Method of test administration as a factor in test validity: The use of a personality questionnaire in the prediction of cancer and coronary heart disease
Date of this Version
The validity of personality inventories has always presented considerable problems, as we are usually dealing with hypothetical constructs, rather than intervening variables (Cronbach and Meehl, 1955). Intervening variables are directly observable and reducible to empirical laws, while hypothetical constructs refer to processes or entities that are not directly observed (Garber & Strassberg, 1991). A construct, according to Cronbach and Meehl (1955), is defined as "some postulated attribute of people, assumed to be reflected in test performance" (p. 283). The difficulty with such constructs is that they are basically unobservable and can only be measured indirectly. "Their existence is inferred through the relations between variables that can be observed and the performance on the tests that presumably measure the hypothesized entity" (Garber & Strassberg, 1991, p. 220). Examples of hypothetical constructs are traits, types, attributes, or any quality that cannot easily be operationally defined.
This document has been peer reviewed.