Using decision modeling to measure second level valences in expectancy theory

Abstract

Abstract This paper employs a decision modeling approach to measure second level valences in Expectancy Theory. As proposed by J. C. Naylor, R. D. Pritchard and D. R. Ilgen (A theory of behavior in organizations, New York: Academic Press, 1980 ) second-level valences are measured across different levels of an outcome. In the first experiment, the job-preference decisions of 24 under-graduates were examined using a decision making exercise involving 24 hypothetical jobs described in terms of three intrinsic instrumentalities. In the second experiment, the job-preference decisions of 57 undergraduates were examined using a decision-making exercise involving 24 hypothetical jobs described in terms of four extrinsic instrumentalities. Factorial designs were used in both experiments to preserve orthogonality and allow a separate interpretation of each of the second-level valences. A regression model was derived for each subject to block on individuals and provide a within-person analyses of the data. In both experiments, the beta weight measures of the second level valences (1) conformed to the concept of Naylor et al. that second-level valence is a relationship across levels of an outcome; (2) operationalized the within-person property of Expectancy Theory; (3) allowed separate interpretations of each second-level valence; and (4) displayed stable, high internal consistency estimates. Therefore, it appears that decision modeling with beta weight measures of second-level valences offers an innovative approach for Expectancy Theory researchers.