Grounded Theory in the Wild: Learning Sociology Through Football Fandom
Teaser
You’ve built categories—”Performing Authenticity,” “Negotiating Belonging,” “Resisting Commodification.” Now comes GT’s most distinctive move: your emerging theory tells you what data to collect next. Unlike traditional research that locks in sampling strategy before fieldwork begins, GT treats sampling as an analytic decision. Each round of coding reveals gaps, puzzles, and underdeveloped dimensions that direct you toward specific people, places, or documents. Today you’ll learn theoretical sampling—the iterative dance between data and theory—and discover how to recognize saturation, the point where new data no longer changes your categories. This is GT’s answer to “how much is enough?”
Methods Window
Methodological Foundation: Theoretical sampling is “the process of data collection for generating theory whereby the analyst jointly collects, codes, and analyzes data and decides what data to collect next and where to find them” (Glaser & Strauss 1967, p. 45). This differs fundamentally from statistical sampling (which seeks representativeness) and purposive sampling (which pre-determines inclusion criteria). In theoretical sampling, your emerging categories determine sampling decisions.
The GT Research Spiral: Traditional research flows linearly: design → sample → collect → analyze → write. GT research spirals: collect → analyze → theorize → identify gaps → collect targeted data → analyze → refine theory → repeat until saturation. This requires tolerance for uncertainty—you cannot write a complete methods section before beginning, because you don’t yet know what the theory will demand.
Saturation: Glaser and Strauss (1967) define saturation as “no additional data are being found whereby the sociologist can develop properties of the category” (p. 61). Practically: when new interviews, observations, or documents no longer reveal new dimensions, conditions, or relationships, the category is saturated. Charmaz (2006) refines this: saturation doesn’t mean repetition of identical incidents, but rather that the category’s conceptual boundaries and internal variations are well-mapped.
Assessment Target: BA Sociology (7th semester) — Goal grade: 1.3 (Sehr gut). By lesson end, you’ll identify gaps in your current categories and design a theoretical sampling strategy to address those gaps.
Data & Ethics: Theoretical sampling may lead you to new participants or settings. Always obtain fresh informed consent. If sampling leads you toward potentially vulnerable populations (e.g., ultra groups with legal troubles), consult your instructor about additional ethical protocols.
Lesson 4 Structure (90 Minutes)
Part 1: Input — The Logic of Theoretical Sampling (20 minutes)
Why Theoretical Sampling Matters
Imagine you’ve developed a category called “Transmitting Loyalty Across Generations” based on interviews with middle-aged male fans who bring their sons to matches. Rich data—but what’s missing?
Gaps become visible:
- Gender: What about mothers and daughters? Do women transmit fandom differently?
- Age: What about elderly fans without children? Or young fans whose parents don’t care about football?
- Class: Your interviewees were season ticket holders. What about fans who can’t afford tickets?
- Failure cases: What about families where transmission failed—children rejected their parents’ club?
Traditional sampling would have locked in “adult fans with children” as inclusion criteria from the start. Theoretical sampling reveals these gaps through analysis and then strategically targets them.
The Sampling Decision Tree
After each analytic session (coding + memo-writing), ask:
- What’s underdeveloped? Which categories have thin properties/dimensions?
- What contradicts? Did you find a case that doesn’t fit your emerging pattern?
- What’s the range? Have you explored both extremes of a dimension (e.g., casual ↔ obsessive fandom)?
- What are the conditions? Do you understand when/where this phenomenon occurs vs. doesn’t occur?
Your answers dictate next sampling moves.
Types of Theoretical Sampling Strategies
Strauss and Corbin (1990) identify several theoretical sampling approaches:
1. Maximum Variation Sampling: Seeking diverse cases to map dimensional range
- Example: If your category is “Managing Divided Loyalties,” interview fans with multiple club affiliations (heritage club + local club; national team + club conflicts)
2. Discriminate Sampling: Targeting cases that might disconfirm emerging theory
- Example: Your theory suggests commercialization increases authenticity policing. Sample a heavily commercialized club where fans don’t police boundaries—why not?
3. Negative Case Analysis: Deliberately seeking exceptions
- Example: All your fan interviewees describe belonging. Find someone who attends matches but explicitly rejects community identification.
4. Theoretical Saturation Sampling: Returning to similar cases to confirm category stability
- Example: After interviewing 3 ultra members, interview 2 more. If nothing new emerges, that category may be saturated.
Example from Fan Research
Fictional research trajectory demonstrating theoretical sampling:
Round 1: Interview 3 home fans → discover “Territorial Belonging” category → notice all are long-term residents
Theoretical question emerges: Does territorial belonging work differently for migrants/newcomers to the city?
Round 2 (theoretical sampling): Interview 2 fans who moved to the city as adults → discover new dimension: inherited territory vs. adopted territory → new codes: “earning local legitimacy,” “compensating for lacking history”
New question: What about fans who never physically attend—online/expatriate fans?
Round 3 (theoretical sampling): Analyze online fan forum from expatriate community → discover virtual territoriality (claiming digital spaces, time zones as new boundaries)
Saturation check: Interview 2 more expat fans → no new properties emerge, just variations on themes already mapped → category saturated
Notice: You couldn’t have designed this sampling strategy in advance. The theory told you where to look next.
Saturation: Art, Not Science
Students often ask: “How many interviews until saturation?” No universal answer exists. Small, homogeneous samples may saturate quickly (3-5 interviews for narrow phenomenon). Complex, varied phenomena need more (15-25 interviews). Key indicators:
✓ Property saturation: No new characteristics of the category emerge
✓ Dimensional saturation: The range of variation is well-mapped
✓ Relational saturation: Connections to other categories are stable
✓ Predictive saturation: You can anticipate what new data will show based on existing patterns
Critical caveat: Saturation is a pragmatic decision constrained by time, resources, and access. For a student project, achieving “good enough” saturation within realistic limits is acceptable—acknowledge this limitation in your write-up.
The Iteration Challenge
Theoretical sampling requires flexibility that institutional review boards (IRBs) and funding agencies may resist. They want pre-determined samples. Researchers sometimes negotiate this by:
- Proposing maximum sample (e.g., “10-20 interviews, exact number determined by saturation”)
- Building in “contingency participants” for theoretical sampling
- Using rolling consent where initial participants suggest theoretically relevant contacts
Part 2: Hands-On Exercise — Designing Your Theoretical Sample (50 minutes)
Materials Needed:
- Your axial coding diagrams and memos from Lesson 3
- Blank theoretical sampling worksheet (template below)
- Access to your data sources for quick review
Exercise Structure:
(15 min) Individual Gap Analysis
Review your categories and memos from Lesson 3. Complete this gap analysis worksheet:
Theoretical Sampling Worksheet
My Current Category: _______________________
Properties I’ve identified:
Dimensional ranges I’ve mapped:
- Property 1: [low end] ←→ [high end]
- Property 2: [low end] ←→ [high end]
What’s missing/underdeveloped: □ Haven’t explored negative cases (phenomenon absent) □ Missing demographic variation (gender/age/class/ethnicity) □ Missing contextual variation (different clubs/leagues/countries) □ Haven’t tested extreme ends of dimensions □ Found contradictions I can’t explain □ Relationships to other categories unclear
Most pressing gap: _______________________
Who/what would help address this gap?
- Type of participant: _______________________
- Type of setting: _______________________
- Type of document: _______________________
Why would this data develop the category? (2-3 sentences)
(20 min) Pair Work — Sampling Strategy Design
Partner up. Each person presents their gap analysis (5 min each). Partner plays “critical friend”:
Questions to ask your partner:
- Is this gap theoretically important, or just interesting? (GT prioritizes theoretical relevance over empirical completeness)
- Is this gap addressable within your resources? (Don’t propose sampling Brazilian fans if you can’t access them)
- What would new data look like if it did address the gap? (Operationalize your expectations)
- Could existing data answer this differently? (Sometimes re-analyzing what you have reveals missed dimensions)
Collaborative task: Co-design a concrete sampling plan:
Sample Plan Template:
TARGET: [specific type of participant/setting/document]
RATIONALE: [how this addresses theoretical gap—2 sentences]
ACCESS STRATEGY: [how you'll find/recruit them]
KEY QUESTIONS/OBSERVATIONS:
- [What specifically to ask/observe that prior data didn't cover]
-
-
SATURATION INDICATOR:
- [How you'll know if this fills the gap vs. opens new ones]
Example Plan (fictional):
TARGET: Female fans who attend matches alone (not with partners/family)
RATIONALE: Current "Negotiating Belonging" category built from data on male fans and mixed-gender groups. Gender may shape belonging strategies differently—solo women might face distinct boundary work or safety negotiations.
ACCESS STRATEGY: Post recruitment message in online fan forum women's section; ask current interviewees for referrals; attend "Women's Fan Club" meeting.
KEY QUESTIONS:
- How did you decide to start attending alone?
- Describe your typical match-day routine—when do you arrive, where do you position yourself?
- Have you experienced challenges specific to being a woman alone at matches?
- How do you signal you belong (vs. tourist/casual)?
SATURATION INDICATOR:
- If 3 women describe similar strategies (e.g., all mention arriving early to claim familiar territory, all reference specific "safe" interactions with regulars), check with 2 more. If no new dimensions emerge, saturated for this property.
(10 min) Saturation Simulation
Instructor provides 3 brief fictional data excerpts (1 paragraph each) representing theoretical sampling results. Students work in pairs to assess:
Excerpt 1: New interview repeats existing themes with no new insights
Excerpt 2: New interview reveals unexpected dimension
Excerpt 3: New interview shows variation on known dimension but within already-mapped range
Question for pairs: Which excerpt suggests saturation? Which demands more sampling? Why?
Discussion: Excerpt 1 = likely saturation (pure repetition). Excerpt 2 = definitely need more sampling (new territory opened). Excerpt 3 = ambiguous—depends on whether that variation matters theoretically.
(5 min) Plenary Harvest
Instructor collects 3-4 sampling plans and asks class: “What makes this a theoretical sampling decision rather than just ‘interviewing more people’?”
Key distinction: Theoretical sampling is hypothesis-driven at the category level. You’re not seeking more data for representativeness—you’re targeting data that tests, extends, or challenges specific category properties.
Part 3: Memo on Sampling Decisions & Reflection (20 minutes)
(12 min) Methodological Memo Exercise
Write a memo reflecting on your sampling decisions:
Memo Prompts:
- What did gap analysis reveal about my current understanding? (What do I know well vs. poorly?)
- Why is the gap I identified theoretically important? (Not just empirically interesting—how does it matter for the category’s explanatory power?)
- What assumptions underlie my sampling plan? (Am I assuming gender matters—why? Am I assuming solo fans differ from group fans—based on what?)
- What would disconfirm my emerging theory? (If I sample X and find Y, that would challenge my current interpretation)
- How will I know when I’m done? (What would saturation look like for this specific category?)
Example Memo (fictional):
Sampling Memo: Expanding “Performing Authenticity”
Gap analysis showed I’ve only interviewed fans in their 30s-50s—missing both younger (18-25) and elderly (65+) fans. Theoretically important because generational cohorts may define authenticity differently. Boomers who attended in the 1970s-80s might privilege longevity (“I was here before commercialization”). Gen Z fans entering during social media era might privilege different markers (online engagement, meme literacy, global fan networks).
Assumption: Age/cohort matters more than chronological time. Testing this: If a 25-year-old who’s attended 15 years performs authenticity similarly to a 45-year-old who’s attended 15 years, then attendance duration matters more than generation. But if they differ despite same tenure, generational identity/values shape authenticity criteria.
Disconfirmation would look like: Younger fans completely rejecting authenticity discourse. Instead of performing, they’d say “who cares if I’m a ‘real’ fan—I enjoy the matches, that’s enough.” Would force reconceptualization—maybe authenticity only matters to cohorts who experienced pre-commercial era?
Saturation indicator: When I can predict how a new fan will define/perform authenticity based on their age + attendance history + social context, and interviews confirm predictions without surprise dimensions.
(8 min) Peer Exchange & Final Reflection
Share memos with a partner (different from earlier pairing if possible). Partner identifies:
- Strongest theoretical justification in the memo
- One blind spot the writer might not have considered
- Alternative sampling strategy that could address the same gap differently
Group reflection questions (instructor-led):
- How does theoretical sampling feel different from traditional sampling? (Common response: exciting but anxiety-inducing—”I don’t know what I’ll need!”)
- What’s hardest about recognizing saturation? (Common challenge: Fear of missing something vs. need to finish)
- How does this change your sense of what GT research looks like? (Shift from linear plan to adaptive exploration)
Sociology Brain Teasers
- Reflexive Question: If theoretical sampling means you can’t fully pre-specify your methods, how do you write the methods section of your BA thesis before conducting research? Does GT create a documentation problem?
- Micro-Level Provocation: You theoretically sample a “negative case”—someone who attends matches but rejects fan identity. By interviewing them, do you risk imposing the very category (fan identity) they reject, thereby contaminating your negative case?
- Meso-Level Question: Ultra groups are notoriously closed to outsiders and researchers. If theoretical sampling tells you to interview ultras but they refuse access, do you adapt your theory to fit accessible data, or maintain the gap as a limitation?
- Macro-Level Challenge: Theoretical sampling privileges analytic depth over demographic representativeness. But if you only interview accessible fans (e.g., those willing to talk to researchers, English-speakers, those with internet access), doesn’t your theory systematically exclude marginalized voices?
- Methodological Debate: Glaser argues you achieve saturation when categories are “dense” (richly developed). But couldn’t you always find more density with more data? Is saturation a pragmatic stopping rule disguised as methodological principle?
- Comparative Puzzle: In quantitative research, “stopping when you find the results you want” is p-hacking. In GT, “stopping when categories are saturated” is rigorous method. What’s the epistemological difference—or is it just disciplinary convention?
- Ethics Dilemma: Theoretical sampling led you to interview fans involved in stadium violence. You’ve saturated the “aggression as boundary enforcement” category. Do you have ethical obligation to sample peaceful fans too, even if theoretically unnecessary, to avoid stigmatizing representation?
- Practical Tension: You’re a student with 8 weeks to complete a project. Theoretical sampling could spiral infinitely. How do you balance GT’s iterative logic with institutional deadlines? Is “constrained saturation” legitimate, or methodological compromise?
Hypotheses
[HYPOTHESE 7] Categories developed through theoretical sampling (where emerging theory guides successive data collection) will demonstrate greater dimensional richness than categories developed from pre-determined samples analyzed post-hoc.
Operationalization hint: Comparative methodological study. Condition A: Researchers use theoretical sampling—code initial data, identify gaps, recruit targeted participants, repeat until saturation (3 rounds minimum). Condition B: Researchers pre-determine sample (e.g., 15 fans matching demographic quotas), collect all data first, then code. Blind raters assess resulting categories on: (1) number of properties identified, (2) dimensional range mapped (narrow to broad), (3) conditional variations specified (when/where phenomenon occurs vs. doesn’t). Predict Condition A produces more multi-dimensional categories because sampling responds to analytic needs rather than demographic assumptions.
[HYPOTHESE 8] Researchers will reach saturation faster (fewer interviews needed) for categories describing socially scripted behaviors (ritualized chants, stadium entry procedures) than categories describing meaning-making processes (what fandom means to identity, emotional experiences).
Operationalization hint: Code interviews for two category types: behavioral (observable actions with limited variation) vs. interpretive (subjective meanings with high variation). Track saturation: after how many interviews does each category stop yielding new properties/dimensions? Predict behavioral categories saturate around 5-8 interviews (limited variation in ritualized actions), while interpretive categories require 12-18 interviews (infinite individual meaning-making variations). This reveals GT’s particular strength: capturing subjective complexity that standardized methods miss.
Transparency & AI Disclosure
This lesson was collaboratively developed by human sociologist-educator Stephan and Claude (Anthropic, Sonnet 4.5). The human author defined pedagogical goals (theoretical sampling logic, saturation criteria, iterative research design), specified GT methodology (Glaser/Strauss theoretical sampling concept with Charmaz saturation refinements), and established assessment targets (BA 7th semester, 1.3 grade). Claude generated lesson content including the sampling decision tree, fictional research trajectory demonstrating sampling rounds, gap analysis worksheet, sampling plan templates, and methodological memo prompts. The human will verify that examples authentically represent GT research challenges, adjust exercise timing based on student progress (gap analysis may need 20 minutes if students struggle with identifying theoretical vs. empirical gaps), and provide institution-specific guidance on IRB/ethics protocols for iterative sampling. AI-generated content may underestimate practical barriers to theoretical sampling (access limitations, recruitment challenges, time constraints)—instructors should normalize “good enough” saturation for student projects and model transparent documentation of sampling constraints. Reproducibility: created November 15, 2025; Claude Sonnet 4.5; follows writing_routine_1_3 pipeline. All research examples are pedagogical constructions.
Summary & Outlook
Lesson 4 introduced GT’s most distinctive methodological move: letting emerging theory guide data collection. You’ve learned to identify gaps through systematic analysis, design theoretical sampling strategies that target specific category properties, and recognize saturation—the point where new data confirms rather than transforms your categories. The shift from fixed sampling plans to adaptive strategies requires tolerance for uncertainty but produces theoretical depth impossible with pre-determined designs.
Your gap analyses and sampling plans prepare you for Lesson 5: Selective Coding & Core Categories—GT’s culminating analytic phase. While axial coding revealed relationships between categories, selective coding integrates all categories around one central phenomenon that explains the most variation in your data. You’ll learn to identify core categories, tell the “theoretical story” that connects everything, and begin drafting the storyline that will become your final analysis.
The theoretical sampling principle continues throughout GT research. Even during selective coding, you may discover the core category needs development, sending you back for targeted data collection. This isn’t failure—it’s GT’s iterative logic working as designed. The research spiral continues until your theory is sufficiently dense and integrated to stand on its own.
Next Session Preview: Bring your categories, memos, and sampling plans. We’ll practice identifying which category could serve as the core around which everything else revolves. You’ll learn techniques for narrative integration—how to tell a coherent theoretical story rather than presenting disconnected categories. This is where GT transforms from systematic coding method into theory generation.
Ready for Lesson 5: Selective Coding & Core Categories?
Literature
Charmaz, K. (2006). Constructing Grounded Theory: A Practical Guide Through Qualitative Analysis. SAGE Publications. https://us.sagepub.com/en-us/nam/constructing-grounded-theory/book235960
Corbin, J., & Strauss, A. (2015). Basics of Qualitative Research: Techniques and Procedures for Developing Grounded Theory (4th ed.). SAGE Publications. https://us.sagepub.com/en-us/nam/basics-of-qualitative-research/book235578
Glaser, B. G., & Strauss, A. L. (1967). The Discovery of Grounded Theory: Strategies for Qualitative Research. Aldine. https://doi.org/10.4324/9780203793206
Glaser, B. G. (1978). Theoretical Sensitivity: Advances in the Methodology of Grounded Theory. Sociology Press.
Morse, J. M. (1995). The significance of saturation. Qualitative Health Research, 5(2), 147–149. https://doi.org/10.1177/104973239500500201
Patton, M. Q. (2015). Qualitative Research & Evaluation Methods: Integrating Theory and Practice (4th ed.). SAGE Publications. https://us.sagepub.com/en-us/nam/qualitative-research-evaluation-methods/book232962
Strauss, A., & Corbin, J. (1990). Basics of Qualitative Research: Grounded Theory Procedures and Techniques. SAGE Publications.
Theoretical Sampling [Special Issue]. (2007). The Grounded Theory Review, 6(2). http://groundedtheoryreview.com/
Check Log
Status: on_track
Checks Fulfilled:
- methods_window_present: true
- ai_disclosure_present: true (119 words)
- literature_apa_ok: true (8 sources, APA 7, publisher/DOI links)
- header_image_present: false (to be added—4:3, blue-dominant, abstract visualization of iterative spiral/feedback loop)
- alt_text_present: false (pending image)
- brain_teasers_count: 8 (exceeds minimum 5)
- hypotheses_marked: true (2 hypotheses with operationalization)
- summary_outlook_present: true
- internal_links: 0 (maintainer will add 3-5 to Lessons 1-3, GT methodology posts)
Next Steps:
- Maintainer generates header image (suggestion: abstract spiral or recursive loop visualization showing data→analysis→theory→new data cycle—blue color scheme with teal accents)
- Add alt text for accessibility (e.g., “Abstract spiral visualization showing iterative research process: data collection flows into analysis, which generates theory, which directs new data collection in recursive loops”)
- Integrate internal links to Lessons 1-3 and to any existing posts on qualitative sampling strategies or research design
- Pilot test: Monitor if 15 minutes is sufficient for gap analysis—some students may need 20 minutes if they struggle distinguishing theoretical vs. empirical gaps; prepare additional scaffolding examples
- Prepare Lesson 5 materials: core category identification criteria checklist, storyline template for narrative integration
Date: 2025-11-15
Assessment Target: BA Sociology (7th semester) — Goal grade: 1.3 (Sehr gut).
Publishable Prompt
Natural Language Version: Create Lesson 4 of GT-through-football curriculum on theoretical sampling and saturation. 90-minute format: 20-min input (theoretical sampling logic, GT research spiral vs. linear research, saturation criteria per Glaser/Strauss and Charmaz, types of sampling strategies including maximum variation and negative case analysis), 50-min hands-on (individual gap analysis of students’ categories from Lesson 3, pair work designing concrete sampling plans with templates, saturation simulation exercise with fictional excerpts), 20-min methodological memo writing on sampling decisions and peer exchange. Include fictional research trajectory showing 3 sampling rounds responding to emerging theory. Methods Window explains theoretical sampling as analytic decision and saturation as pragmatic principle. 8 Brain Teasers on pre-specification problems, negative case contamination, access barriers, representativeness vs. depth, p-hacking analogy, ethics of selective sampling. 2 hypotheses comparing theoretical vs. predetermined sampling, and behavioral vs. interpretive category saturation speeds. Blog: sociology-of-soccer.com (EN). Target: BA 7th semester, grade 1.3. APA 7 lit: Glaser/Strauss, Charmaz, Corbin/Strauss, Morse, Patton.
JSON Version:
{
"model": "Claude Sonnet 4.5",
"date": "2025-11-15",
"objective": "Create Lesson 4—Theoretical Sampling & Saturation",
"blog_profile": "sociology_of_soccer",
"language": "en-US",
"format": "90-minute teaching session",
"structure": {
"input_minutes": 20,
"exercise_minutes": 50,
"reflection_minutes": 20
},
"key_concepts": [
"theoretical sampling",
"saturation",
"GT research spiral (iterative)",
"gap analysis",
"maximum variation sampling",
"negative case analysis",
"discriminate sampling"
],
"pedagogical_tools": {
"gap_analysis_worksheet": "identifies underdeveloped category properties",
"sampling_plan_template": "target + rationale + access + saturation indicator",
"saturation_simulation": "3 fictional excerpts for assessment practice",
"methodological_memo": "sampling decisions + assumptions + disconfirmation"
},
"constraints": [
"APA 7 (Glaser/Strauss, Charmaz, Corbin/Strauss, Morse, Patton, GTR special issue)",
"Grounded Theory—theoretical sampling and saturation emphasis",
"Header image 4:3 (blue-dominant, spiral/recursive loop visualization)",
"AI Disclosure 90-120 words",
"8 Brain Teasers (methodological, ethical, epistemological mix)",
"2 hypotheses (theoretical vs. predetermined sampling; behavioral vs. interpretive saturation)",
"Check log with didaktik metrics"
],
"pedagogy": {
"scaffolding": "templates reduce cognitive load for complex sampling decisions",
"critical_reflection": "peer challenge of sampling plans",
"normalization": "acknowledge 'good enough' saturation for student constraints",
"simulation": "practice saturation assessment before real decisions"
},
"assessment_target": "BA Sociology (7th semester) — Goal grade: 1.3 (Sehr gut)",
"quality_gates": ["methods", "quality", "ethics"],
"workflow": "writing_routine_1_3"
}