The Effect of Directional Goals on the Perceived Credibility of Data Analytics.

• In Event: Behavioral Financial and I/S Research

Abstract

We examine the joint effect of advice source (data analytics vs. human expert) and advice valence (good news vs. bad news) on the perceived credibility of the source of the advice. Drawing on motivated reasoning theory, we predict that individuals will perceive human experts as being more credible than data analytics, but only when the advice suggests bad news. Furthermore, we predict that this effect is mediated by individuals’ perceived understanding of the reasoning process behind the advice. Using a demand forecasting task, we find evidence that is largely consistent with our predictions. Furthermore, we find that, when advice suggests bad news, individuals utilize the advice from human experts to a greater extent as a result of greater perceived understanding and perceived source credibility. However, as we predicted, we do not find this result if advice suggests good news. Overall, we demonstrate the importance of directional goals for the perceived credibility of data analytics. In doing so, we contribute to accounting practice by cautioning organizations to be aware of potential impediments to the successful implementation and use of data analytics for judgments and decision-making.

Authors

• Xiaoling Chen, University of Illinois-Urbana-Champaign

• Ryan Matthew Hudgins, University of Illinois-Urbana-Champaign

• William F Wright, University of Illinois-Urbana-Champaign