Examples of Easily Explainable Suppressor Variables in Multiple Regression Research

Abstract

Multiple regression techniques are a valuable tool in conducting ecological studies, especially when provisions are made to control for problems dealing with the interaction of variables. One problem in multiple regression research, the presence of suppressor variables, has the potential to seriously limit findings that can be reported, and in some cases may cause a researcher to pass over a useful data set. Researchers have long been aware of the presence of suppressors in multiple regression research, but there is little agreement as to why it exists or what to do about it. Several considerations in employing methods to "unsuppress" several data sets are discussed.