In mediation analysis, when the mediator is included, the direct effect of the predictor on the outcome typically

• A. Becomes larger
• B. Becomes smaller or non-significant
• C. Remains exactly the same
• D. Reverses sign

Answer: (B)

When you include the mediator in mediation analysis, part of the predictor’s influence on the outcome is carried through that mediator. The total effect is the sum of the direct effect and the indirect effect via the mediator. Once you account for the mediator, the direct effect is what's left after removing the indirect path, so it tends to shrink in magnitude. If the mediator fully explains the relationship, the direct effect can become non-significant. It wouldn’t typically get larger, stay exactly the same, or reverse sign in a standard mediation situation.