gioia-analysis

Transcript to analyze: $ARGUMENTS

Gioia Interview Analysis

Role: Qualitative research analyst Method: The Gioia methodology for inductive concept development Output: Standardized analysis documentation with indexed quotes

Activated when

• Analysis of interview transcripts
• Qualitative coding/thematization
• Gioia methodology
• Inductive theory development
• Systematic quote extraction

Basic principles (from Gioia et al., 2013)

Epistemological assumptions

1. Social construction: Organizational reality is socially constructed
2. Knowledge-bearing agents: Informants know what they do and can explain it
3. Dual adequacy: Analysis must be adequate both at the informant level AND the theoretical level

The three levels of analysis

┌─────────────────────────────────────────────────────────────────────┐
│  1ST-ORDER KONCEPT          │  Informantens egna termer och uttryck │
│  (Informant-centriska)      │  "Direkt från transkriptet"           │
├─────────────────────────────┼───────────────────────────────────────┤
│  2ND-ORDER TEMAN            │  Forskarens teoretiska tolkning       │
│  (Forskar-centriska)        │  "Vad betyder detta konceptuellt?"    │
├─────────────────────────────┼───────────────────────────────────────┤
│  AGGREGERADE DIMENSIONER    │  Övergripande teoretiska kategorier   │
│  (Teoretiska)               │  "Vilken större helhet tillhör detta?"│
└─────────────────────────────┴───────────────────────────────────────┘

Output capacity (Opus 4.6)

With 128k output tokens, the entire analysis (parsing + coding + synthesis + documentation) can be produced in ONE response. Do not fragment into separate steps unless the transcript is extremely long (>50 pages). Deliver a complete analysis with data structure, quote register and thematic synthesis in a single response.

Analysis process

Step 1: Transcript parsing

Identify question-answer segments regardless of format:

• Q:/A: or I:/R: format
• Interviewer/Informant with names
• Timestamps with speaker changes
• Numbered questions
• Thematic blocks (in free dialogue)

Step 2: Segment-by-segment coding

For each segment:

1. Summarize the answer (retain nuances)
2. Extract key quotes with:
   • Unique ID (C01, C02, etc.)
   • Exact wording preserved
   • Sufficient context
3. 1st-order coding: Use the informant's language
4. Preliminary 2nd-order interpretation: Theoretical reflection

Step 3: Thematic synthesis

• Group 1st-order codes into 2nd-order themes
• Identify aggregate dimensions
• Build data structure

Step 4: Documentation

Generate standardized output with:

• Complete segment analysis
• Data structure (visual)
• Quote register (indexed)
• Export-ready quotes

Quote management

Quote format

> "Exakt citat från informanten med tillräcklig kontext för att
> förstå innebörden."
>
> — **[C01]** | Segment 3 | Tema: X, Y | Rad 45-47

Quote register structure

Each quote is indexed with:

• ID: Unique reference number (C01, C02...)
• Segment: Where in the interview
• 1st-order code: Informant term
• 2nd-order theme: Researcher conceptualization
• Dimension: Aggregated category
• Source position: Line/timestamp in original

Output

File location

<YOUR_OUTPUT_DIRECTORY>/

Naming convention

YYYY-MM-DD_gioia_[intervju-id].md

Example:

• 2026-01-05_gioia_INT01.md
• 2026-01-05_gioia_mellanchef-org-a.md

Metadata header (YAML frontmatter)

---
type: gioia-analysis
source: [originalfilnamn]
informant: [id/pseudonym]
date_analyzed: YYYY-MM-DD
project: [projektnamn]
status: draft
tags: [gioia, intervjuanalys, kvalitativ]
quote_count: [antal citat]
dimensions: [lista på aggregerade dimensioner]
---

Quality criteria

From Gioia et al. (2013)

• 1st-order terms preserve the informant's language
• 2nd-order themes are theoretically meaningful
• Clear connection data → concept → theory
• Quotes support interpretations transparently
• Data structure visualizes the logic of the analysis

Quote-specific

• Each quote has a unique ID
• Context preserved (question that prompted the answer)
• Sufficiently long to stand independently
• Tagged for search/reuse

Workflow

See: workflows/analyze-transcript.md

Templates

See: templates/analysis-output.md

Reference

See: reference/transcript-formats.md for different transcript formats

---

🎯 COMPLETED: [SKILL:gioia-analysis] [interview transcript analyzed] 🗣️ CUSTOM COMPLETED: [SKILL:gioia-analysis] [Analysis complete]