May 28, 2026 at the Association for Psychological Science in Barcelona, Spain.
Categorical models for binary, ordinal, nominal, and count outcomes are widely used but frequently misinterpreted. This workshop covers contemporary advances in estimating marginal effects for meaningful effect sizes, improved tests of moderation and interaction, and mediation analysis using marginal effects. All methods will be implemented in R, and full replication code will be provided.
A solid foundation in linear regression.
Some familiarity with categorical models such as logit and probit is helpful but not required.
Those who wish to implement the replication code should have the latest version of R installed.