Which statement is true about mediators and moderators?

Which statement is true about mediators and moderators?

• A. A mediator explains how an effect occurs, while a moderator affects the strength or direction of the effect across levels of another variable.
• B. A mediator is a sampling error.
• C. A moderator is a bias.
• D. They are the same concept.

Answer: (A)

Mediators explain how an effect occurs, while moderators affect the strength or direction of the effect across levels of another variable. A mediator is an intermediary variable that transmits the influence of an independent variable to a dependent variable, revealing the mechanism by which the effect happens. A moderator is a variable that changes the size or even the direction of the relationship between the independent and dependent variables, depending on its level, by creating an interaction with the main predictor. For example, imagine a training program designed to boost job performance. If the program works by increasing employees’ motivation, and that increased motivation leads to better performance, motivation is a mediator—it explains the path from program to outcome. If the program’s impact on performance is stronger for younger employees and weaker for older employees, age is a moderator—it changes the effect size across different groups. Sampling error is random variation that occurs because you’re looking at a subset of the population, not a mechanism linking variables. Bias is a systematic distortion in how data are collected or measured. They are not the same concept as mediators or moderators and do not describe how or when effects occur or change.

Mediators explain how an effect occurs, while moderators affect the strength or direction of the effect across levels of another variable. A mediator is an intermediary variable that transmits the influence of an independent variable to a dependent variable, revealing the mechanism by which the effect happens. A moderator is a variable that changes the size or even the direction of the relationship between the independent and dependent variables, depending on its level, by creating an interaction with the main predictor.

For example, imagine a training program designed to boost job performance. If the program works by increasing employees’ motivation, and that increased motivation leads to better performance, motivation is a mediator—it explains the path from program to outcome. If the program’s impact on performance is stronger for younger employees and weaker for older employees, age is a moderator—it changes the effect size across different groups.

Sampling error is random variation that occurs because you’re looking at a subset of the population, not a mechanism linking variables. Bias is a systematic distortion in how data are collected or measured. They are not the same concept as mediators or moderators and do not describe how or when effects occur or change.