Moderation

What are the steps to run a moderation?

1. Create a product term by multiplying the two main effects. 2. Regress your DV on your three predictor variables in a hierarchical fashion.

What is a buffer?

A buffer is a moderator that reduces the main effect- the high group yields the flattest line.

What are moderators?

A moderator is a third variable that affects the strength of the relationship between an IV and a DV. It is usually a stable, relatively unchanging variable (e.g., gender, ethnicity).

What are you looking for in a moderation?

A nice smile.

What are you looking for in a moderation?

A significant main effect for the IV, you usually have this. You may or may not have a main effect for the moderating variable. You also check to see whether the interaction term is a statistically significant predictor (p< .05) although this is rarre to find.

What is an exacerbator?

An exacerbator is a moderator that increases the main effect. To exacerbate something is to make it worse- the high group yields the steepest line.

Can you explain a moderation effect without looking at a graph?

No, not typically, (unless you are statistical superbeing Paul Jose).

What are simple slopes?

Simple slopes refer to the steepness/flatness of the lines in the figure. Steep lines depict a stronger correlation between variables, a flat line depicts a weak or null correlation between variables.

What does a moderation tell you?

Under what circumstances something has an effect, i.e., it qualifies the main effect.

What are the two basic ways to interpret an interaction in a moderation?

Where does the spread occur? Or the slope of the lines.

Can a variable be both a mediator and a moderator? What is the difference?

Yes, a rule of thumb is moderators are stable variables that don't change over time. A mediator is a changeable variable that does change in relation to the other two variables studied (IV and DV). If a variable doesn't change it cannot be a mediator, but some variables (e.g., coping stratergies) are partly 'changable' and partly stable, so they can be used in both analyses.