Examination of Contextual Effects and Changes in Task Predictability on Auditor Calibration.

Research on calibration is mainly located in the psychology literature, where the general finding is that individuals are overconfident in their judgments. On the other hand, the results of studies examining auditor calibration suggest auditors are less confident, tending towards underconfidence. There appear to be two competing explanations for the divergence, namely task predictability [Lichtenstein et al. 1982] and contextual effects of auditing [Solomon et al. 1985]. This study aims to reconcile these findings by obtaining auditors' and non-auditors' responses to both auditing and general knowledge questions across different levels of task predictability. This enables the consideration of both the task predictability and the contextual effects explanations in a single experimental design. When examining the explanations separately, both appear to account for the observed differences in calibration. However, when task predictability is controlled for, most of the task contextual effect disappears while some subject contextual effect remains. The findings of this study suggest that task predictability is a means for explaining and reconciling the conflicting findings of the audit and psychology calibration literatures.