what Hsieh and Shannon call conventional content analysis and yields a summary of categories, not a theory.

Inductive thematic saturation

Inductive thematic saturation is saturation judged by the appearance of new codes or themes during analysis: it is reached when further data produce no new codes. Saunders and colleagues identify it as one of four models of saturation, centred on analysis rather than on data collection or on building a theory.

Inquiry audit

Also called: external audit, dependability audit, confirmability audit, enquiry audit

An inquiry audit is a review of a qualitative study by an outside auditor who examines its records, such as raw data, coding notes, memos and meeting minutes. Examining the process tests dependability, and checking that findings can be traced back to the data tests confirmability, which is why the two parts are often called the dependability and confirmability audits.

Insider research

Also called: insider researcher, insider status

Insider research is research carried out by someone who shares experiences, identity or membership with the people studied, such as a nurse studying nurses. Insider status can deepen analysis and interpretation, and insider and outsider researchers alike need reflexivity about how their position, and any power imbalance with participants, shapes the work.

Institutional ethnography

Also called: IE

Institutional ethnography is an approach developed by the sociologist Dorothy Smith that starts from the everyday experience of people, often those with little power, and traces how that experience is organised by institutions through texts, rules and procedures. Its aim is to map these ruling relations and show how things actually work.

Intercoder reliability

Also called: inter-coder reliability, intercoder agreement, inter-coder agreement, ICR, coder agreement

Intercoder reliability is the degree to which two or more researchers who code the same qualitative data independently apply the codes in the same way, often reported with a statistic such as Cohen's kappa. Coding reliability approaches treat it as evidence of rigour, while reflexive thematic analysis rejects it, since agreement shows shared training rather than accurate coding.

Interpretative phenomenological analysis

Also called: IPA, interpretive phenomenological analysis

Interpretative phenomenological analysis is a qualitative approach, developed in psychology by Jonathan Smith, that examines in detail how particular people make sense of a significant lived experience. It rests on phenomenology, hermeneutics and idiography: each case is analysed in depth before any claims are made across cases, usually with a small sample.

Interpretive description

Interpretive description is a qualitative approach proposed in nursing by Thorne and colleagues in 1997 for answering applied clinical questions without adopting a traditional methodology such as phenomenology or grounded theory wholesale. It aims to build practical knowledge about experiences of health and illness, and it sits close to qualitative description.

Interpretive phenomenology

Also called: hermeneutic phenomenology, Heideggerian phenomenology, interpretative phenomenology

Interpretive phenomenology, or hermeneutic phenomenology, is the approach to phenomenology rooted in Martin Heidegger and Gadamer, which holds that experience is always understood through a person's lifeworld and so can be interpreted but never simply described. The researcher acknowledges and uses their own preconceptions instead of bracketing them. Max van Manen's work is a widely respected guide.

Investigator triangulation

Also called: researcher triangulation, analyst triangulation

Investigator triangulation is the involvement of two or more researchers in coding, analysis and interpretation, so that each can check and challenge the others' readings of the data. Differences are discussed until an agreed interpretation is reached. It is one of the kinds of triangulation used to strengthen the credibility and confirmability of qualitative findings.

Jefferson transcription

Also called: Jeffersonian transcription, Jefferson transcription system, Jefferson notation, CA transcription

Jefferson transcription is the system of transcription notation devised by Gail Jefferson for conversation analysis, which records not only the words spoken but pauses, overlaps, stress, intonation and other features of how talk is delivered. It aims to capture the details participants themselves respond to while remaining readable without expert knowledge.

Latent coding

Also called: latent code, latent codes

Latent coding is coding that looks beneath what participants explicitly say to the ideas, assumptions and meanings that underpin it. In Braun and Clarke's thematic analysis it sits at one end of a continuum with semantic coding, and the choice between them depends on the research question and theoretical approach, not on which is better.

Latent content

Also called: latent content analysis

Latent content is the underlying meaning of a text, what is implied rather than stated, which content analysis can reach only by interpretation. Latent content analysis relies on the researcher's judgement, and often on theory, to read meaning beneath the surface, in contrast to manifest content, which is taken at face value.

Lifeworld

Also called: life-world, life world

The lifeworld, in phenomenology, is the world as a person lives and experiences it before reflecting on it, shaped by their history, culture and relationships. Hermeneutic phenomenology holds that no one can step outside their lifeworld, so any experience has to be interpreted in its light. Lifeworld research is a phenomenological approach built around the idea.

Line-by-line coding

Also called: line by line coding

Line-by-line coding is coding each line of a transcript or document in turn, naming what is happening in it, as a way of opening up the data at the start of analysis. It is a common technique for open or initial coding in grounded theory because it keeps the researcher close to the data.

Lived experience

Also called: lived experiences

Lived experience is a person's first-hand, subjective experience of something as it was actually lived, including what it meant to them, as opposed to an abstract or second-hand account of it. It is the object of study in phenomenology and interpretative phenomenological analysis, which ask what an experience is like and how people make sense of it.

Manifest content

Also called: manifest content analysis

Manifest content is what a text visibly and literally says, the words and elements on its surface, which content analysis can describe and often count with little interpretation. Manifest content analysis stays close to the text and suits questions about what is present, in contrast to latent content, which concerns underlying meaning.

Meaning saturation

Meaning saturation is the point at which further data add no new understanding of the issues already identified, so that each is understood in depth, with its dimensions and nuances. Hennink and colleagues distinguish it from code saturation and found that it took many more interviews, 16 to 24 against about nine in their study.

Meaning unit

Also called: unit of meaning, meaning units

A meaning unit, in qualitative content analysis, is a segment of text, such as a phrase, a sentence or a conversational turn, that holds one idea relevant to the research question and is coded as a single unit. Deciding how large meaning units should be is an early step, and long ones are often condensed before coding.

Member checking

Also called: member check, member checks, respondent validation, participant validation, participant checking, respondent checking, member validation

Member checking is the practice of returning data, interpretations or findings to the participants they came from, to see whether they recognise them as accurate and true to their experience. It is widely used to support credibility, but critics note that it can sit awkwardly with interpretive work, where the researcher's reading may legitimately differ from the participant's.

Memo

Also called: memos, memoing, memo writing, analytic memo, analytical memo, theoretical memo

A memo is an informal note a qualitative researcher writes during a study to record ideas about the data: what a code means, how categories relate, why a decision was taken. Memo writing is essential in grounded theory, where memos have been called the mortar of the theory, and memos also form part of the audit trail.

Methodological triangulation

Also called: method triangulation, methods triangulation

Methodological triangulation is the use of more than one method of data collection to study the same phenomenon, such as interviews, focus groups and observation, so that each can offset the others' limits. In qualitative research it strengthens credibility, and when it combines qualitative and quantitative methods it overlaps with mixed methods research.

Narrative analysis

Also called: analysis of narratives

Narrative analysis is a family of approaches to texts that take the form of a story, which study how people select, order and evaluate events to make sense of experience. Riessman distinguishes thematic analysis of what is told, structural analysis of how it is told, and interactional and performative analysis of how teller and audience shape a story together.

Narrative inquiry

Also called: narrative enquiry

Narrative inquiry is a qualitative methodology that studies experience through the stories people live and tell, with researcher and participants in a continuing relationship. Named in educational research by Connelly and Clandinin in 1990, drawing on Dewey, it attends to time, social relationships and place, and moves from field texts to research texts.

Negative case analysis

Also called: deviant case analysis, negative case, negative cases, deviant case, disconfirming case analysis

Negative case analysis is the deliberate search for cases or data that do not fit an emerging explanation, followed by revision of the explanation until it accounts for them or their difference is understood. It guards against seeing only confirming evidence, and Lincoln and Guba list it among the strategies that establish credibility.

Netnography

Also called: netnographic research

Netnography is a qualitative method, developed by Robert Kozinets in the 1990s, that adapts ethnography to online communities by systematically observing and interpreting the interactions and cultures that occur naturally in forums, social media and other digital spaces. It is a defined procedure, narrower than online ethnography in general, and raises particular ethical questions about consent.

Ontological authenticity

Ontological authenticity is the authenticity criterion asking whether taking part in a qualitative study helped participants understand their own situation more fully or in more informed ways. It is one of the five authenticity criteria of constructivist inquiry, and it differs from educative authenticity, which concerns understanding other people's situations.

Open coding

Also called: initial coding, open codes

Open coding is the first stage of coding in grounded theory, in which the researcher breaks the data into small pieces and labels them with as many codes as needed, without a predefined list, asking what each piece is about. Charmaz calls the equivalent stage initial coding. The term is also used loosely for any first, inductive pass through qualitative data.

Outcome space

An outcome space, in phenomenography, is the final set of categories describing the qualitatively different ways a group of people understand a phenomenon, together with the logical, often hierarchical, relationships between those categories. It is the main result of a phenomenographic study and shows the range of understandings rather than a single shared essence.

Peer debriefing

Also called: peer debrief, peer debriefer

Peer debriefing is the discussion of a study's data, emerging analysis and findings with one or more knowledgeable colleagues, who ask questions and challenge interpretations to refine the researcher's thinking. It is one of Lincoln and Guba's strategies for establishing credibility, and it can take place at several stages of the analysis.

Persistent observation

Persistent observation is sustained, focused attention on the features of a setting or data set most relevant to the research question, examining them in depth rather than taking in everything superficially. It is one of Lincoln and Guba's strategies for credibility and complements prolonged engagement.

Phenomenography

Also called: phenomenographic research

Phenomenography is a qualitative research approach, developed by Marton and colleagues in Sweden in the 1970s, that describes the qualitatively different ways in which a group of people experience or understand the same phenomenon. Unlike phenomenology, which looks for the shared essence of an experience, it maps the variation between understandings in what it calls an outcome space.

Phenomenological reduction

Also called: transcendental phenomenological reduction

Phenomenological reduction, in Husserl's descriptive phenomenology, is the series of steps by which the researcher moves from the everyday, taken-for-granted attitude to a description of the essence of an experience. It begins with epoché, or bracketing, continues with a full description of each participant's experience, and ends with imaginative variation to distil the essences.

Phenomenology

Also called: phenomenological research, phenomenological approach, phenomenological study, phenomenological design

Phenomenology is a qualitative approach, rooted in the philosophy of Husserl and Heidegger, that studies lived experience: what a phenomenon is like for those who experience it and what it means to them. Its two main branches are descriptive phenomenology, which seeks the essence of an experience, and interpretive phenomenology, which holds that experience can only ever be interpreted.

Photovoice

Also called: photo voice, photo-voice

Photovoice is a participatory research method in which people photograph their own lives and community, discuss the pictures in groups to identify strengths and concerns, and present the results to decision-makers. Described by Wang and Burris in 1997, it aims to help people record their community, prompt critical dialogue and reach policymakers.

Positionality

Also called: researcher positionality

Positionality is the researcher's position in relation to a study: how their social identity, background, professional role and personal experience shape the questions they ask, their relationships with participants and the way they interpret data. Recognising it openly is part of reflexivity, and many qualitative papers now include a positionality statement.

Positionality statement

Also called: researcher positionality statement

A positionality statement is a short passage in which researchers describe the identities, experiences and assumptions they bring to a study and how these may have shaped it. It gives readers what they need to judge the researcher's influence, and it can be written for each author separately or for the research team as a whole.

Post-intentional phenomenology

Post-intentional phenomenology is a contemporary approach to phenomenology, associated with Mark Vagle, that treats a phenomenon as multiple, partial, contextual and always in flux, both produced by the world and producing it, rather than as having one fixed essence. It is one of several newer approaches that bridge the descriptive and interpretive traditions.

Progressive subjectivity

Progressive subjectivity is a credibility strategy in which researchers state their assumptions and expectations and keep track of how their understanding shifts as the study proceeds, so that they can see how their own views may be shaping the interpretation. It is one of the techniques associated with Guba and Lincoln's criteria for constructivist inquiry.

Prolonged engagement

Prolonged engagement is spending enough time in a setting or with participants to understand the context, build trust and rapport, and test first impressions and misinformation. Lincoln and Guba list it as a strategy for credibility. It is central to ethnography and hard to achieve in short studies such as rapid or focused ethnographies.

Qualitative content analysis

Qualitative content analysis is a method for systematically describing the meaning of text or other material by coding it and grouping the codes into categories, and sometimes themes. It developed from quantitative content analysis, which counts features of texts. Hsieh and Shannon describe conventional, directed and summative forms, and others distinguish manifest from latent analysis.

Qualitative data analysis

Also called: QDA, qualitative analysis, analysis of qualitative data

Qualitative data analysis is the process of making sense of non-numerical data, such as transcripts, field notes, documents and images, by organising, coding, comparing and interpreting it to answer a research question. It usually runs alongside data collection rather than after it, and its form depends on the approach chosen, from thematic analysis to grounded theory.

Qualitative description

Also called: qualitative descriptive design, qualitative descriptive study, descriptive qualitative research, qualitative descriptive approach

Qualitative description is a qualitative design, described by Sandelowski in 2000, that aims for a rich, straightforward account of an experience or event in everyday language, staying close to the data rather than building theory or seeking an essence. It usually uses content or thematic analysis and is common in nursing and health research.

Qualitative research

Also called: qualitative methods, qualitative inquiry, qualitative study, qualitative methodology, qualitative approach

Qualitative research is research that studies the nature and meaning of phenomena through non-numerical data, such as interviews, observation, documents and images, to understand how and why people experience and act as they do. It is flexible and iterative, moving back and forth between data collection and analysis, and it treats the researcher's interpretation as part of the method.

Rapid qualitative analysis

Also called: rapid analysis, RQA, Hamilton's rapid qualitative analysis

Rapid qualitative analysis is a streamlined approach, used especially in health services and implementation research, in which each interview is summarised by domain in a structured template and the summaries are gathered into a matrix to compare participants, instead of full line-by-line coding. It is designed to produce findings faster than full coding allows.

Referential adequacy

Also called: referential adequacy materials

Referential adequacy is a credibility strategy in which researchers keep a separate collection of materials from the field, such as documents and field notes, against which interpretations can be checked during and after analysis. It is one of the strategies Lincoln and Guba proposed for establishing credibility in qualitative research.

Reflexive journal

Also called: reflexive diary, research diary, researcher diary, reflective journal

A reflexive journal is a diary a qualitative researcher keeps throughout a study to record their reactions, assumptions, decisions and thoughts about the data and about their own influence on it. It supports reflexivity and confirmability, feeds the audit trail and gives readers evidence of how the researcher's perspective shaped the work.

Reflexive thematic analysis

Also called: reflexive TA, RTA, Braun and Clarke's thematic analysis

Reflexive thematic analysis is Braun and Clarke's approach to thematic analysis, in which themes are patterns of shared meaning that the researcher actively creates through six recursive phases, from familiarisation and coding to writing up. It treats the researcher's subjectivity as a resource, so it avoids fixed codebooks and rejects intercoder reliability, saturation and talk of themes emerging.

Reflexivity

Also called: researcher reflexivity, reflexive practice

Reflexivity is the researcher's continuing critical reflection on how their own background, assumptions, preferences and relationships with participants shape every stage of a study, from the questions asked to the interpretation of data. In qualitative research it is treated as essential to transparency and quality, and it is usually recorded in a reflexive journal and reported.

Researcher as instrument

Also called: human instrument, researcher as research instrument

Researcher as instrument is the idea that in qualitative research the researcher, not a questionnaire or scale, is the main tool of data collection and analysis, so their skills, knowledge and perspective directly affect the quality of the findings. It is why credibility depends partly on the researcher's familiarity with the setting and experience of the methods.

Saturation

Also called: qualitative saturation, saturation point

Saturation is the point in a qualitative study at which more data or more analysis would add nothing new, often used to decide when to stop collecting data. The idea began in grounded theory but now has several senses, among them theoretical, data, code and meaning saturation, and some methodologists argue that it does not suit every approach.

Selective coding

Also called: theoretical integration, advanced coding

Selective coding is the final stage of coding in Straussian grounded theory, in which the researcher chooses a core category and relates all the other categories to it to form the theory. In classic grounded theory the term instead means coding only the data that relate to the core category once it has been found. Later Straussian texts call the stage theoretical integration.

Semantic coding

Also called: semantic code, semantic codes

Semantic coding is coding that captures the explicit, surface meaning of what participants say, staying close to their own words. In Braun and Clarke's thematic analysis it sits at one end of a continuum with latent coding and tends to support descriptive or realist accounts of experience, though neither kind of coding is better in itself.

Sensitising concept

Also called: sensitizing concept, sensitising concepts, sensitizing concepts

A sensitising concept is a broad idea, often drawn from theory, that gives a qualitative researcher a starting point and a sense of where to look in the data without fixing in advance what will be found. Unlike an a priori code, it guides attention rather than being applied as a label, and grounded theorists often use one to begin analysis.

Situational analysis

Situational analysis is an extension of grounded theory developed by Adele Clarke, a student of Anselm Strauss, that takes a whole situation, including human and non-human elements, discourses and social worlds, as the focus of analysis. It works through three kinds of map, situational maps, social worlds and arenas maps, and positional maps, to show the messiness and variety of a field.

Straussian grounded theory

Also called: Strauss and Corbin grounded theory, evolved grounded theory, Straussian GT

Straussian grounded theory is the version of grounded theory set out by Anselm Strauss and Juliet Corbin in Basics of Qualitative Research (1990), a book that led Glaser to a sharp public critique. It offers detailed procedures, open, axial and selective coding with tools such as the coding paradigm, and rests on pragmatism and symbolic interactionism.

Substantive coding

Substantive coding is the first phase of analysis in classic grounded theory, made up of open coding and then selective coding, in which the researcher builds categories and their properties from the data and identifies the core category. It is followed by theoretical coding, which links the categories into a theory.

Summative content analysis

Summative content analysis is the approach to qualitative content analysis that begins by identifying and counting keywords or content in texts, then interprets the context in which they are used. It is one of the three approaches Hsieh and Shannon described and the one that stays closest to quantitative content analysis.

Symbolic interactionism

Also called: symbolic interactionist perspective

Symbolic interactionism is a sociological perspective holding that people act on the meanings things have for them, and that those meanings, and the self, are formed and changed through interaction with others. Herbert Blumer developed it from the ideas of G. H. Mead, and it underpins grounded theory, particularly the Straussian version.

Tactical authenticity

Tactical authenticity is the authenticity criterion asking whether a qualitative study left participants empowered and able to act, with some influence over what is done as a result. It is one of the five authenticity criteria of constructivist inquiry, and it goes a step beyond catalytic authenticity, which asks whether the study prompted a wish for change.

Template analysis

Also called: template approach

Template analysis is a form of thematic analysis, associated especially with Nigel King at the University of Huddersfield, in which the researcher builds a hierarchical coding template, often starting from a few a priori themes, and revises it as it is applied to the data. It is flexible about epistemology and is one of the codebook approaches to thematic analysis.

Thematic analysis

Also called: TA, thematic analyses, theme analysis

Thematic analysis is a widely used method for developing, analysing and reporting patterns of meaning, called themes, across a qualitative data set. It is a method rather than a full methodology, so it can serve many theories and research questions. Braun and Clarke distinguish coding reliability, codebook and reflexive forms, which rest on different assumptions.

Theme

Also called: themes, qualitative theme

A theme is a pattern of shared meaning across a qualitative data set, organised around a central concept or idea, that captures something important about the research question. Braun and Clarke distinguish it from a topic summary, which only collects what was said about a subject. Themes are usually built from codes and categories and are more interpretive than either.

Theoretical coding

Also called: theoretical codes

Theoretical coding is the final stage of coding in classic grounded theory, in which the researcher works out how the substantive categories relate to one another and weaves them into an integrated theory, often with the help of Glaser's theoretical coding families. Constructivist grounded theorists use such theoretical codes only if they earn their place in the analysis.

Theoretical saturation

Theoretical saturation is the point in grounded theory at which new data no longer add properties to the categories of the developing theory. It is the original sense of saturation, from Glaser and Strauss, and is tied to theoretical sampling: the researcher keeps sampling to develop categories until they are fully described, not simply until data repeat.

Theoretical sensitivity

Theoretical sensitivity is a grounded theorist's ability to recognise which parts of the data matter for the developing theory and to see their meaning and connections. It grows through immersion in the data, coding, memo writing and reading, and Glaser devoted a 1978 book to it. Straussian grounded theory holds that the researcher's own experience can increase it.

Theoretical sufficiency

Also called: sufficiency

Theoretical sufficiency is the term Dey (1999) suggested in place of theoretical saturation, on the view that a researcher can judge categories to be developed enough for the purpose but can never know they are complete. It reflects wider doubts about saturation, which Saunders and colleagues describe as a prediction about unobserved data based on what has been observed.

Theory triangulation

Also called: theoretical triangulation

Theory triangulation is the use of more than one theoretical perspective to interpret the same qualitative data, to see what each brings out and to avoid reading the data through a single lens. It is one of the kinds of triangulation used to strengthen the credibility and confirmability of qualitative findings.

Thick description

Thick description is a detailed account of behaviour or experience together with its context and meaning, rich enough that an outsider can understand why things happened as they did. The term comes from the anthropologist Clifford Geertz. In qualitative research it also serves transferability, since it lets readers judge whether findings apply to their own setting.

Three-dimensional narrative inquiry space

Also called: three-dimensional space, narrative commonplaces, three commonplaces of narrative inquiry

The three-dimensional narrative inquiry space is Clandinin and Connelly's framework for narrative inquiry, in which every experience is examined along three dimensions: temporality, sociality, meaning personal and social relationships, and place. Researchers attend to all three at once when composing field texts from stories and turning them into research texts.

Topic summary

A topic summary, in Braun and Clarke's terms, is a so-called theme that only summarises what participants said about a subject, such as the benefits of an intervention, without a central idea that unites it. They warn that topic summaries are often presented as themes, which should instead capture a pattern of shared meaning.

Transferability

Transferability is the trustworthiness criterion concerned with applicability: how far a qualitative study's findings may apply in other settings or with other people. The researcher supports it with thick description of participants, context and methods, but the judgement about transfer belongs to the reader. It is Lincoln and Guba's counterpart to external validity.

Triangulation

Also called: triangulate, triangulated

Triangulation, in qualitative research, is the use of more than one data source, method, investigator or theory to study the same phenomenon, so that the strengths of each offset the weaknesses of the others. Its main kinds are data, investigator, methodological and theory triangulation, and it is used to strengthen the credibility and confirmability of findings.

Trustworthiness

Also called: trustworthiness criteria, Lincoln and Guba criteria, four-dimension criteria

Trustworthiness is the overall quality of a qualitative study, meaning how far its findings can be trusted. Lincoln and Guba proposed it in 1985 in place of validity and reliability, with four criteria, credibility, transferability, dependability and confirmability, each supported by particular strategies. Authenticity criteria were added later for constructivist research.

Turn-taking

Also called: turn taking, turn-taking system

Turn-taking, in conversation analysis, is the organisation by which people in talk coordinate who speaks next and when, so that gaps and overlaps stay short. Sacks, Schegloff and Jefferson described it as a set of practices through which listeners anticipate where a turn may end, the point at which a change of speaker becomes relevant.