conjoint-design

Design and analyze conjoint experiments with attribute architecture and power analysis.

39|1|Updated Jan 21, 2026
One-click install
npx skills add https://github.com/scdenney/open-science-skills --skill conjoint-design
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: conjoint-design
Source: https://github.com/scdenney/open-science-skills/tree/main/plugin/skills/conjoint-design
Command: npx skills add https://github.com/scdenney/open-science-skills --skill conjoint-design

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Specialized logic for designing conjoint and factorial vignette experiments, enabling researchers to plan and execute rigorous, policy-relevant attribute-based studies.

Core Features & Use Cases

  • Attribute architecture guidance: ensure orthogonality, randomization strategy, and valid design constraints.
  • Power analysis and estimation: AMCE/MM estimation, interaction considerations, and robust statistical best practices.
  • Design variant guidance: choose between paired-choice, rating, and factorial vignette formats; draft pre-analysis plans and regression specifications.
  • Use Case: Design a new conjoint study with 5 attributes and 4 levels per attribute, compare design variants, and prepare a preregistered analysis plan.

Quick Start

Specify your attributes and levels, then request a design and power analysis for your conjoint study.

Frequently Asked Questions about conjoint-design

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I design a conjoint experiment with multiple attributes and levels?

To design a conjoint experiment, you specify your attributes and levels, then request guidance on attribute architecture to ensure orthogonality, randomization strategy, and valid design constraints for rigorous policy-relevant studies.

What is the difference between paired-choice, rating, and factorial vignette conjoint designs?

Conjoint design variants differ by respondent task format: paired-choice forces a binary comparison, rating scales measure preference intensity, and factorial vignettes evaluate single profiles, with guidance available to select the best fit for your study.

How do I calculate statistical power for a conjoint experiment?

Power analysis for conjoint experiments involves calculating the required sample size based on your attribute architecture, estimating AMCE or MM effects, and applying robust statistical best practices to ensure reliable policy inference.

Can I draft a pre-analysis plan for a factorial vignette experiment?

Yes, you can draft a pre-analysis plan for factorial vignette experiments by specifying regression specifications, defining AMCE estimation methods, and preregistering interaction considerations before data collection begins.

How do I ensure orthogonality when randomizing conjoint attribute levels?

Ensuring orthogonality in conjoint attribute randomization requires applying valid design constraints that prevent attribute correlation, allowing unbiased estimation of marginal component effects across all profile dimensions.