修行会

A most useful tool is the scientific method, and what better person, to re-illuminate it, than Barney Glaser. This guide is condensed from Glaser’s books from Sociology Press (and others). I am forever grateful and indebted to his commitment to the craft of science.

“Description writing comes naturally, conception writing seldom does and must be learned with discipline.”

Barney Glaser (2012, p. 70)

Grounded Theory: A Practical Guide to Building Theory from Data

Grounded Theory (GT; Glaser & Strauss, 1967; Glaser, 1978) is the study of a concept that names a pattern as a core category with general implications ”every which way you look” (Glaser, 2010). GT, within the context of the sciences, is a qualitative method for generating theory from data through constant comparison, coding, demoing, and theoretical sampling, culminating in an integrated, modifiable theory that explains participants’ main concern.

Compare incidents with incidents, then incidents with codes, then codes with codes to raise abstraction and refine properties. This iterative comparing is the engine of GT.

Why Use Grounded Theory?

Use GT when existing theories don’t explain a process you’re studying, and you need a data-grounded, portable explanation that can travel beyond one site. Quality is judged by fit, work, relevance, and modifiability (often with “grab” which is the theory’s immediate resonance).

The GT Cycle at a Glance

• Collect data (e.g., interviews, observations, documents).
• Open code line-by-line (name what‘s happening). Capture in vivo (participants‘ own words) where they carry analytic punch.
• Compare constantly (incident to incident, incident to code, code to code).
• Write memos early and often (mini-essays that push concepts and relationships in any way possible in your own language while avoiding the descriptive and sticking to concepts).
• See a core category emerging? Shift to selective coding (code only what relates to the core) and theoretical sampling (seek data that develop categories).
• Sort memos to integrate using theoretical codes (weave categories into an explanatory model).
• Stop when saturated (no new properties/relationships for core categories. Write up!

Key Methods and How to Do Them Well

Data Collection

• Set aside preconceived ideas (Glaser & Strauss, 1967).
• Use broad inquiries to take the data as accurate as possible without modification (Glaser, 2001b). Interview guides while in-depth, are forced and imprecise (Charmaz, 2000; Glaser, 2002).
• Consider using a “grand tour question” like, ”to what do I owe the honor of a visit?” (Linden & Simmons, 2025).
• Very passive listening (Glaser, 2002)
• Consider Max Weber‘s vertehen, “whereby the investigator understands a group‘s behavior by viewing their action through their eyes” (Glaser, 1996).

Coding

• Substantive Coding
  • Open Coding (first pass): conceptualize line-by-line; allow in-vivo codes.
  • Selective Coding: once the core category emerges, code what relates.
• Theoretical Coding (if used): integrate categories (e.g., process, cause-consequence, context-contingency, degree-dimension, strategy). Use them as emergent linkages, not templates to force data.

Constant Comparison

Memoing & Memo Sorting

Memoing is the core analytic work capturing conceptual ideas and categorical relationships as they emerge. Memos have no right way to be written (Glaser, 2014). Later, sort memos. Just sort them. More memos and writups will emerge. This will automatically assemble the integrated grounded theory toward final writeup and provide outline components. Avoid excessive “story talking”. Conceptualizations are not connected to a specific location, time or people (Glaser, 2012). Stop. Write!

Theoretical Sampling

After initial coding, sample to develop the theory. Seek cases, people, times, or sites that elaborate, contrast, or challenge the categories and their conditions/consequences until saturation.

Saturation (What It Is / Isn‘t)

Stop when additional data (i.e., indicators) add no new properties of the core categories and relationships are well accounted for, not when hitting some arbitrary quota.

Where Approaches Differ

Use the approach that fits the philosophical stance and institutional requirements; keep GT’s comparative, emergent core intact. Use one method and stick with it otherwise experience procedural disorder (Glaser, 2005).

• Classic/Glasserian GT: Emphasizes emergence over forcing; substantive (open to selective coding) then theoretical coding; postpone deep literature review in the substantive area; heavy memoing/sorting.
• Straussian GT: Open, axial, selective coding; diagramming often used; more explicit procedures. Note, that this is not GT (Chametzky, 2025).
• Constructivist GT (Charmaz & Thornberg, 2021): Initial/focused/theoretical coding; co-construction and reflexivity emphasized; literature engagement is more flexible. Note, that this is not GT (Chametzky, 2025).

Literature: Read When and How?

Classic GT recommends delaying deep, topic-specific reviewing to avoid preconception, but many contexts require initial review. A pragmatic resolution: do broad, sensitizing reading early; brakes it in memos; conduct focused engagement later during sorting to position and extend theory.

Evaluate GT: The 4+1 Criteria (Glaser, 1978)

Use these criterion rather than traditional validity tests for qualitative theory building.

• Fit: concepts match incidents.
• Work: explains how participants resolve their main concern.
• Relevance: addresses what matters to the field/participants.
• Modifiability: updatable with new data.
• Grab: The theory‘s intuitive, immediate pull.

Common Derailments (and fixes)

• Over-talking instead of memoing; stop, write when ready; channel ideas into memos and sorting.
• Over-coding/indicator chasing; focus on properties that earn their relevance; don‘t fragment endlessly.
• Forcing frameworks/diagrams; use theoretical codes and diagrams emergently; don‘t retrofit data to a model.
• Staying descriptive; push to conceptualize and relate categories; reduce reliance on illustrative anecdotes.
• Premature literature forcing; keep early reading sensitizing, revisit literature after core integration.

Writing a Grounded-Theory Paper (A Lean Template)

• Title: Signal the core category; aim for grab.
• Abstract: State the main concern, core category, and how the theory works.
• Introduction: What problem does this theory explain? Keep it conceptual.
• Methods (brief, but clear): Data sources; constant comparison; memoing/sorting; theoretical sampling; saturation indicator.
• Results/Theory: Present the core category, its properties, variation conditions, relationships to other categories; integrate via theoretical codes; include modifiability note(s).
• Discussion: General implications, scope conditions; how it fits/works; relate to selective literature (as more data).
• Conclusion: What the theory explains and how it can be modified as contexts change.

References

Breckenridge, J. (2009). Demystifying theoretical sampling in Grounded Theory Research. Grounded Theory Review 8(2), 113–126.

Chametzky, B. (2025). When is Grounded Theory (GT) not Grounded Theory: Methodological convergences and divergences. Grounded Theory Review, 24(1), pp. 25–47.

Charmaz, K. (2000). Grounded Theory: Objectivist and constructivist methods. In N. K. Denzin & Y. S. Lincoln (Eds.), Handbook for qualitative research 2nd ed., Sage.

Charmaz, K., & Thornberg, R. (2021). The pursuit of quality in grounded theory. Qualitative Research in Psychology, 18(3), 305–327.

Chun Tie, Y., Birks, M., & Francis, K. (2019). Grounded theory research: A design framework for novice researchers. SAGE open medicine, 7, 2050312118822927. https://doi.org/10.1177/2050312118822927

Glaser, B. (2014). Memoing: A vital grounded theory procedure. Sociology Press.

Glaser, B. (2012). Stop. Write! Writing grounded theory. Sociology Press.

Glaser, B. (2010). Grounded Theory is the study of a concept! YouTube. https://www.youtube.com/watch?v=OcpxaLQDnLk

Glaser, B. (2005). The grounded theory perspective III: Theoretical coding. Sociology Press.

Glaser, B. (2004). Remodeling Grounded Theory. Forum Qualitative Sozialforschung Social Research 5(2), Art. 4. Retrieved from https://sites.ualberta.ca/~iiqm/backissues/1_2Final/html/glaser.html

Glaser, B. (2002). Constructivist Grounded Theory? Forum Qualitative Social Research, 3(3), Article 12.

Glaser, B. (2001a). Conceptualization: On theory and theorizing using grounded theory. International Journal on Qualitative Methods, 1(2), Article 3. Retrieved from https://sites.ualberta.ca/~iiqm/backissues/1_2Final/html/glaser.html

Glaser, B. (2001b). The grounded theory perspective: Conceptualization contrasted with description. Sociology Press.

Glaser, B. (1996). Gerund Grounded Theory: The basic social process dissertation. Sociology Press.

Glaser, B. (1978). Theoretical sensitivity: Advances in the methodology of Grounded Theory. Sociology Press.

Glaser, B., & Strauss, A. L. (1967). The discovery of grounded theory: Strategies for qualitative research. Aldine Transaction.

Hernandez, C. A. (2009). Theoretical coding in Grounded Theory methodology. Grounded Theory Review 8(3), 51–66.

Holton, J. (2010). The coding process and its challenges. Grounded Theory Review, 9(01), 21–40.

Linden, K. L., & Simmons, O. E. (2025). A front-row seat to the development of Grounded Theory: An interview with Odis Simmons. Grounded Theory Review, 24(1), 48–68.

Appendix 1: Glosssary

• Incident: A discrete chunk of data (line, event).
• Code (substantive): A conceptual label for incidents.
• Category: A higher-level concept grouping several codes (multi-indicator).
• Core Category: The central pattern that explains how the main concern is resolved.
• Theoretical Code: An emergent idea for relating categories (e.g., process, cause-consequence; context-contingency; degree-dimension).
• Theoretical Sampling: Collecting more data driven by the emerging theory to develop categories.
• Saturation: No new properties of core categories are emerging.

Appendix 2: Additional Guidance

The following is more detailed and specific to Classic Glasserian GT.

It’s the concept that’s important. Concepts are procedurally generated, emergent, and discovered. Minimize preconception; sort your memos; constant comparison; open → selective → theoretical coding; theoretical codes integrate the theory; saturate concepts; state modifiability.

• Stay close to, yet out of the data.
• Be mindful.
• Avoid the descriptive.
• You should have written it down.
• Be kind.
• Sort your memos.
• Minimize preconception.
• Substantive, open, selective, theoretical coding.
• Recording, taping, or talking leads to description.
• Theoretically sample; only when necessary.
• Delay peer/literature review; when necessary.
• Momentum of indicators, unchecked, leads to reversibility of indicators.

Constant Comparison Method

• Incident to incident generates concepts.
• Incident to concept generates categories.

Open Coding

• What is the study of?
• What category does the problem incident indicate?
  What property of the core category does the incident indicate?
• What theoretical codes may apply to integrate the emerging theory?

Natural Coding

• Concepts name pattern between 5-7 indicators; multi-indicator concepts.
• Saturate concepts.

Selective Coding

• Delimit coding to variables relating to core variables [emerged].

Theoretical Code

• [ to be annotated ]

Working Theory

• Guides
  • Resort thoughtfully.
  • Think theory and writing substantive concepts leads to conceptual integration
  • Relate concept to concept maintains theory conceptual generalization and general implications.
  • One general implication.
• Derailments
  • Unresearched jargon and conjecture leads to search for substance.
  • Talking leads to overgeneralizing and excessive indicators.
  • Insufficient sorting leads to paralyzed write-up.
  • Diagramming leads to audience seeking relevance.
  • Illustration leads to lightening conceptual density.
  • Analogy leads to introducing new concepts; leading to potential irrelevance and lack of fit.
  • Drama leads to introducing the concept of performance.
  • Academic field jargon leads to theoretical dilution.
  • Professional concern leads to dilution and/or derailment of theoretical transference.
  • Relating concept to description leads to lower conceptualization.
  • Relating concept to space, time, and people loses generality.
• Flow
  • Reading poetry leads to finding rhythm.
  • Known rhythms leads to stable audiences.

Working Paper

• Title states core category with and for grab.
• Subtitle may convey attribution of grounded theory method.
• Introduction states general nature of core category, a few properties, and reduce to core category.
• Opening summarizes theory.
• Opening states core category and a few, at most three (3) sub core concepts resolving a non-preconceived main concern to the field which emerged in area of interest.
• Content shares systematic construction of integrated emergent conceptual relationships.
• Content underscores modifiability of theory.

like a scientist in a laboratory, we do our best
to avoid contaminating the reagent,
that reagent is your mind, body, and heart.
this is how, we, touch the ground.¹

Notes

¹ see 真言