Abstract
The primary objective of this study was to empirically test the constructs embodied in social exchange theory through a comparison of two self-report measures that were designed to assess individuals' perceptions of relationship interactions: the Behavior Exchange Inventory Form-I (BEI-I) and the Spousal Inventory of Desired Changes and Relationship Barriers (SIDCARB). The secondary aim of this study was to enhance the construct validity of the BEI-I through a correlational analysis with the SIDCARB. The BEI-I contains eight, 6-item scales that are scored on a 5-point Likert scale which assess individuals' perceptions of behavior exchange patterns within a relationship. The SIDCARB consists of twenty-four items divided into three subscales which assess individuals' appraisals of the conjugal exchange process.A total of 111 subjects between the ages of 18 and 55 who were involved in a monogamous relationship for at least six months participated in the study. Both instruments were administered to each subject.Both univariate statistical analyses and the multivariate canonical correlation analysis were performed to assess the nature of the relationships between the scales of the two instruments. Bivariate correlations between the BEI-I and the SIDCARB revealed significant correlations between the perceived directionality of exchange and the perception of relationship equity. Additionally, certain factors were significantly correlated with individuals' willingness to remain in their present relationships.The result of a canonical correlation analysis yielded two significant canonical variates. The first accounted for approximately 53% of the variance between linear composites of the BEI-I and the SIDCARB, the second accounted for 22% of the remaining variance. Six of the eight BEI-I scales and two of the three SIDCARB scales loaded on the first variate. A redundancy analysis revealed approximately 18% of the BEI-I scale variance was predictable from the linear combination of the SIDCARB scales. The results reveal an overlap between the two instruments, suggesting that the BEI-I and the SIDCARB appear to be measuring different aspects of the domain of behavior exchange.