Qualitative Coding Methodology Reference

The Coding Hierarchy

Theme (interpretive pattern)
  └── Category (grouping)
        └── Code (label)
              └── Segment (data excerpt)
                    └── Source (document/media)

2. Saldana's Coding Methods Taxonomy

Reference: Saldana, J. (2021). The Coding Manual for Qualitative Researchers (4th ed.). SAGE.

First Cycle Coding (Initial pass — applied directly to data)

Grammatical Methods

Elemental Methods (foundational, widely used)

Affective Methods (subjective experience)

Literary and Language Methods

Exploratory Methods (preliminary investigation)

Procedural Methods

Themeing the Data (4th edition addition)

Second Cycle Coding (Analytical refinement of first cycle codes)

Second cycle methods work with the codes themselves, not raw data:

Implementation Guidance for QualCoder v2

First Cycle → User applies codes to data segments
              QualCoder supports: code creation, hierarchy, color, memo
              AI can: suggest codes, auto-code with provisional lists

Second Cycle → User reorganizes codes into categories/themes
               QualCoder supports: drag-drop hierarchy, merge codes, code reports
               AI can: suggest groupings, identify patterns, co-occurrence analysis

3. Major Qualitative Frameworks

Grounded Theory (GT)

Origins: Glaser & Strauss (1967). Three major schools:

Core GT Procedures:

• Constant comparison: Continuously compare data↔codes, codes↔codes, codes↔categories
• Theoretical sampling: Let emerging theory guide further data collection
• Theoretical saturation: Stop when new data adds no new insights
• Memo writing: Record analytical thinking throughout

QualCoder implications: Support iterative coding, code merging, code frequency tracking, memo linking, saturation indicators.

Braun & Clarke's Reflexive Thematic Analysis (RTA)

Reference: Braun, V. & Clarke, V. (2006, 2022). Most cited qualitative method (190,000+ citations).

Six Phases:

TA as a family of methods (Braun & Clarke 2022 typology):

• Coding reliability TA: Multiple coders, ICR, codebook-driven
• Codebook TA: Structured codebook, more deductive
• Reflexive TA: Researcher subjectivity as resource, iterative

2024 Update: Reflexive Thematic Analysis Reporting Guidelines (RTARG) published as alternative to COREQ/SRQR.

Content Analysis

Reference: Krippendorff, K. (2018). Content Analysis: An Introduction to Its Methodology (4th ed.).

• Systematic, replicable analysis of text
• Quantitative orientation: code frequencies, co-occurrence matrices
• Strong emphasis on inter-coder reliability
• QualCoder support: Code frequency reports, co-occurrence matrices, coding comparison

Framework Analysis (Ritchie & Spencer)

• Developed at UK National Centre for Social Research
• Matrix-based: cases (rows) × themes (columns)
• Good for applied policy research with specific questions
• Supports both inductive and deductive approaches
• QualCoder support: Case management, cross-case comparison, matrix displays

Interpretative Phenomenological Analysis (IPA)

Reference: Smith, Flowers & Larkin (2009).

• Focus on lived experience and meaning-making
• Idiographic: each case analyzed in depth before cross-case patterns
• Double hermeneutic: researcher interprets participant's interpretation
• Small samples (typically 3-6 participants)
• QualCoder support: Case-by-case coding, individual case memos, cross-case themes

Discourse Analysis

• Analyzes how language constructs social reality
• Focus on rhetorical strategies, power dynamics, positioning
• Codes are about linguistic function, not content
• QualCoder support: Fine-grained text coding, verbal exchange coding

Narrative Analysis

• Analyzes story structures (setting, characters, plot, resolution)
• Codes story elements and narrative arcs
• QualCoder support: Narrative coding, temporal ordering, case linking

4. Quality & Rigor Criteria

Lincoln & Guba's Trustworthiness (1985)

The gold standard for qualitative research quality:

Inter-Coder Reliability (ICR)

When multiple coders work on same data:

Interpretation (Landis & Koch, 1977):

• < 0.00 = no agreement
• 0.21–0.40 = fair
• 0.41–0.60 = moderate
• 0.61–0.80 = substantial
• 0.81–1.00 = near-perfect

Best practices:

• Calculate ICR throughout project, not just at end
• Train coders on codebook before independent coding
• Use at least 10-20% of data for reliability testing
• Document and resolve disagreements through discussion

Codebook Development

A codebook should include for each code:

1. Code name — Short, descriptive label
2. Definition — Clear boundary of what the code means
3. Inclusion criteria — When to apply this code
4. Exclusion criteria — When NOT to apply
5. Example — Typical data excerpt
6. Counter-example — Data that looks similar but doesn't fit

5. AI-Assisted Qualitative Coding

Current Landscape (2024-2026)

Commercial QDA tools with AI:

• ATLAS.ti 25: GPT-powered auto-coding, AI Lab for intent-based coding
• NVivo 15: ML-based autocode, AI summarization, suggest codes
• MAXQDA 24/Tailwind: AI-assisted coding and theme generation

Three Emerging Paradigms

1. Rigorous Manual Analysis — Traditional hand-coding, transparent audit trail
2. AI-Augmented Manual Analysis — AI suggests, human decides (QualCoder v2 approach)
3. AI-Native Structured Analysis — AI-first but maintaining academic rigor

QualCoder v2's AI Philosophy

QualCoder uses AI transparently:

• Prompts are visible and modifiable by the researcher
• AI suggestions require human approval (never auto-applied)
• Full audit trail of AI-generated vs. human-generated codes
• Researcher remains the analytical authority

Best Practices for AI-Assisted Coding

DO:

• Use AI for initial code suggestions on large datasets
• Let AI identify potential patterns for human review
• Use AI to check coding consistency across documents
• Maintain audit trail: save prompts, AI outputs, human decisions
• Validate AI coding against manual coding of a subset
• Use provisional coding with AI-generated start lists

DON'T:

• Auto-accept AI-generated codes without review
• Use AI as sole coder (no human verification)
• Assume AI understands context, power dynamics, or cultural nuance
• Skip reflexivity because "the AI did it"
• Claim methodological rigor without human analytical engagement

Prompt Engineering for Qualitative Coding

When implementing AI coding features, prompts should:

1. Specify the coding method (e.g., "Use descriptive coding to label topics")
2. Provide codebook context (existing codes, definitions)
3. Include research question for structural/theoretical alignment
4. Request rationale alongside code suggestions
5. Ask for confidence levels to guide human review priority

6. Feature-to-Methodology Mapping

When implementing QualCoder features, map them to methodology:

7. Glossary of Key Concepts

8. Key References

Foundational Texts

• Saldana, J. (2021). The Coding Manual for Qualitative Researchers (4th ed.). SAGE.
• Braun, V. & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2), 77-101.
• Braun, V. & Clarke, V. (2022). Thematic Analysis: A Practical Guide. SAGE.
• Charmaz, K. (2014). Constructing Grounded Theory (2nd ed.). SAGE.
• Corbin, J. & Strauss, A. (2015). Basics of Qualitative Research (4th ed.). SAGE.
• Glaser, B.G. & Strauss, A.L. (1967). The Discovery of Grounded Theory. Aldine.
• Krippendorff, K. (2018). Content Analysis: An Introduction to Its Methodology (4th ed.). SAGE.
• Lincoln, Y.S. & Guba, E.G. (1985). Naturalistic Inquiry. SAGE.
• Smith, J.A., Flowers, P. & Larkin, M. (2009). Interpretative Phenomenological Analysis. SAGE.
• Ritchie, J. & Spencer, L. (1994). Qualitative data analysis for applied policy research. In Analyzing Qualitative Data.

Recent Developments

• Braun, V. & Clarke, V. (2024). RTARG: Reflexive Thematic Analysis Reporting Guidelines. Palliative Medicine.
• O'Connor, C. & Joffe, H. (2020). Intercoder reliability in qualitative research. International Journal of Qualitative Methods.
• Lumivero (2025). The State of AI in Qualitative Research.

9. Decision Guide: Which Coding Method?

START
  │
  ├── Do you have pre-existing codes from theory?
  │     YES → Provisional Coding / Hypothesis Coding
  │     NO ↓
  │
  ├── Is this your first pass through the data?
  │     YES ↓
  │     │   ├── Large dataset, need overview? → Holistic Coding
  │     │   ├── Want participant voice? → In Vivo Coding
  │     │   ├── Want topic inventory? → Descriptive Coding
  │     │   ├── Studying processes/actions? → Process Coding
  │     │   └── Grounded theory study? → Initial Coding (line-by-line)
  │     │
  │     NO → Second Cycle Methods ↓
  │           ├── Need to group codes? → Pattern Coding / Focused Coding
  │           ├── Need relationships? → Axial Coding
  │           └── Building theory? → Theoretical Coding
  │
  ├── What aspect interests you?
  │     ├── Emotions → Emotion Coding
  │     ├── Values/beliefs → Values Coding
  │     ├── Conflicts → Versus Coding
  │     ├── Program effectiveness → Evaluation Coding
  │     ├── Stories → Narrative Coding
  │     └── Performance/identity → Dramaturgical Coding
  │
  └── Multiple coders?
        YES → Codebook approach + ICR measurement
        NO → Reflexive approach + memo trail