Why Aren't All Marketers Bayesians?
(A2022-107311)

AUTHORS

Charles Hofacker, Florida State University

ABSTRACT

Bayesian statistical inference has long ago reached mainstream status in quantitative marketing (Rossi & Allenby, 2003). Surprisingly, Bayesian Decision Theory is far less used despite its deep connection to what both marketing managers and academics need to do after making an inference; in the one case make decisions about which marketing actions to take and in the other decide what marketing actions to recommend to managers. Contrarily, the dominant inferential paradigm within marketing is to make decisions based on null hypothesis significance testing (NHST). The goal of this presentation is to suggest that we as a field turn away from NHST and towards Bayesian inference and decision-making. In this presentation I will briefly describe the current null hypothesis testing convention, compare it to Bayesian inference, describe Bayesian Decision Theory, and illustrate its application with two practical and theoretical examples from the domain of marketing communication.