Abstract
Despite the importance of the concept of involvement, it remains misunderstood. No attempt has been made to establish the reliability and validity of the concept. To resolve apparently conflicting research findings, Houston and Rothschild posit a paradigm which classifies involvement as situational, enduring, and response. The author assesses the reliability and validity of this tripartite classification of involvement by using a multitrait-multimethod matrix approach and a linear structural relations analysis approach. Subsequently, the S-O-R formulation and causality are also tested.