ux-research-methods
UX research methodology: usability testing protocols, user interviews, contextual inquiry, surveys, card sorting, and diary studies, plus synthesis frameworks including JTBD, journey mapping, and affinity diagramming. Use when planning a study, writing a discussion guide, or analyzing research data.
- License 许可
- MIT
- 最近更新
- 2026-11-10
- Stars
- ★ 0
- Forks
- 0
01 前言 Front Matter
- name
- ux-research-methods
- description
- UX research methodology: usability testing protocols, user interviews, contextual inquiry, surveys, card sorting, and diary studies, plus synthesis frameworks including JTBD, journey mapping, and affinity diagramming. Use when planning a study, writing a discussion guide, or analyzing research data.
- allowed-tools
- Read, Grep, Glob
02 资源文件 2
- 参考文档references/analysis-frameworks.md39.3 KB
- 参考文档references/research-protocols.md33.6 KB
03 SKILL.md 正文
UX Research Methods
Research Philosophy
Great design is grounded in evidence, not assumption. UX research exists to reduce the risk of building the wrong thing and to illuminate the gap between designer intent and user reality. Every design decision should trace back to user evidence — whether from direct observation, behavioral data, or validated heuristics.
The Research Hierarchy (adapted from NNG Group)
- Behavioral data (what users do) > Attitudinal data (what users say)
- Qualitative research (why) informs quantitative research (how much)
- Generative research (discover opportunities) precedes evaluative research (validate solutions)
- Continuous discovery (ongoing) outperforms project-based research (periodic)
Method Selection
| Question type | Method | Typical sample |
|---|---|---|
| Workflows, workarounds, unarticulated needs | Contextual inquiry | 4-8 |
| Motivations, mental models | Semi-structured interviews | 5-8 per segment |
| Behavior over time | Diary study | 10-15 (dropout) |
| Information architecture | Card sorting / tree testing | 15-30 / 50+ |
| Can users complete the task? | Moderated usability test | 5 finds ~85% of issues |
| Scaled task evidence | Unmoderated remote test | 10-20 |
| Expert review, no users | Heuristic evaluation (3-5 evaluators) or cognitive walkthrough | n/a |
| How many, how much | Surveys / analytics | 100+ (400+ for segments) |
| Which variant performs better | A/B test | traffic for significance, define MDE first, 1-2 weeks min |
Research Operations
Planning a Research Study
- Define research questions (3-5 focused questions per study)
- Select appropriate method(s) based on questions and constraints
- Write a 1-page research plan: goals, method, participants, timeline, deliverables
- Recruit representative participants — screen rigorously for target criteria
- Prepare materials: discussion guide, task scenarios, consent forms
- Pilot test the protocol with 1-2 internal participants
Participant Recruitment
- Screen for behavioral criteria, not just demographics
- Include users with disabilities (minimum 1 per study for inclusive insights)
- Offer fair compensation respecting participants' time and expertise
- Maintain a participant panel for longitudinal research
- Avoid "professional testers" — seek real target users
Ethical Research Practice
- Obtain informed consent with clear explanation of data usage
- Allow participants to withdraw at any time without consequence
- Protect participant privacy and anonymize data
- Never manipulate or deceive participants
- Share findings back with participants when possible
AI-Assisted Analysis Rules
Treat LLM themes and codes as a first pass only: review 100% of AI output for the first three studies, then spot-check 30-50%; trace every insight to a participant quote; the researcher authors the report; anonymize transcripts and get consent before any upload; disclose AI assistance.
Routing
- Choosing and running a study — the full method selection matrix (discovery, evaluative, quantitative), moderator scripts, interview guides, contextual inquiry, diary studies, card sorting and tree testing, A/B protocols, survey design, screener and consent templates, research ops: read
references/research-protocols.md. - Making sense of what you collected — affinity diagramming, thematic coding, Jobs-to-Be-Done, journey maps, personas, SUS and significance tests, mixed methods, repositories, atomic research, AI-assisted synthesis with prompts and ethics: read
references/analysis-frameworks.md.
Cross-Referencing
- For heuristic evaluation methodology, reference
nng-ux-heuristics. - For accessibility research requirements, reference
accessibility-inclusive-design. - For mobile-specific testing considerations, reference
mobile-ux-design. - For design system validation, reference
design-systems-architecture. - For UX metrics and measurement frameworks, reference
ux-metrics-measurement. - For AI-specific research methods, reference
agentic-ai-generative-ux. - Turning findings into design direction with structured critique:
design-critique-case-studies
Key Sources
- NNG Group: "Which UX Research Methods" and research methodology articles
- Portigal, S. (2013). "Interviewing Users"
- Krug, S. (2014). "Rocket Surgery Made Easy"
- Sauro, J. & Lewis, J. (2016). "Quantifying the User Experience"
- Arxiv: GenAI for UX Research (Dec 2025)
- Arxiv: EvAlignUX — Evaluating AI Alignment in UX Research (Sep 2024)
04 同仓库更多 Skills
no-description
暂无简介
good-skill
Fixture skill that satisfies every rule.
broken-link
Links to a reference file that does not exist.
Bad_Name
Name violates the kebab-case rule and does not match the directory.
skill-authoring
Write, review, and validate portable agent skills in SKILL.md format. Use when creating a new skill, editing an existing one, or asked whether a skill follows the Agent Skills spec.
weigh-bags
Record scale readings for each bag — empty (tare), outbound, current, final — and reconcile them against the itemised inventory to find unaccounted weight. Use when the user gives a bag weight, says how heavy something is, weighs their luggage, or asks how close to the limit they are. Appends to bags.yaml and runs scripts/reconcile.py.