Debiasing Auditor Judgments from the Influence of Information Foraging Behavior

• In Event: Behavioral Audit Research

Abstract

This study examines a potential bias in the collection of audit evidence that originates in information foraging behavior. In an electronic information environment, foraging behavior suggests that auditors selecting items for a sample will select those items that will “cost” the least in terms of time and effort. Specifically, in our setting, where an auditor must haphazardly select sample items across multiple divisions with multiple managers, foraging theory predicts that auditors will avoid those divisions whose managers are slower in responding to information requests. Results demonstrate an interactive effect where auditors collect fewer items from a slow manager, but this reduction in sampling is mitigated when a deviation previously occurred that is associated with the slow manager. By definition, haphazard sampling requires a representative sample, such that the avoidance of a single division introduces bias to the sample. Since information foraging is self-serving, the bias is eliminated by a simple intervention of having one auditor select sample items to be tested by another auditor. Additionally, we provide evidence of how additional contextual cues in an audit context can change auditors’ interpretations of client actions. Given the importance of sampling to audit quality, we contribute to the literature by identifying a potentially harmful bias, and perhaps more importantly, we propose a simple and efficient intervention that can eliminate such a bias.

Authors

• Benjamin Commerford, University of Alabama-Tuscaloosa

• Richard Hatfield, University of Alabama-Tuscaloosa

• Richard W Houston, University of Alabama-Tuscaloosa

• Curtis Mullis, University of Alabama-Tuscaloosa