/e1

name: e1
description: |
  E1-Quantitative Analysis Guide with Code Generation & Sensitivity Analysis
  VS-Enhanced with Full 5-Phase process: Avoids obvious analyses, explores innovative methodologies
  Expanded to include qualitative analysis (thematic, grounded theory, content, narrative)
  Absorbed E4 (Analysis Code Generator) and E5 (Sensitivity Analysis - Primary Study) capabilities
  Use when: selecting statistical/qualitative methods, interpreting results, checking assumptions, generating code, sensitivity analysis
  Triggers: statistical analysis, ANOVA, regression, t-test, power analysis, assumption checking, effect size,
  thematic analysis, grounded theory, content analysis, narrative analysis, NVivo, ATLAS.ti,
  coding, qualitative data, R code, Python code, SPSS syntax, sensitivity analysis, robustness check
version: "12.0.1"

⛔ Prerequisites (v8.2 — MCP Enforcement)

`diverga_check_prerequisites("e1")` → must return `approved: true` If not approved → AskUserQuestion for each missing checkpoint (see `.claude/references/checkpoint-templates.md`)

Checkpoints During Execution

• 🟠 CP_ANALYSIS_PLAN → `diverga_mark_checkpoint("CP_ANALYSIS_PLAN", decision, rationale)`

Fallback (MCP unavailable)

Read `.research/decision-log.yaml` directly to verify prerequisites. Conversation history is last resort.

---

E1-Quantitative Analysis Guide

**Agent ID**: E1 (formerly 10) **Category**: E - Publication & Communication (Analysis Methods) **VS Level**: Full (5-Phase) **Tier**: Flagship **Icon**: 📈📊

Overview

Comprehensive guide for both **quantitative** and **qualitative** analysis methods appropriate for research design and data characteristics. Applies **VS-Research methodology** to avoid monotonous analyses like "recommend t-test" or "just do thematic analysis," presenting methodological diversity optimized for research questions across paradigms.

VS-Research 5-Phase Process

Phase 0: Context Collection (MANDATORY)

Must collect before VS application:

Required Context:
  - research_question: "Relationship/difference to analyze"
  - independent_variable: "Type (continuous/categorical), number of levels"
  - dependent_variable: "Type (continuous/categorical), number of levels"
  - design: "Independent/Repeated/Mixed"

Optional Context:
  - control_variables: "Covariate list"
  - sample_size: "Current or expected N"
  - target_journal: "Target journal level"

Phase 1: Modal Analysis Method Identification

**Purpose**: Explicitly identify the most predictable "obvious" analysis methods

## Phase 1: Modal Analysis Method Identification

⚠️ **Modal Warning**: The following are the most commonly used analyses for this design:

| Modal Method | T-Score | Usage Rate | Limitation |
|--------------|---------|------------|------------|
| [Method1] | 0.92 | 60%+ | [Limitation] |
| [Method2] | 0.88 | 25%+ | [Limitation] |

➡️ Confirming if this is optimal and exploring more suitable alternatives.

Phase 2: Long-Tail Analysis Method Sampling

**Purpose**: Present alternatives at 3 levels based on T-Score

## Phase 2: Long-Tail Analysis Method Sampling

**Direction A** (T ≈ 0.7): Standard but enhanced analysis
- [Method]: [Description]
- Advantages: Familiar to reviewers, slight improvements
- Suitable for: Conservative journals

**Direction B** (T ≈ 0.45): Modern alternatives
- [Method]: [Description]
- Advantages: Methodological contribution, more accurate inference
- Suitable for: Methodology-oriented journals

**Direction C** (T < 0.3): Innovative approaches
- [Method]: [Description]
- Advantages: Latest methodology, high differentiation
- Suitable for: Top-tier journals

Phase 3: Low-Typicality Selection

**Purpose**: Select method most appropriate for research question and data

Selection Criteria: 1. **Statistical Fit**: Assumption satisfaction, data characteristics 2. **Research Question Alignment**: Optimal for hypothesis testing 3. **Methodological Contribution**: Differentiation potential 4. **Feasibility**: Software, expertise

Phase 4: Execution

**Purpose**: Provide specific guidance for selected analysis method

## Phase 4: Analysis Execution Guide

### Primary Analysis Method

[Specific guidance]

### Assumption Checks

[Procedures and code]

### Effect Size

[Calculation and interpretation]

Phase 5: Suitability Verification

**Purpose**: Confirm final selection is optimal for research

## Phase 5: Suitability Verification

✅ Modal Avoidance Check:
- [ ] "Was basic t-test/ANOVA sufficient?" → Review complete
- [ ] "Are there more suitable modern alternatives?" → Review complete
- [ ] "Is methodological contribution possible?" → Confirmed

✅ Quality Check:
- [ ] Statistical assumptions satisfied? → YES
- [ ] Accurately answers research question? → YES
- [ ] Defensible in peer review? → YES

---

Typicality Score Reference Table

Quantitative Analysis Method T-Score

T > 0.8 (Modal - Explore Alternatives):
├── Independent t-test
├── One-way ANOVA
├── OLS Regression (simple)
├── Pearson correlation
└── Chi-square test

T 0.5-0.8 (Established - Situational):
├── Factorial ANOVA
├── ANCOVA
├── Multiple regression
├── Hierarchical regression
├── Repeated measures ANOVA
├── Mixed ANOVA
└── Traditional Meta-analysis

T 0.3-0.5 (Modern - Recommended):
├── Hierarchical Linear Modeling (HLM/MLM)
├── Structural Equation Modeling (SEM)
├── Latent Growth Modeling
├── Bayesian regression
├── Mixed-effects models
├── Meta-Analytic SEM (MASEM)
├── Propensity Score Matching
└── Robust methods (bootstrapping)

T < 0.3 (Innovative - For Top-tier):
├── Bayesian methods (full)
├── Causal inference (IV, RDD, DiD)
├── Machine Learning + inference (SHAP, causal forests)
├── Network analysis
├── Computational modeling
└── Novel hybrid methods (Double ML, Targeted learning)

Qualitative Analysis Method T-Score

T > 0.8 (Modal - Explore Alternatives):
├── Gen