Professional Documents
Culture Documents
Moderation Implied An Interaction Effect, Where Introducing A Moderating Variable
Moderation Implied An Interaction Effect, Where Introducing A Moderating Variable
Moderation Implied An Interaction Effect, Where Introducing A Moderating Variable
Moderation
A moderator is a variable that specifies conditions under which a given predictor is
related to an outcome. The moderator explains ‘when’ a DV and IV are related.
Moderation implied an interaction effect, where introducing a moderating variable
changes the direction or magnitude of the relationship between two variables. A
moderation effect could be
a) Enhancing, where increasing the moderator would increase the effect of the
predictor (IV) on the outcome (DV);
b) Buffering, where increasing the moderator would decrease the effect of the predictor
on the outcome; or
c) Antagonistic, where increasing the moderator would reverse the effect of the
predictor on the outcome.
Hierarchical multiple regression is used to assess the effects of a moderating variable. To
test moderation, we will in particular be looking at the interaction effect between X and
M and whether or not such an effect is significant in predicting Y.
Steps in Testing Moderation
In order to confirm a third variable making a moderation effect on the relationship
between the two variables X and Y, we must show that the nature of this relationship
changes as the values of the moderating variable M change. This is in turn done by
including an interaction effect in the model and checking to see if indeed such an
interaction is significant and helps explain the variation in the response variable better
than before.
SLIDE 2
Model 1, which takes into account the variables A in predicting C, indicates a weak
correlation (r = .12) between A and C. The coefficient of determination,
R squared, indicates that 1.4 % of the change in C is due to A. The relationship
indicated by Model 1 is confined to the data and is not generalizable/significant
because the p value is more than .05 (p = .64). The regression equation is
C = 13.95 – .05A.
SLIDE 6
Encode the moderator, B
SLIDE 7
Click Analyses, then Regression, then Linear Regression. Transfer the variables as
indicated. Do the write up for the Linear Regression output
Model 1 (not the same as the previous one), which takes into account variables A and B,
in predicting variable C, indicates a strong correlation (r = .63) between C and the covariates A
and B. The coefficient of determination, R squared, indicates that 40.0% of the change in C
is due to A and B. The relationship of A to C is not confined to the data and is generalizable/
significant because the p value is less than .05 (p = .017). The same is true for the relationship
between B and C (p=.007). The regression equation is C = 19.04 + .64A – 2.30B.
SLIDE 8
Click Analyses, then medmod, then Moderation.
This new Model takes into account variables A and B and the interaction term A*B
in predicting variable C. The Moderation Estimates indicate that A is a significant predictor
of C (p < .05), B is a significant predictor of C (p < .05), and the interaction term A*B is a
significant predictor of C (p < .05) when the Model takes into account A, B, and A*B in
predicting C.
SLIDE 10