Vicarious Learning And Inferential Accuracy in Adoption Processes

Abstract

We develop a model of inferential vicarious learning that explicates how firms can learn whether to adopt a practice whose value varies across organizations. We first theorize how firms can infer this variation from samples of adopters, nonadopters, abandoners, and nonabandoners. Surprisingly, incomplete samples may allow more accurate inferences than complete ones. We then propose how firms, equipped with an understanding of the variation, can select referent organizations that differ from those identified in previous research.