🧠 UX Research & Design · Research Methods

Synthesize research without bias collapse

Affinity mapping with evidence tagging separating observation from interpretation.

advanced~40 minUX ResearchersProduct DesignersDesign Leads

Steps

1Extract observations verbatim onto atomic notes with source refs
2Cluster by pattern, NOT by person or session order
3Name clusters descriptively before interpretively
4Tag insight strength: single-source vs multi-user convergence
5Separate insights from recommendations explicitly
6Store the map; future decisions cite evidence lineage

Common Pitfalls

▲Loudest stakeholder's pet theory shaping clusters
▲Insights so general they justify any decision

Commands

Install with skills CLI

$ npx skills add aniruddhaadak80/skills --skill research-methods-synthesis-affinity

Install globally

$ npx skills add aniruddhaadak80/skills --skill research-methods-synthesis-affinity -g

Tags

#synthesis#analysis#workshop#ux-research#research-methods

Install this skill

$ npx skills add aniruddhaadak80/skills --skill research-methods-synthesis-affinity -a opencode

SKILL.md

# Synthesize research without bias collapse

> Affinity mapping with evidence tagging separating observation from interpretation.

## When to Use This Skill

Affinity mapping with evidence tagging separating observation from interpretation.

## Steps

1. Extract observations verbatim onto atomic notes with source refs
2. Cluster by pattern, NOT by person or session order
3. Name clusters descriptively before interpretively
4. Tag insight strength: single-source vs multi-user convergence
5. Separate insights from recommendations explicitly
6. Store the map; future decisions cite evidence lineage

## Common Pitfalls

- Loudest stakeholder's pet theory shaping clusters
- Insights so general they justify any decision

View source on GitHub ↗

Task-based sessions with neutral scripting, think-aloud discipline, and severity-rated findings.

🧠 UX Research & Design·~35m

Open →

Write surveys without garbage-inintermediate

Question construction, ordering effects, and response-scale discipline producing analyzable data.

🧠 UX Research & Design·~30m