THE LOGICAL STATUS OF SUPPRESSOR VARIABLES.

While social scientists are aware that the relationship between an independent and dependent variable may be spurious, they are much less aware that the absence of relationship may be equally spurious. The chief reason is that the relationship is concealed by a suppressor test factor. This paper supplements earlier discussions of this topic by considering the logical status of suppressor test factors and compensating influences. Examples of antecedent, intervening, and component suppressor variables and of compensating influences are provided. It shows that, in a causal sense, standard test factors and suppressor test factors generate opposite interpretations of spuriousness, but in a formal sense they are the same. Several substantive contributions of suppressor variables to social science are discussed. Suppressor variables are not rarities, but may actually appear as often as standard test factors in research. [ABSTRACT FROM AUTHOR]