APA Dictionary of Psychology

n. in regression analysis, the situation in which two independent variables are so highly associated that one can be closely or perfectly predicted by the other. For example, collinearity likely is present if a researcher examines how height and age contribute to children’s weight, since the two predictors are highly interrelated (i.e., as children grow older they get taller). Collinearity leads to difficulties in interpreting the unique influences of the independent variables and requires the use of partialing procedures to distinguish their separate effects. See also multicollinearity. —collinear adj.