Grounded theory is a method for generating theory out of data rather than for testing theory against data. Its object is a set of concepts and propositions that are induced from systematic fieldwork and remain traceable to it — a theory that is, in the founders' word, grounded. It was set out by Barney Glaser and Anselm Strauss in The Discovery of Grounded Theory (1967), written out of their study of hospital care of the dying, and directed against the sociology of their day, in which grand theories of the Parsonian kind were elaborated with little empirical contact while quantitative research busied itself verifying hypotheses derived from them.
The reversal is fundamental. In the hypothetico-deductive design, concepts are defined, hypotheses stated, instruments built and data collected to test them. In grounded theory the researcher begins with a sensitising area of enquiry rather than a hypothesis, and lets the analytic problem emerge from the field. The resulting theory is judged not by its statistical significance but by four criteria the founders proposed: fit with the data, relevance to the participants' concerns, workability in explaining what happens, and modifiability as new data arrive.
The procedures
The distinctive procedural rule is that collection and analysis proceed simultaneously. Data are coded as soon as they are gathered, and the codes direct the next round of collection.
Coding moves through levels. Open coding breaks the transcript into small segments and labels each with a concept, often in the participants' own words — so-called in vivo codes — generating a large and deliberately untidy set of categories. Axial coding then relates categories to one another, specifying conditions, actions, interactions and consequences, and collapsing codes into higher-order categories with their properties and dimensions. Selective coding identifies a core category around which the others are organised, and states the theory as a set of relationships centred on it.
Constant comparison runs throughout: each new incident is compared with those already coded, incident with incident and then incident with category, which forces the analyst to specify what a category includes and excludes. Memo writing is the running record of analytic thought — why a code was created, how two categories connect, what to look for next — and is the medium in which the theory actually takes shape. Theoretical sampling replaces the statistical logic of sample selection: further cases are chosen because they promise to develop or challenge an emerging category, so the sample cannot be fixed in advance. Collection stops at theoretical saturation, when additional cases yield no new properties of the core categories. The founders also distinguished substantive theory, applying to one empirical area such as dying in hospitals, from formal theory, applying to a general process such as awareness contexts or status passage.
The suspension of prior literature
The most contested instruction is that the researcher should delay the literature review, entering the field without a settled framework so that the data are not forced into borrowed categories. Glaser was the more insistent on this, and the criticism it attracts is direct: no trained sociologist can hold a blank mind, since the very act of noticing presupposes concepts, and a researcher ignorant of existing work risks rediscovering the obvious or missing the significance of what is found. The defensible version is not ignorance but suspension — knowing the literature while deliberately withholding commitment to its categories until the data have been allowed to suggest their own, and treating existing concepts as sources of sensitising questions rather than as a coding frame.
The divergence and constructivist grounded theory
Glaser and Strauss later split publicly. Strauss and Corbin's Basics of Qualitative Research (1990) introduced a prescribed coding paradigm and conditional matrix that fixed the questions to be asked of every category. Glaser replied in Emergence vs Forcing (1992) that these devices imposed a template on the data and produced full conceptual description rather than emergent theory, abandoning the original logic. The dispute matters because it exposes a real tension between the method's promise of emergence and the demand for teachable, auditable procedure.
Kathy Charmaz offered a third position. Constructivist grounded theory retains the procedures — coding, comparison, memos, theoretical sampling — but rejects the founders' residual objectivism. Categories do not emerge from data as though the researcher were absent; they are constructed in the interaction between researcher and field, and the theory bears the marks of the researcher's standpoint. Charmaz therefore makes reflexivity part of the method and treats grounded theory as a flexible set of guidelines rather than a set of rules, which is the version most widely used today.
Application and appraisal
The strategy travels well to any setting where existing theory is thin and process matters. Indian work has used it in medical sociology and public health — patient and caregiver careers, help-seeking pathways, stigma around tuberculosis, HIV or mental illness — in studies of women's self-help groups, and in research on migrant labour, where the object is to explain how people manage an unfolding situation rather than to measure its incidence. Its virtues are that concepts fit the world they describe, that participants' concerns drive the analysis, and that it produces theory where none existed.
The criticisms are serious. The blank-mind premise is untenable. The procedures can harden into a mechanical ritual of coding that generates categories without producing a theory, and the volume of coding is severe. Samples are small and purposive, so any claim beyond the substantive area is contestable, and saturation is a judgement rather than a measurable point. The requirement that theory be induced also sits awkwardly with the recognition that the analyst's prior concepts are always at work — the very point Charmaz concedes.
For the UPSC answer
State the core inversion first: grounded theory generates theory from data instead of testing pre-formed hypotheses, and Glaser and Strauss advanced it in 1967 against both ungrounded grand theory and hypothesis-verifying empiricism. Then give the procedural chain in order — simultaneous collection and analysis, open, axial and selective coding, constant comparison, memos, theoretical sampling, saturation — because examiners reward the named sequence. Use the literature-review controversy and the Glaser–Strauss split to show the method's internal tension between emergence and prescribed procedure, and close with Charmaz's constructivist version, which concedes that categories are constructed and thereby answers the blank-mind objection.
References & further reading
- Glaser, B. G. and Strauss, A. L. (1967). The Discovery of Grounded Theory: Strategies for Qualitative Research. Aldine.
- Strauss, A. L. and Corbin, J. (1990). Basics of Qualitative Research: Grounded Theory Procedures and Techniques. Sage.
- Glaser, B. G. (1992). Basics of Grounded Theory Analysis: Emergence vs Forcing. Sociology Press.
- Charmaz, K. (2006). Constructing Grounded Theory: A Practical Guide through Qualitative Analysis. Sage.
- Glaser, B. G. and Strauss, A. L. (1965). Awareness of Dying. Aldine.
- Bryman, A. (2012). Social Research Methods. Oxford University Press.