Moderator variables (February 15, 2005).
This page is moving to a new website.
I've always disliked the excessive use of detailed terminology, but when someone asked me
about moderator variables, I had to look up the details. Basically, a moderator variable is
one that interacts with the exposure or treatment variable. It effectively forces you to
qualify your findings. Here's a published example with abstract (I have not read the full
article):
- The relation between locus of control and coping in two contexts: age as a moderator
variable. Blanchard-Fields F, Irion JC. Psychol Aging 1988: 3(2); 197-203.
[Medline] We examined the relation between locus of control and coping as moderated
by age and context. A total of 96 adolescents, young adults, middle-aged adults, and older
adults were administered Lazarus's Ways of Coping questionnaire, a situation-specific
controllability measure, a defensive coping questionnaire, and Levenson's Internality,
Powerfulness of Others, and Chance global locus of control scales for challenging and
threatening contexts. Age moderated the relation between locus of control and coping.
Internality was positively related to escape-avoidance, hostile reaction, and self-blame for
younger individuals and negatively related for older adults. A belief in powerfulness of
others was positively related to planful problem-solving and self-controlling in older
adults and negatively related for younger individuals. Finally, a belief in chance was
positively related to distancing and self-controlling for older adults and negatively
related for younger individuals. The implication that dimensions of control vary with age is
discussed.
As you can see, the relationship between locus of control and coping is quite different
for younger individuals compared to older individuals.
Some web resources that offer definitions of moderator variables are:
There are additional terms used in the research literature such as mediator variable and
intervening variable. I generally try to avoid using these terms, and even shy away from the
commonly used term "confounding variable."