conjoint-diagnostics

Conjoint diagnostics systematically evaluate attribute selection, randomization, and task design to identify estimation and external validity issues. The skill applies a structured checklist covering design, measurement error, and interpretation to real or simulated studies.

AMCE estimand clarity is checked by evaluating reference levels, standard errors, and subgroup analysis. The diagnostic framework guides estimation and interpretation to ensure conjoint study results are transparent and correctly specified.

Measurement error diagnostics assess inter-rater reliability (IRR), swapping error, and bias correction methods. The framework applies sensitivity analyses to evaluate how measurement inconsistencies impact conjoint experimental results.

Yes, the diagnostic checklist applies to both real and simulated conjoint studies. It systematically evaluates profile restrictions, sampling considerations, forced-choice validity, and reporting contexts across simulated attributes and levels.

External validity checks address profile distributions and the gap between forced-choice tasks and real-world behavior. The diagnostic framework evaluates reporting transparency to ensure conjoint experimental interpretations match real-world contexts.

Conjoint diagnostics flag multiple testing issues to caution against misinterpreting subgroup analysis. The framework enforces standard error evaluation and estimand clarity, providing a final evaluation summary of interpretation risks.