ABSTRACT
There are two different complicating issues that may arise about an observed relationship between an IV and a DV. First, there may be an antecedent variable that has both a direct and an indirect influence on the DV, the indirect influence being via its relationship with the IV. The second kind of complicating issue arises when there is an intermediate or mediating variable – one that is influenced by the IV and in turn influences the DV. The usual bivariate correlation is sometimes referred to as a zero-order correlation. A partial correlation with one variable controlled is a first-order correlation. Multicollinearity refers to the existence of a strong correlation between the IV and the mediator.
