Understand every analysis

Estimate population-average associations for clustered or repeated responses.

Model subject-specific associations for repeated binary or count outcomes.

Use when the outcome/model family is not yet clear or the needed model is not implemented.

Compare a continuous outcome between exactly two independent groups without assuming equal variances.

Compare two independent group means when equal variances are defensible.

Compare two measurements from the same unit, person, or matched pair.

Compare one continuous variable against a known reference value.

Compare a binary outcome proportion against a specified reference proportion.

Compare two independent groups when rank/ordinal behavior is more appropriate than means.

Compare paired ordinal or non-normal continuous measurements.

Compare three or more independent groups using ranks.

Compare three or more paired or repeated measurements using ranks.

Compare a continuous outcome across three or more groups.

Estimate main effects and interactions for multiple categorical factors.

Compare three or more repeated measurements from the same unit.

Compare adjusted group means while controlling continuous covariates.

Measure linear association between two continuous variables.

Describe how two numeric variables vary together in their original units.

Measure monotone association using ranks.

Measure ordinal or monotone association with a rank-based coefficient robust to small samples.

Measure association between one continuous variable and one binary variable.

Measure association between a continuous variable and a categorical grouping variable.

Measure association between two continuous variables after linear adjustment for covariates.

Test association between two categorical variables.

Analyze a small 2x2 contingency table.

Compare paired binary responses.

Model a continuous outcome with one or more predictors.

Model a binary outcome with one or more predictors.

Fit a linear regression when predictors are correlated and coefficient shrinkage is useful.

Fit a linear regression with coefficient shrinkage and possible variable selection.

Model count outcomes or rates.

Reduce several correlated numeric features to a smaller set of variance directions.

Explore whether observations form compact groups across several numeric features.

Estimate latent factors that explain correlations among observed numeric variables.

Assess the internal consistency of multiple items intended to measure one construct.

Explore nested group structure without fixing a cluster count in advance.

Evaluate how a numeric score separates a binary outcome across thresholds.

Plan sample size or estimate power for a mean comparison.

Compare survival curves between groups.

Model time-to-event outcomes with covariates.

Model continuous outcomes with repeated or clustered observations.

Measure agreement between two categorical ratings of the same observations.

Measure categorical agreement across three or more raters.

Assess absolute agreement among numeric ratings or repeated measurements.

Test whether a two-group mean difference is small enough to be practically equivalent.

Test whether one numeric predictor changes the association of another with an outcome.

Estimate an indirect association through a numeric mediator.

Model an unordered categorical outcome with three or more levels.