Partial least squares structural equation modeling for composite-based path models with reflective or formative indicators.

Covariance-based structural equation modeling and confirmatory factor analysis, fit via maximum likelihood.

Generalized Structured Component Analysis -- an alternating least squares component-based alternative to PLS-SEM.

Multiple linear regression with the full standard diagnostic output set.

Binary logistic regression for categorical outcome variables.

Nonparametric resampling for significance testing, available across every analysis in ResearchPLS.

Bootstrap-based testing of indirect effects through one or more mediating variables.

Interaction-effect testing for continuous or categorical moderator variables.

Conditional process analysis combining mediation and moderation in a single model.

Tests whether path coefficients differ significantly between predefined groups.

Configural, compositional, and full invariance testing -- a required precondition for a valid MGA.

Importance-Performance Map Analysis -- contrasts a construct's total effect against its average performance.

Necessary Condition Analysis -- identifies whether a predictor is a necessary (not just sufficient) condition for an outcome.

Finite mixture segmentation that detects unobserved heterogeneity in a PLS-SEM sample.

PLS prediction-oriented segmentation for identifying latent classes in the structural model.

A statistical correction for endogeneity when a valid instrumental variable is not available.

Tests whether a block of indicators is better modeled as reflective or formative.

Compares competing model specifications using in-sample information criteria.

A priori sample-size planning and post-hoc power sensitivity for a fitted structural model.