Examples
Worked examples
- Is an instance
A study of how early-career researchers navigate first authorship disputes, using constant comparison across interviews as they're collected and theoretical sampling toward disconfirming cases, stopping once new interviews stop adding new properties -- producing a substantive theory of dispute navigation as a staged, trust-dependent process.
- Is an instance
A Straussian grounded theory study of staff adapting to a new grants-management system: open coding, axial coding relating categories as conditions-actions-consequences, theoretical sampling across both adopters and resistant staff, and selective coding integrating the categories around a core category explaining the adaptation process.
Counter-examples
Looks similar, but isn't
- Not an instance
A study that starts from an existing framework (e.g., the Technology Acceptance Model), codes interview transcripts against that framework's pre-specified constructs, and recruits a fixed sample decided in advance for demographic coverage. This is framework-driven qualitative content analysis or an instrumental case study, not grounded theory -- the theory is imposed before analysis rather than generated from the data, and sampling is representativeness-driven rather than theoretical.
Editorial commentary
What makes a study grounded theory
Grounded theory is a qualitative research methodology in which a theory is built inductively from data, rather than proposed in advance and then tested against it. A study counts as grounded theory only when three things hold together across the whole design, not just in one part of it: simultaneous data collection and analysis — the researcher analyzes early data while still collecting more, rather than gathering a full dataset and analyzing it afterward; the constant comparative method — every new piece of data is systematically compared against the codes and categories built from data already analyzed, so categories are continuously refined rather than fixed at the outset; and theoretical sampling directed toward saturation — who or what to sample next is decided by what the emerging categories need (a case likely to extend, test, or challenge a category), not by a sampling plan fixed before analysis began, and data collection continues until new data stops adding new properties to the core categories. The output is a substantive theory: a set of interrelated concepts, grounded in and traceable back to the data, that explains a process or pattern of social action — not a description of a setting, a summary of themes, or a test of an existing theory’s fit.
Grounded theory is one of several established qualitative traditions — alongside phenomenology, ethnography, case study, and narrative research (Creswell & Poth’s widely cited five-tradition framework) — and it is the tradition specifically suited to research questions about how a process unfolds or how people construct a course of action over time, as distinct from questions about lived experience, a bounded case, or a cultural group. See Qualitative Research for the four criteria that make a study qualitative at all, and Research Study Types for how grounded theory sits alongside the wider taxonomy of study designs.
Origins: Glaser and Strauss (1967)
Grounded theory was introduced by sociologists Barney Glaser and Anselm Strauss in The Discovery of Grounded Theory: Strategies for Qualitative Research (Sociology Press, 1967). Their argument was methodological and, at the time, a genuine departure from prevailing practice: sociology had come to treat theory generation and theory verification as separate jobs, with qualitative fieldwork relegated mostly to illustrating theories developed elsewhere. Glaser and Strauss proposed a systematic, replicable procedure for generating theory directly from data, giving qualitative research the same claim to methodological rigor that quantitative hypothesis-testing had, without borrowing quantitative logic to do it. The book’s procedures — constant comparison, theoretical sampling, and saturation — are the same core mechanics still used today, though how strictly they are applied, and what they are used for, split into different schools not long after (see below).
The core mechanics
Constant comparative method
The constant comparative method is the analytic engine of grounded theory: each new incident, interview segment, or document is compared against previously coded data, category by category, rather than coded once and set aside. Glaser’s original four-step description of the process is still the clearest summary: (1) comparing incidents applicable to each category as they are coded, (2) integrating categories and their properties as comparison continues, (3) delimiting the emerging theory as categories stabilize, and (4) writing the theory once delimitation is substantially complete. This is what distinguishes grounded theory coding from a one-pass thematic coding exercise — the comparison is continuous and drives what gets sampled next, not just how the existing data gets organized.
Theoretical sampling
Theoretical sampling is the sampling logic that follows directly from constant comparison: instead of drawing a sample sized and structured in advance for demographic representativeness, the researcher selects the next data source specifically because it is likely to test, extend, or challenge a category that is still underdeveloped. A category that looks solid based on interviews with one kind of participant might be deliberately checked against a participant expected to be a poor fit for it. This is a defining, not incidental, feature — a study that recruits its full sample before analysis begins, on criteria fixed in advance, is not using theoretical sampling even if the resulting analysis is otherwise inductive.
Theoretical saturation
Theoretical saturation is the stopping rule: data collection continues until new data no longer adds new properties, dimensions, or relationships to the core categories. Saturation is a property of the categories, assessed through ongoing analysis, not a sample-size target decided in advance — two grounded theory studies of similar scope can reach saturation at meaningfully different sample sizes depending on how much variation the categories need to account for.
Two worked examples
Example 1: A straightforward inductive study
A researcher wants to understand how early-career researchers navigate a first authorship dispute. Interviews begin with a handful of postdocs; each transcript is open-coded and compared against the codes from every prior transcript as it comes in, rather than batched for analysis at the end. Early codes point to a recurring pattern the researcher labels “testing trust before escalating.” Theoretical sampling then deliberately seeks out participants likely to challenge that emerging category — someone who escalated a dispute immediately, and someone from a discipline with an explicit authorship policy already in place — rather than recruiting more of the same kind of interview. Data collection stops once new interviews stop adding new properties to the core category. The output is a substantive theory of authorship-dispute navigation as a staged, trust-dependent process, built from and traceable to the interview data.
Example 2: A Straussian study with axial coding
A research-administration team studies how staff adapt to a newly implemented grants-management system. Following Strauss and Corbin’s procedure (see below), the researcher open-codes early interviews into discrete concepts, then uses axial coding to relate those concepts to one another as conditions, actions, and consequences, and uses theoretical sampling to deliberately interview both enthusiastic early adopters and staff who resisted the change, continuing until no new relationships among categories emerge. Selective coding then integrates the categories around a core category — “renegotiating expertise” — that explains the adaptation process across both groups.
Counter-example: when a qualitative study isn’t grounded theory
A researcher starts from an existing framework — say, the Technology Acceptance Model — interviews staff about a new system, and codes each transcript against the model’s pre-specified constructs (perceived usefulness, perceived ease of use) to see whether the data fits, recruiting a fixed sample decided in advance for demographic coverage rather than by what the emerging analysis needs. This may be a perfectly good qualitative study, but it is not grounded theory: the theoretical framework is imposed before analysis rather than generated from it, sampling is representativeness-driven rather than theoretical, and there is no constant-comparison-driven emergence of a new, data-grounded theory. This is closer to framework-driven qualitative content analysis, or an instrumental case study — a researcher who labels it “grounded theory” anyway because the raw data is interview transcripts is describing the data source, not the analytic method.
Three schools of grounded theory
Glaser and Strauss’s original collaboration split into distinct methodological schools, and a manuscript or thesis committee will often expect the author to name which one is being used, since the schools differ on real procedural points, not just terminology:
- Glaserian (classic) grounded theory — Glaser’s later, solo-authored work insisted on staying close to the original 1967 procedure: theory should emerge from the data with minimal preconception, and Glaser was critical of more structured coding procedures as forcing data into a preconceived framework rather than letting categories emerge.
- Straussian grounded theory — Strauss and Juliet Corbin’s Basics of Qualitative Research (1990; later editions 1998, 2008, 2015) introduced a more explicit, structured coding sequence: open coding (identifying concepts), axial coding (relating categories to one another via a conditions-actions-consequences paradigm), and selective coding (integrating categories around a core category). This is the most widely taught version of grounded theory coding today, in part because its procedure is easier to teach and to demonstrate rigor with in a methods section.
- Constructivist grounded theory — Kathy Charmaz’s Constructing Grounded Theory (2006; 2nd ed. 2014) keeps the core mechanics — coding, memo-writing, theoretical sampling — but rejects the idea that theory is neutrally “discovered” in the data. Charmaz frames the researcher as a co-constructor of meaning alongside participants, and the resulting theory as an interpretation shaped by that relationship, not a fully objective account.
When to use grounded theory
Grounded theory fits a research question that asks how a process unfolds, how people construct or negotiate a course of action, or how a social pattern develops over time, in an area where existing theory is thin, contested, or a poor fit for the setting being studied. It is a weaker fit when the goal is to describe a bounded case in depth (case study), to capture the structure of a lived experience (phenomenology), to describe the shared practices of a cultural group (ethnography), or to test an existing theory or model against new data — grounded theory is specifically for generating theory, not for confirming or elaborating one that already exists going into the study. Because theoretical sampling and saturation-based stopping rules don’t fit a sample size fixed in a proposal in advance, grounded theory also requires a study design, timeline, and (for human-subjects research) an IRB protocol that can accommodate an iterative, not fully pre-specified, sampling plan.
Why the distinction matters for research administration
Getting the label right affects three concrete things a research office or thesis committee checks: whether the methods section actually does what it claims (a paper that calls itself grounded theory but shows a fixed sample and a single coding pass at the end reads, to a reviewer familiar with the method, as a mislabeled thematic analysis); which reporting checklist applies (the Standards for Reporting Qualitative Research, SRQR, is the checklist most directly applicable across qualitative designs including grounded theory, and journals increasingly ask authors to reference it); and how a human-subjects protocol should describe recruitment, since theoretical sampling means the final sample composition and size genuinely cannot be fully specified before data collection begins, which an IRB reviewer needs to understand rather than read as a vague or incomplete protocol. See How to Write the Methodology Section of a Qualitative Research Paper for how to report a grounded theory design, and Qualitative Research vs. Quantitative Research for how saturation-based sampling logic differs from statistical power calculations.
Related CASRAI terms
- Qualitative Research — the broader approach grounded theory is one tradition within.
- Mixed Methods Research — how a grounded theory strand can be combined with a quantitative strand in a single study.
- Research Study Types — where grounded theory sits within the wider taxonomy of study designs.
- Research Approach: Definition, Types, and How to Choose One — the broader plan-level decision (quantitative, qualitative, or mixed methods) that a choice of grounded theory sits underneath.
References
- Glaser, B. G., & Strauss, A. L. The Discovery of Grounded Theory: Strategies for Qualitative Research. Sociology Press, 1967.
- Strauss, A., & Corbin, J. Basics of Qualitative Research: Grounded Theory Procedures and Techniques. SAGE, 1990 (later editions 1998, 2008, 2015).
- Charmaz, K. Constructing Grounded Theory. SAGE, 2006 (2nd ed. 2014).
- Creswell, J. W., & Poth, C. N. Qualitative Inquiry and Research Design: Choosing Among Five Approaches (4th ed., 2018).
- O’Brien, B. C., et al. Standards for Reporting Qualitative Research (SRQR), endorsed by the EQUATOR Network (equator-network.org) — applies across qualitative designs including grounded theory.
Also known as
Grounded theory research · Grounded theory methodology · Grounded theory approach
Machine-readable encodings
Use in your systems
<role vocab="credit"
vocab-identifier="https://casrai.org/dictionary/"
vocab-term="Grounded Theory"
vocab-term-identifier="https://casrai.org/dictionary/term/grounded-theory" />{
"@context": "https://schema.org",
"@type": "DefinedTerm",
"@id": "https://casrai.org/dictionary/term/grounded-theory",
"name": "Grounded Theory",
"identifier": "https://casrai.org/dictionary/term/grounded-theory",
"description": "Grounded theory is a qualitative research methodology in which a theory is generated inductively from data, rather than proposed in advance and tested against it. A study counts as grounded theory only when three things hold together: data collection and analysis proceed simultaneously and iteratively, not sequentially; codes and categories are developed and refined through the constant comparative method, systematically comparing each new piece of data against categories built from earlier data; and further data collection is directed by theoretical sampling -- deliberately seeking cases likely to test, extend, or challenge the emerging categories -- continuing until theoretical saturation, the point at which new data stops adding new properties to the core categories. The output is a substantive theory: interrelated concepts explaining a process or pattern of social action, grounded in and traceable back to the data itself.",
"inDefinedTermSet": "https://casrai.org/dictionary/domain/research-outputs#set",
"url": "https://casrai.org/dictionary/term/grounded-theory",
"sameAs": [
"Grounded theory research",
"Grounded theory methodology",
"Grounded theory approach"
],
"license": "https://creativecommons.org/licenses/by/4.0/",
"publisher": {
"@id": "https://casrai.org/#organization"
},
"dateModified": "2026-07-23T16:25:55",
"inLanguage": "en"
}
