Mediator - 2D Illustration
- Author:
- Shu Fai Cheung
This demonstration visualizes the effect from the predictor (X) to the outcome (Y) in a simple mediation model.
This activity illustrates a mediation model with three variables: X (predictor / independent variable), M (mediator), and Y (outcome variable / dependent variable).
The graph on the left is the one-predictor regression model predicting M by X.
The graph on the right is the partial regression line predicting Y by M, with X partialled out.
Change the three sliders to specify the unstandardized coefficients for this model (the a path, b path, and c' path). You can also move the constants (intercepts) in the two graphs, represented by crosses, to move the two lines up and down.
Then change the X value of Case A (the black point on the X axis on the left graph), and see
(a) how the predicted values of Case A on M and Y, respectively, change;
(b) whether the line on the right, depicting the partial regression line predicting Y by M (X partialled out), change as you move Case A.