Step by step Manual
Mediation Analysis
1. Hypothesis for mediation: the effect of X on Y is mediated by M, such that higher/lower
levels of X lead to higher/lower levels of M, which in turn lead to higher/lower levels of Y
2. Draw a conceptual model for yourself to visually depict the relations between the
different variables (IV, DV, mediator) and in which you (can) specify the effects.
3. Check assumptions using a multiple regression analysis in SPSS. The procedure is the
same here as for hierarchical/multiple regression analysis. Mediators are treated as
predictors in this case.
4. Analyze > Regression > PROCESS
1. Drag outcome variable to ‘Y variable’
2. Drag the predictor to ‘X Variable’
3. Drag the mediator(s) to ‘M Variable(s)’
4. Model number: 4 (simple mediation)
5. Check on ‘Bootstrap inference for model coefficients
6. Options:
1. Effect size
2. Sobel test (produces significance test)
3. Show total effect model
4. Compare indirect effects (only select when you have 2 or more mediators: when
you have more than one mediator in the model, it estimates the effect and
confidence interval for the difference between the indirect effects resulting from
these mediators)
5. Centering data > all variables that define products
7. Multicategorical
1. If predictor variable is categorical with more than two categories, click this, which
makes PROCESS automatically dummy-code.
5. Output
1. Look at ‘Model summary’ of the outcome variable to see if the model with mediator
added is a significant improvement over de null model.
2. Total effect model is the effect of the predictor on the outcome when the mediator is
not present in the model
1. b = x.xx, 95% CI [x.xx, x.xx], t = x.xx, p = x.xxx (coeff = b)
2. R2 = the model explains X% of the variance in X
3. When the CIs does not cross zero, there is significant mediation.
Total effect = a*b + c'
The total effect is the sum of direct and indirect effects of the X on the outcome (Y)
Direct effect = c'
The direct effect of X on Y when taking the mediator into account.
Indirect effect = a*b
The mediated effect is also called the indirect effect. This is because it is the part of the model
that indirectly affects the outcome through the mediator.
If the effect of X on Y is zero when the mediator is included (c' = 0), there is evidence
for full mediation.
Mediation Analysis
1. Hypothesis for mediation: the effect of X on Y is mediated by M, such that higher/lower
levels of X lead to higher/lower levels of M, which in turn lead to higher/lower levels of Y
2. Draw a conceptual model for yourself to visually depict the relations between the
different variables (IV, DV, mediator) and in which you (can) specify the effects.
3. Check assumptions using a multiple regression analysis in SPSS. The procedure is the
same here as for hierarchical/multiple regression analysis. Mediators are treated as
predictors in this case.
4. Analyze > Regression > PROCESS
1. Drag outcome variable to ‘Y variable’
2. Drag the predictor to ‘X Variable’
3. Drag the mediator(s) to ‘M Variable(s)’
4. Model number: 4 (simple mediation)
5. Check on ‘Bootstrap inference for model coefficients
6. Options:
1. Effect size
2. Sobel test (produces significance test)
3. Show total effect model
4. Compare indirect effects (only select when you have 2 or more mediators: when
you have more than one mediator in the model, it estimates the effect and
confidence interval for the difference between the indirect effects resulting from
these mediators)
5. Centering data > all variables that define products
7. Multicategorical
1. If predictor variable is categorical with more than two categories, click this, which
makes PROCESS automatically dummy-code.
5. Output
1. Look at ‘Model summary’ of the outcome variable to see if the model with mediator
added is a significant improvement over de null model.
2. Total effect model is the effect of the predictor on the outcome when the mediator is
not present in the model
1. b = x.xx, 95% CI [x.xx, x.xx], t = x.xx, p = x.xxx (coeff = b)
2. R2 = the model explains X% of the variance in X
3. When the CIs does not cross zero, there is significant mediation.
Total effect = a*b + c'
The total effect is the sum of direct and indirect effects of the X on the outcome (Y)
Direct effect = c'
The direct effect of X on Y when taking the mediator into account.
Indirect effect = a*b
The mediated effect is also called the indirect effect. This is because it is the part of the model
that indirectly affects the outcome through the mediator.
If the effect of X on Y is zero when the mediator is included (c' = 0), there is evidence
for full mediation.
