PSYC 3380 Lecture Notes - Lecture 14: Sampling Distribution, Confidence Interval, Type I And Type Ii Errors

• School: University of Guelph
• Department: Psychology
• Course: PSYC 3380
• Professor: Jeffrey Spence

Document Summary

The efect of one variable on another is moderated by another variable. Variable w is a moderation of the relationship between x and y. We mean that the form or strength of the relation between x and y depends on the level of w. There may be a level of w under which the relation of x and y is much weaker or even absent. Causal change, why questions, this thing + this thing + this thing happens. Variable x has a relation with y. When we say that m is a mediator, we mean that x has an efect on m, and m in turn has an efect on y. M is a mediator: a middleman, a link in the causal chain. Tested by multiple regressions p values are usually examined to see if mediation is present or absent. A represents the efect of x on m. B represents the efect of m on y, holding x constant.