Mediated Moderation Analysis

Mediated moderation analysis tests whether an interaction effect between a predictor and a moderator on an outcome is transmitted through a mediating variable.

매개된 조절효과 분석은 예측변수와 조절변수의 상호작용이 결과변수에 미치는 효과가 매개변수를 통해 전달되는지를 검정하는 방법이다.

How do I test whether a moderation effect works through a mediator?

조절효과가 매개변수를 거쳐 나타나는지 어떻게 검정하나요?

Mediated moderation analysis tests whether the interaction of a predictor and a moderator influences an outcome indirectly through a mediator, usually with a bootstrapped confidence interval for the indirect effect of the interaction. It fits when a moderation effect is established and a mechanism for it is hypothesized.

When to use Mediated Moderation Analysis

• An established interaction between a predictor and moderator on an outcome
• A hypothesized mechanism variable that carries that interaction
• Continuous or binary predictors with a continuous mediator and outcome

When to use something else

• The question is whether an indirect effect varies across moderator levels; use moderated mediation analysis
• No moderator is involved; use simple mediation analysis

Assumptions and requirements

• Correct temporal or causal ordering of predictor, mediator and outcome
• No unmeasured confounding of the mediator-outcome relationship
• Linear relationships and correctly specified interaction terms

What the analysis reports

• Interaction coefficients in mediator and outcome models
• Indirect effect of the interaction with bootstrap confidence interval
• Direct effects
• Simple slopes or conditional effects plots

How to report it

The interaction of [predictor] and [moderator] had a significant indirect effect on [outcome] through [mediator], b = [value], 95% bootstrap CI [[value], [value]].

Related methods

• Moderated Mediation Analysis
• Mediation Analysis
• Moderation Analysis

Run Mediated Moderation Analysis on Quorant