What to decide about missing data before regression
Missingness affects the analysis population, estimand, uncertainty, and credibility long before a model is fitted.
Missingness affects the analysis population, estimand, uncertainty, and credibility long before a model is fitted.
Describe the observation process
Summaries by variable, time, group, and relevant predictors are more informative than one overall missing percentage.
State what makes the method plausible
Complete-case analysis, likelihood approaches, weighting, and multiple imputation rely on different assumptions. None repairs unavailable information automatically.
Plan sensitivity to untestable assumptions
When conclusions could change under plausible departures, show that dependence rather than presenting one analysis as certain.
Failure modes to check
- Choosing a method because it is familiar rather than because it matches the estimand and data structure.
- Treating assumptions as a generic pass/fail checklist instead of evidence about a specific model.
- Changing the workflow after seeing results without recording the decision or its effect on uncertainty.
- Reporting software output without tracing the claim to final code, diagnostics, and limitations.
Continue the decision path
Explore the related statistical consulting service, or discuss the decision in your own study.
Related decision guides
Methodological note
This guide provides general educational information. The appropriate decision depends on the study question, design, data, evidence, ethics, and institutional requirements.