Moderation and Mediation Flashcards
Moderator
A variable that affects the strength and/or direction of the relationship between a predictor and the outcome.
Direct Effects
The direct effect of the IV on the DV. To assess, regress the DV on the IV.
Additive Effects
The indirect effect of another variable on the DV. To assess, regress the DV on the other variable.
Testing Moderation
Need to centre means then enter the centred predictor and moderator into block one and the interaction into block two.
Centering Means
Aids interpretation of the interaction and reduces the multicollinearity between the predictor and moderator.
Actual score - mean scorefor predictor and moderator
Mediator
A variable that accounts for the relationship between a predictor and an outcome.
A causes B which causes C.
Baron and Kenny (1986)
Method for testing mediation
- Regress DV on IV
- Regress Mediator on IV
- Regress DV on Mediator and IV
Criteria for Baron and Kenny
Analysis 1 and 2 must be significant and the mediator in analysis 3 must be significant. Sobel’s test must also be significant.
Sobel’s Test
A z-test that identifies whether the indirect effect of the predictor on the outcome via the mediator is significantly different from zero.
Assumptions of Mediation
Mediation assumes that the mediator is free of error. This can lead to biased coefficients.
If multicollinearity is too high, the test power is reduced
Issues with Baron and Kenny
Methods no longer used because they are outdated and flawed, but it does provide a good basis for understanding mediation and moderation.