qualitative-research-pro 1.0.0

package/rules/finding-output-format.md

@@ -0,0 +1,47 @@
+# Finding Output Format (Qualitative / GT)
+
+Use this rule when presenting findings, sharing interim results, or preparing audit-friendly summaries of categories.
+
+## Category Presentation
+
+For each category, provide:
+
+1. **Name** — concise conceptual label.
+2. **Definition** — what the category denotes and excludes.
+3. **Properties** — characteristics that vary within the category.
+4. **Dimensions** — continua or poles along which properties vary (when empirically grounded).
+
+Order categories to reflect analytic story (often core-related first in GT presentations).
+
+## Data Extract Formatting
+
+- Use **block quotes** for verbatim material; indent or style per APA 7 for long quotations.
+- Attribute with **pseudonym** and minimal context (role, setting type) without re-identifying.
+- Provide **line or timestamp references** when available (`lines 12–18`, `00:03:22–00:04:10`).
+- Keep excerpts **representative**, not cherry-picked outliers, unless labeled as negative or deviant cases.
+
+## Analytical Commentary Format
+
+- **Claim** — state the interpretive point in one or two sentences.
+- **Evidence** — cite excerpt or summarized pattern.
+- **Significance** — link to broader category, process, or theory (what this instance shows about the pattern).
+- Avoid repeating the quote in different words without adding conceptual value.
+
+## Theoretical Proposition Format
+
+- State relationships between categories as **conditional or processual** claims grounded in comparison.
+- Example pattern: “When {condition}, participants tend to {action/strategy}, which {consequence}, unless {boundary condition}.”
+- Number propositions for cross-referencing in discussion and models.
+
+## Visual Model Format
+
+- Provide a **simple figure** (boxes/arrows or staged process) mapping core category to related categories.
+- Legend: define shapes/line types; keep labels consistent with the text.
+- Caption: what the model represents, scope, and known limitations.
+- Prefer clarity over decorative complexity; update figures when theory is delimited.
+
+## Packaging Checklist
+
+- [ ] Each category has definition, properties/dimensions, and at least one anchored excerpt.
+- [ ] Commentary interprets, not merely restates.
+- [ ] Propositions and visuals align with memo trails and coding definitions.

package/rules/gt-coding-standards.md

@@ -0,0 +1,40 @@
+# Grounded Theory Coding Standards
+
+Use this rule for Glaserian-oriented grounded theory coding, memoing, and phase transitions. Adapt labels if using a variant approach, but preserve comparative logic.
+
+## Open Coding
+
+- Code **line-by-line** or **incident-to-incident** early in the project; treat each segment as potentially theoretical.
+- Prefer **gerunds** for process codes when they fit the data (e.g., *managing uncertainty*).
+- Stay open: hold imported frameworks lightly; let categories emerge through **constant comparison**.
+- Revise codes when comparisons reveal misfit; retire codes that do not earn their place.
+
+## Selective Coding
+
+- Begin selective coding **only after** a plausible **core category** has emerged and earned centrality through comparison.
+- Focus coding and memoing on relationships to the core, delimiting the theory’s scope.
+- Do not force a core category for narrative convenience; return to data and memos if fit is weak.
+
+## Theoretical Coding
+
+- Apply **coding families** (e.g., causal, consequence, strategy, dimension) to integrate categories theoretically.
+- Use theoretical codes to specify **how** categories relate, not merely that they co-occur.
+- Diagram relationships only when they clarify comparison results, not as decoration.
+
+## Code Naming Conventions
+
+- Use concise, conceptual labels; avoid long sentences as code names.
+- Maintain a **codebook** with definitions, inclusion/exclusion notes, and exemplar excerpts.
+- Distinguish **in vivo** codes (participant language) from analyst-constructed concepts when useful.
+
+## Memo Triggers During Coding
+
+- Stop coding when a conceptual leap, tension, or comparison insight appears; **memo first**.
+- Record comparisons (incident-to-incident, incident-to-concept, concept-to-concept) and emergent hypotheses.
+- Date memos; link them to codes and data locations for auditability.
+
+## When to Shift From Open to Selective Coding
+
+- Categories show **density** and **variation**; new data mainly refine rather than introduce new properties.
+- A core category **recurs**, **relates widely**, and **explains** much of what is going on.
+- Theoretical sampling targets **gaps** in the emerging theory rather than broad exploration alone.

package/rules/methodological-rigor.md

@@ -0,0 +1,56 @@
+# Methodological Rigor (Qualitative Research)
+
+Use this rule when designing studies, analyzing data, or evaluating trustworthiness and theory quality in qualitative inquiry.
+
+## Lincoln & Guba: Trustworthiness
+
+**Credibility** — Confidence that findings are grounded in participants’ experiences.
+
+- Prolonged engagement and persistent observation where feasible.
+- Triangulation across data types, investigators, or methods when appropriate.
+- Member checking (use judiciously; treat feedback as more data, not validation).
+- Peer debriefing and negative case analysis.
+- Thick description and audit trails linking claims to data.
+
+**Transferability** — Sufficient contextual detail for readers to judge applicability elsewhere.
+
+- Rich description of setting, participants (anonymized), and processes.
+- Clear boundaries of the analytic claims (“in this context…”).
+
+**Dependability** — Stability and traceability of the analytic process.
+
+- Documented procedures, versioned coding schemes, dated memos, decision logs.
+- Reflexive notes on how interpretations shifted and why.
+
+**Confirmability** — Findings are tied to evidence rather than researcher bias alone.
+
+- Transparent chain of evidence from excerpt → code → category → interpretation.
+- Explicit separation of inferences from raw data in write-ups.
+
+## Glaser: Grounded Theory Quality Criteria
+
+**Fit** — Categories emerge from and pattern the data; minimal forcing.
+
+**Work** — The emerging theory explains variation, processes, and relationships in the substantive area.
+
+**Relevance** — The theory addresses what actually matters to participants and the phenomenon.
+
+**Modifiability** — Categories remain open to refinement as new data appear.
+
+## Techniques Mapped to Criteria
+
+| Criterion        | Example techniques                                      |
+|-----------------|-----------------------------------------------------------|
+| Credibility     | Constant comparison, memoing, deviant cases               |
+| Transferability | Contextual narrative, sampling rationale                  |
+| Dependability   | Audit trail, coding definitions, dated analytic memos     |
+| Confirmability  | Exemplars, intercoder logs (if used), reflexivity notes   |
+| Fit/work/etc.   | Theoretical sampling, selective coding, theoretical coding |
+
+## Pre-Submission Checklist
+
+- [ ] Every major claim is supported by data with traceable excerpts or field evidence.
+- [ ] Procedures for coding, memoing, and sampling are described transparently.
+- [ ] Limitations and scope (who/where/when) are explicit.
+- [ ] Reflexivity is addressed without displacing participant meaning.
+- [ ] GT claims align with fit, work, relevance, and modifiability (if using GT).

package/rules/quality-criteria.md

@@ -0,0 +1,41 @@
+# Quality Criteria (GT and Qualitative Research)
+
+Use this rule to choose evaluation frameworks, design review criteria, and report how quality was assessed.
+
+## Glaser’s Four Criteria (Grounded Theory)
+
+**Fit** — Indicators: categories arise from comparisons; deviant cases are accommodated or explain boundaries; minimal retrofitting of external frameworks.
+
+**Work** — Indicators: theory explains processes, consequences, and variation; handles “how” questions in the substantive area.
+
+**Relevance** — Indicators: core problems and strategies mirror participant concerns; theoretical sampling targets emergent gaps.
+
+**Modifiability** — Indicators: categories refine with new data; the researcher revises rather than defends misfitting labels.
+
+## Lincoln & Guba: Trustworthiness (Naturalistic Inquiry)
+
+- **Credibility**, **transferability**, **dependability**, **confirmability** (see `methodological-rigor.md` for techniques).
+- Appropriate when reporting constructivist or interpretive studies emphasizing contextual understanding and auditability.
+
+## Charmaz (Constructivist GT): Credibility, Originality, Resonance, Usefulness
+
+- **Credibility** — Adequate evidence, researcher visibility, and persuasive warranting.
+- **Originality** — Fresh insights, not mere redescription of common sense.
+- **Resonance** — Rings true to lived experience; handles emotional and tacit dimensions.
+- **Usefulness** — Offers interpretive or practical traction for audiences.
+
+Use for **comparison** or **supplement** when writing within constructivist GT or when reviewers expect explicit usefulness framing.
+
+## When Each Framework Fits
+
+| Situation | Primary framework |
+|-----------|-------------------|
+| Classic/Glaserian GT emphasis | Glaser’s fit, work, relevance, modifiability |
+| Thick contextual, interpretive account | Lincoln & Guba + transparent methods |
+| Constructivist GT / engaged scholarship | Charmaz + trustworthiness as appropriate |
+| Mixed or pragmatic qualitative | Declare criteria upfront; avoid checklist stacking without rationale |
+
+## Reporting Guidance
+
+- State **which criteria** you used and **how** you operationalized them (memo trails, sampling, negative cases, peer debriefs).
+- Do not claim saturation or credibility by volume alone; tie claims to analytic process evidence.

package/rules/reflexivity-requirements.md

@@ -0,0 +1,40 @@
+# Reflexivity Requirements
+
+Use this rule to document researcher positionality, manage bias, and align reflexive practice with grounded theory without conflating concepts.
+
+## Positionality Statement (Typical Components)
+
+- Social identities, roles, and stakes relevant to the topic (e.g., insider/outsider status).
+- Prior experience, training, and theoretical preferences that could shape attention.
+- Relationships with sites, gatekeepers, or communities and how they evolved.
+- Strategies used to bracket or interrogate assumptions during fieldwork and analysis.
+
+## Reflexive Journal Requirements
+
+- Maintain dated entries for **decision points**, emotional responses, surprises, and rapport dynamics.
+- Log **coding dilemmas**, alternative interpretations, and conversations with peers or supervisors.
+- Note how memos and categories shifted; link entries to data locations or analytic artifacts.
+
+## Assumptions and Preconceptions
+
+- Explicitly list entering hypotheses or worries; revisit them after major coding passes.
+- Treat unexamined preferences as risks to **confirmability**; seek disconfirming evidence and negative cases.
+- Distinguish personal reactions from analytic claims—label each clearly in journals.
+
+## Reflexivity vs. Glaser’s Theoretical Sensitivity
+
+- **Reflexivity** interrogates the researcher’s standpoint, power, and influence on co-construction of meaning.
+- **Theoretical sensitivity** (Glaser) names the cultivated ability to see meaning in data, informed by broad reading and comparison—not forcing preconceived theory onto data.
+- Both can coexist: reflexivity **curtails** imposition; theoretical sensitivity **sharpens** recognition of emergent patterns when grounded in comparison.
+
+## Ongoing Practice Throughout the Study
+
+- Schedule periodic reflexivity reviews (e.g., after each interview wave or coding sprint).
+- Integrate succinct reflexivity into methods sections; avoid displacing participant-centered claims.
+- Use peer debriefs or audit partners to surface blind spots identified in the journal.
+
+## Integration Into Reporting
+
+- Brief, focused reflexivity in the main text; extended logs in appendices only when ethical and permitted.
+- Align claims with audit trails so readers can assess how reflexive awareness shaped (and constrained) interpretation.
+- Where teams code together, document how differing standpoints were negotiated and recorded.

package/rules/research-ethics-standards.md

@@ -0,0 +1,44 @@
+# Research Ethics Standards
+
+Use this rule for human-participant research, IRB protocols, fieldwork, and any storage or publication of qualitative materials.
+
+## IRB and Regulatory Compliance
+
+- Obtain IRB or equivalent ethical review before collecting identifiable or sensitive data.
+- Follow approved protocol; document amendments and serious adverse events per institutional policy.
+- Retain consent documentation and approval letters per retention schedules.
+
+## Informed Consent
+
+- Use plain language; disclose purpose, procedures, risks, benefits, and contacts.
+- Clarify recording, future use, data sharing, and any secondary analysis plans.
+- For digital or asynchronous participation, verify understanding and voluntary agreement.
+- Assent processes for minors; guardian permission where required.
+
+## Confidentiality and Pseudonyms
+
+- Assign pseudonyms at the earliest stable point in the workflow; use them consistently in notes and drafts.
+- Separate keys linking pseudonyms to identities from analytic files; restrict key access.
+- Avoid indirect identifiers (unique job titles, rare events, small communities) in published excerpts.
+
+## Data Security
+
+- Encrypt devices; use institutional-approved storage (not personal cloud defaults for restricted data).
+- Role-based access; unique accounts; no shared passwords for restricted datasets.
+- Secure transfer (institutional VPN, approved file exchange); avoid unencrypted email for sensitive files.
+
+## Vulnerable Populations
+
+- Additional safeguards for children, incarcerated persons, trauma-exposed groups, and others with heightened risk.
+- Minimize burden; offer referrals; monitor distress protocols with trained personnel.
+- Avoid coercion or undue influence; compensate fairly without inducing vulnerability.
+
+## Autonomy and Power Dynamics
+
+- Affirm the right to withdraw at any time without penalty; clarify what can and cannot be withdrawn post-publication.
+- Acknowledge asymmetry; avoid exploitative rapport; do not leverage gatekeeper power.
+
+## Version Control and Repositories
+
+- **Never** commit raw audio, video, transcripts with identifiers, consent PDFs, or key files to git or public repos.
+- Use `.gitignore` and environment-specific paths; share only de-identified excerpts in supplements when permitted.

package/skills/.gitkeep

@@ -0,0 +1,2 @@
+# Sandbox directory for testing skills
+# Skills in this directory are used for flywheel testing and should not be used in production

package/skills/academic-writing/SKILL.md

@@ -0,0 +1,73 @@
+---
+name: academic-writing
+description: Use when writing academic prose for qualitative research — findings, methods, discussion sections with appropriate voice and conventions.
+---
+
+# Academic Writing for Qualitative Research
+
+Qualitative writing must **show analytic work** while remaining readable. The goal is disciplined interpretation: claims tethered to evidence, voice appropriate to genre, and structure that guides readers through complex meaning.
+
+## Conventions that matter
+
+- Prefer **precision** over vague intensifiers (“very,” “extremely”) unless justified.
+- Define specialized terms on first substantive use.
+- Use **past tense** for what you did; **present tense** for claims you treat as stable findings in the narrative (discipline norms vary—mirror target journal).
+- Keep **parallel structure** in lists of themes/categories.
+
+## Voice: active vs passive; first person
+
+First person (“we/I”) is increasingly acceptable in qualitative methods writing because agency and positionality matter. Use active voice for clarity (“We analyzed transcripts”) while avoiding self-centered prose that eclipses participants’ words.
+
+## Hedging language
+
+Qualitative claims are often **probabilistic and situated**. Appropriate hedges: “suggests,” “indicates,” “in this sample,” “participants described.” Avoid hedging so heavily that claims disappear; avoid over-certainty without evidence.
+
+## Paragraph structure
+
+One main idea per paragraph; opening sentence states the claim; following sentences support with reasoning, excerpt references, or cross-case comparison; final sentence transitions or deepens significance.
+
+## Signposting
+
+Use headings, forecasting sentences, and explicit roadmap paragraphs in long documents (dissertations). In articles, respect word limits—signpost minimally but clearly.
+
+## Integrating quotes
+
+Introduce excerpts; avoid “quote dumping.” Trim quotes with ellipses only ethically and transparently. Follow APA block quote rules when applicable.
+
+## Presenting data extracts
+
+Pair excerpts with **analytic commentary**: what the excerpt illustrates, how it varies across cases, and how it links to a category. Name the participant pseudonym and data type (interview, field note).
+
+## Analytical commentary patterns
+
+- **Excerpt → code label → interpretation → boundary condition.**
+- **Contrast case:** show two excerpts to clarify dimensional range.
+
+## Writing for different audiences
+
+Dissertations can show process (audit trail, extensive methods). Journal articles foreground **contribution** and compress procedural detail into supplements. Practitioner briefs emphasize actionable insights with ethical caveats.
+
+## Common mistakes
+
+- Theme lists without mechanisms.
+- Methodological buzzwords without operational meaning.
+- Ignoring disconfirming evidence.
+- Conflating description with theory.
+
+## Grounded theory writing tip
+
+Move from **categories** to **theoretical statements** about relationships among categories (conditions, strategies, consequences). Use memos as drafting scaffolding.
+
+## Checklist
+
+- [ ] Each major claim anchored to data displays or citations to corpus.
+- [ ] Voice and tense consistent with venue guidelines.
+- [ ] Quotes introduced, trimmed ethically, interpreted.
+- [ ] Disconfirming cases addressed.
+- [ ] Contribution and limits explicit.
+
+## References (starting points)
+
+- Wolcott, H. F. *Writing up qualitative research* (3rd ed.).
+- Belcher, W. L. *Writing your journal article in twelve weeks* — adaptable to qualitative articles.
+- Glaser, B. G. *Doing Grounded Theory* — theory presentation discipline.

package/skills/action-research/SKILL.md

@@ -0,0 +1,96 @@
+---
+name: action-research
+description: "Use when conducting participatory action research (PAR) involving collaborative inquiry, cycles of action and reflection."
+---
+
+# Action Research and Participatory Action Research (PAR)
+
+**Action research (AR)** links **inquiry** and **action** in iterative cycles. Stakeholders examine practices, implement changes, and reflect on outcomes. **Participatory action research (PAR)** foregrounds **collaboration**, **power-sharing**, and often **social justice**, positioning community or practitioner co-researchers as knowledge producers.
+
+## Core action research cycle
+
+While models differ in labeling, a robust cycle includes:
+
+1. **Plan** — Identify a problem or opportunity with stakeholders; review evidence; define aims and ethical safeguards.
+2. **Act** — Implement a contextually feasible intervention, change, or trial practice.
+3. **Observe** — Collect data on process and outcomes (field notes, interviews, artifacts, metrics as appropriate).
+4. **Reflect** — Analyze findings collectively; revise understanding; decide the next cycle’s focus.
+
+Cycles are **spiraling**, not one-off: each turn refines questions and practice.
+
+## Types of action research
+
+### Participatory action research (PAR)
+
+Emphasizes **democratic participation**, **local knowledge**, and **structural critique**. Common in community health, education, and international development when done ethically and non-extractively.
+
+### Community-based participatory research (CBPR)
+
+Partnership between researchers and communities across **all phases**—agenda setting, design, analysis, dissemination. Governance and benefit-sharing are explicit.
+
+### Classroom / practitioner action research
+
+Teachers or clinicians study their own practice for **local improvement** and **professional learning**. Often smaller scale; still requires rigor and ethical clarity.
+
+### Appreciative inquiry (AI)
+
+Builds change from **strengths** and **positive exemplars** rather than deficit framing. Uses **discovery, dream, design, destiny** phases in organizational contexts.
+
+## Researcher positionality in AR
+
+Facilitators may be **insiders** (employees, members) or **outsiders** (academic partners). Critical issues:
+
+- **Who sets the agenda?**
+- **Who benefits from publications and grants?**
+- **How is conflict among stakeholders handled?**
+
+Reflexive journaling, transparent agreements, and rotating facilitation can reduce **extractive** dynamics.
+
+## Collaborative inquiry principles
+
+- **Co-learning** — Expertise is distributed; academic theory meets local knowledge.
+- **Shared decision-making** — Protocols for consent, data ownership, and authorship.
+- **Action orientation** — Knowledge claims tied to **practical effects** and **next steps**.
+- **Inclusivity** — Design for marginalized voices to shape analysis, not only data extraction.
+
+## Quality criteria in action research
+
+Traditional positivist criteria fit poorly. Common alternatives include:
+
+- **Outcome validity** — Actions produce meaningful, context-appropriate improvement.
+- **Process validity** — Inquiry is fair, inclusive, and methodologically sound for the setting.
+- **Democratic validity** — Diverse stakeholders influence conclusions and actions.
+- **Catalytic validity** — Research **energizes critical reflection** and **empowerment** (Lather).
+- **Dialogic validity** — Truth tested through **sustained dialogue** and **challenge** within the community (Anderson & Herr).
+
+Combine these with **ethical validity** (do no harm, consent, privacy) and **ecological validity** (fits real-world constraints).
+
+## Kemmis & McTaggart’s framework
+
+Kemmis, McTaggart, and Nixon emphasize **participatory** and **public** qualities of AR. Cycles are not merely technical but **social practices** that can be:
+
+- **Self-reflective** — Practitioners examine their own practices.
+- **Collaborative** — Joint inquiry across roles.
+- **Public** — Knowledge is shared beyond private teams to wider communities.
+
+They stress **communicative space** where arguments are tested and **practice architectures** (material, social, cultural-discursive arrangements) are transformed.
+
+## Methods toolbox
+
+- **Meeting facilitation**, **visual methods** (mapping, timelines), **focus groups**, **interviews**, **observation**, **document review**, **surveys** where appropriate.
+- **Reflective journals** for practitioners.
+- **Member validation** sessions before scaling changes.
+
+## Ethics and IRB
+
+AR can challenge IRB templates designed for fixed protocols. Use **amendment-friendly** designs, document **evolving consent**, and clarify **risks** when studying workplace power or community conflict.
+
+## When to use action research
+
+Use AR when **change and knowledge** must advance together, stakeholders can **sustain cycles**, and **contextual adaptation** matters more than controlled experimentation.
+
+Use **traditional experiments** when **randomization** and **treatment fidelity** are paramount and participatory control is inappropriate.
+
+Use **grounded theory** when the primary output is **substantive theory** rather than **intervention cycles**—though GT can inform AR data analysis.
+
+Use this skill for PAR design, cycle planning, validity arguments, or Kemmis-style participatory framing.

package/skills/apa-formatting/SKILL.md

@@ -0,0 +1,85 @@
+---
+name: apa-formatting
+description: Use when formatting academic work in APA 7th edition style — citations, references, headings, tables, and manuscript structure.
+---
+
+# APA 7th Edition Formatting (Qualitative-Friendly)
+
+APA style governs **manuscript structure, citation, reference entries, tables/figures, and mechanics**. Qualitative papers still benefit from strict reference hygiene even when narrative methods sections are long.
+
+## In-text citations
+
+- **Parenthetical:** (Glaser, 1998, p. 45) — use page/para/time for direct quotes.
+- **Narrative:** Glaser (1998) argued that…
+- **Two authors:** (Smith & Jones, 2020); three or more: (Lee et al., 2021) on subsequent cites.
+
+## Reference list rules
+
+- Alphabetize; hanging indent.
+- Journal article: Author, A. A. (Year). Title. *Journal*, *volume*(issue), pages. https://doi.org/…
+- Book: Author, A. A. (Year). *Title*. Publisher. https://doi.org/… (when available)
+
+## Heading levels (student/professional papers)
+
+Level 1: Centered, Bold, Title Case.
+
+Level 2: Flush Left, Bold, Title Case.
+
+Level 3: Flush Left, Bold Italic, Title Case.
+
+Level 4: Indented, Bold Title Case, ends with period. Text follows.
+
+Level 5: Indented, Bold Italic Title Case, ends with period.
+
+Match journal instructions if they override APA defaults.
+
+## Tables and figures
+
+Number sequentially; each has a **title** (italic for table title in APA tables) and note if needed. In qualitative work, tables may display **themes, exemplar quotes, coding frequencies** (if used)—ensure tables support claims without replacing analysis.
+
+## Manuscript sections (empirical article pattern)
+
+Title page; abstract; keywords; introduction; method; results/findings; discussion; references; appendices.
+
+Qualitative articles may label findings “Results” or “Findings” per venue—follow journal.
+
+## DOI formatting
+
+Prefer https://doi.org/10.xxxx/xxxxx. If no DOI, use stable URL; omit “Retrieved from” except when needed for archival pages (see APA manual for current arcane cases).
+
+## “et al.” rules (in-text)
+
+For works with 3+ authors, use first author + et al. after first citation (APA 7 simplifies from APA 6 in many cases—verify examples for your source type).
+
+## Block quotes (40+ words)
+
+Indent 0.5 in.; omit quotation marks; cite after final period.
+
+## Numbers
+
+Spell out zero through nine in general prose; numerals for 10+ and for units, ages, dates, statistics (many exceptions—consult APA tables).
+
+## GT methodology references (examples to copy-check against current edition)
+
+- Glaser, B. G., & Strauss, A. L. (1967). *The discovery of grounded theory: Strategies for qualitative research*. Aldine.
+- Glaser, B. G. (1978). *Theoretical sensitivity: Advances in the methodology of grounded theory*. Sociology Press.
+- Glaser, B. G. (1998). *Doing grounded theory: Issues and discussions*. Sociology Press.
+- Charmaz, K. (2014). *Constructing grounded theory* (2nd ed.). Sage.
+
+Always re-verify italics, capitalization, and publisher details against your copy of APA 7 and publisher records.
+
+## Qualitative-specific tips
+
+When citing **unpublished interviews**, follow APA interview citation formats; protect confidentiality in public documents. For reproduced excerpts in appendices, note IRB permissions if required.
+
+## Checklist
+
+- [ ] In-text matches reference list (spelling/year).
+- [ ] DOIs/URLs formatted consistently.
+- [ ] Heading levels logical; no skipped levels.
+- [ ] Tables/figures numbered with notes as needed.
+- [ ] Block quotes formatted and cited correctly.
+
+## References (starting points)
+
+- American Psychological Association. (2020). *Publication manual of the American Psychological Association* (7th ed.).

package/skills/case-study-methods/SKILL.md

@@ -0,0 +1,96 @@
+---
+name: case-study-methods
+description: "Use when designing or conducting case study research following Yin's multiple case study methodology or Stake's approach."
+---
+
+# Case Study Research Methods
+
+A **case study** examines a **bounded system** (a case) using multiple sources of evidence. It is suited to **“how”** and **“why”** questions where the researcher has little control over events, and context matters. Robert Yin and Robert Stake offer influential, partially complementary frameworks.
+
+## Yin’s logic of the case study
+
+Yin emphasizes **rigor through design**, not just data collection. A case is an empirical unit (e.g., organization, program, community, event) analyzed in depth.
+
+### Single vs multiple case designs
+
+- **Single case** — Justified when the case is **critical**, **unique**, **revelatory**, or **longitudinal**. Risk: weak generalization.
+- **Multiple case** — **Replication logic** (literal or theoretical): cases are chosen to predict similar results (literal replication) or contrasting results for predictable reasons (theoretical replication). Strengthens analytic generalization to theory, not statistical populations.
+
+### Holistic vs embedded
+
+- **Holistic** — The case as a single unit of analysis.
+- **Embedded** — Subunits (e.g., teams within a firm) analyzed; requires clarity about **how subunit findings aggregate** to the main case.
+
+## Case selection
+
+Cases should align with **research questions** and **propositions** (if used). For multiple cases, define **selection criteria** before data collection when possible to reduce cherry-picking. Document why cases were included or excluded.
+
+## Theoretical propositions
+
+Optional but powerful: explicit expectations guide **data collection priorities** and **analytic focus** (e.g., “Implementation success depends on leadership support”). Propositions differ from hypotheses in quant studies—they orient inquiry while allowing qualitative refinement.
+
+## Data collection in case studies
+
+Typical sources:
+
+- **Interviews** — Key actors, triangulated roles.
+- **Documents** — Minutes, policies, reports, communications.
+- **Direct observation** — Meetings, workflows, artifacts in use.
+- **Archival records** — Metrics, histories, legal filings.
+
+Yin stresses **converging evidence** (triangulation) and a **case study database** (organized raw data, logs, and analysis notes) separate from the final report.
+
+## Chain of evidence
+
+Readers should be able to trace **claims → displays → data**. Maintain:
+
+- Citation to specific sources (interview ID, document type, date).
+- Clear protocols linking evidence to assertions.
+- Analytic memos documenting interpretive leaps.
+
+## Analytic strategies (Yin)
+
+- **Pattern matching** — Compare empirically based patterns to predicted ones; may involve **rival explanations** ruled out by evidence.
+- **Explanation building** — Iterative narrative linking causes and outcomes; guard against researcher bias through triangulation and disconfirming evidence.
+- **Time-series / logic models** — When temporal ordering or program theory is central.
+
+## Cross-case synthesis (multiple cases)
+
+For each case, produce a **standalone description** and **within-case analysis**. Then conduct **cross-case comparison** using a uniform framework (e.g., tables of constructs). Look for **patterns**, **contrasts**, and **theoretical replication** support.
+
+## Stake’s approach: intrinsic vs instrumental
+
+### Intrinsic case study
+
+The case itself is of primary interest (e.g., a particular school). Findings emphasize **understanding that case**, not broad laws.
+
+### Instrumental case study
+
+The case is examined to **illuminate a broader issue** or refine a theory. The case remains particular, but the lens is **instrumental** to external understanding.
+
+### Collective case studies
+
+Several instrumental cases shed light on a phenomenon; resonates with Yin’s multiple-case design but retains Stake’s emphasis on **experience** and **interpretation**.
+
+## Stake’s emphases
+
+- **Thick portrayal** of cases and **reader naturalistic generalization**.
+- Attention to **emic views** and **interpretive rigor** rather than only deductive propositions.
+- **Progressive focusing** — Early field visits inform what matters next.
+
+## Yin vs Stake (practical contrast)
+
+| Feature | Yin | Stake |
+|---------|-----|--------|
+| Emphasis | Design, replication, propositions | Interpretation, portrayal, experience |
+| Generalization | Analytic generalization to theory | Naturalistic, reader-led |
+| Rigor | Database, chain of evidence, methods triangulation | Interpretive discipline, reflexivity |
+
+## Reporting
+
+- Clear **case boundaries** and **context**.
+- **Methods** section with data sources and analysis steps.
+- **Within-case** chapters/sections before **cross-case** synthesis.
+- Explicit **limitations** and **rival interpretations** considered.
+
+Use this skill when the user designs multi-site evaluations, organizational studies, or compares Yin and Stake for dissertation methodology.