w5 Flashcards
moderators
- A moderator is a variable that influences the strength or direction of the relationship between an independent variable (predictor) and a dependent variable (outcome).
- which treatment works best for whom or under what conditions
- effect depends on x
mediator
- explains the mechanism through which the independent variable affects the dependent variable.
- what rae the underlying mechanisms of treatment success
- effect explained by x
mediation or moderation
Research into who benefits most from a therapy or under what conditions a therapy works best
moderation
mediation or mediation
- Research into the working mechanisms
mediation
can we conclude what is the mechanism of change through mediation
no mediation is the first tep in unraveling the mechanism of change by an intervention
how do we know if there si evidence of mediation
if there is a strong association between the IV and mediator and mediator and DV
T or F
the relationship between mediator and IV and DV is a causal one
F - the relationship between IV and mediator is a causal one but relationship between mediator and DV is not a causal one
why cant we conclude that a moderator causes a change on dv
depending on at what time poitns we measured the mediator and the outcome variable - sometimes we cant conclude which changed 1st
For more insight into causal relationships:
* Repeated assessing mediator and outcome variable
* Change in mediator must (in time) precede change in outcome
* But… never causal statements about mediators, because no
manipulation!
how do we know if mediation exists
we assess whether the indirect effect (a*b) of the independent variable (IV) on the dependent variable (DV) through the mediator (M) is significant. (though bootstapping)
what is the direct, indirect and total effect
Direct effect: c’ - The effect of the independent variable (IV) on the dependent variable (DV) without considering the mediator
Indirect effect: (a × b)The effect of the IV on the DV through the mediator
Total effect: c - The overall effect of the IV on the DV, combining both direct and indirect effects
- Total effect (c) = Direct effect (c’) + Indirect effect (a×b).
when is there a partial mediation
when both the direct effect (c’) and the indirect effect (a × b) are significant. This means the independent variable influences the dependent variable both directly and through the mediator.
how do you analyse nominal moderation
ANOVA
how do you analyze continous moderation
multiple regression
