Improving the Detection of Interactions in Selling and Sales Management Research

Abstract

Because there is little guidance for detecting interactions among unobserved variables in theory tests, the paper provides suggestions to selling and sales management researchers for improved detection of these variables. After a review of situations where including interactions might be appropriate, the paper describes detection techniques for these variables. Since nonlinear structural equation analysis and errors-in-variables techniques are less accessible than regression and generally unknown to researchers in Marketing, the paper discusses the interaction detection capabilities of several regression-based techniques. It also reviews the effects the data used to test a model containing interactions can have on their detection. These effects are illustrated in detecting an interaction between salesperson role clarity and closeness of supervision in their association with satisfaction. The paper concludes with suggestions for improving the detection of interactions among unobserved variables.