Control Variables

In experimental and observational design and data analysis, the term control variable refers to variables that are not of primary interest (i.e., neither the exposure nor the outcome of interest) and thus constitute an extraneous or third factor whose influence is to be controlled or eliminated. The term refers to the investigator's desire to estimate an effect (such as a measure of association) of interest that is independent of the influence of the extraneous variable and free from bias arising from differences between exposure groups in that third variable.

Extraneous variables of this class are usually those variables described as potential confounders in some disciplines. Controlling for a potential confounder, which is not an effect modifier or mediator, is intended to isolate the effect of the ...