The effect of explanation source and type on auditors' judgment performance.

Abstract

The purpose of this paper is to investigate the effect of others' explanations for unexpected test results on auditors' judgments of the likelihood that the explanation is correct. Auditors may inherit explanations from various sources. Two primary sources of such explanations are the auditee (client) and a fellow auditor. Two basic types of explanations are error and non-error. Without gathering additional information, auditors cannot know whether a particular unexpected difference is caused by an error (misstatement) or non-error (no misstatement). A between-subjects design is employed to examine how different combinations of source and type of explanation influence auditors' likelihood assessments. This study utilizes an audit scenario wherein subjects inherit an explanation for an unexpected test result. The explanation is attributed to either a fellow auditor or to an auditee. The explanation specifies either an error cause or a non-error cause for the unexpected test result. The description of each source is identical with respect to competency. However, professional skepticism suggests that subjects will attribute varying reliability to the two sources. The two explanations, although different in type, are equally plausible (as determined by a separate of group of subjects who responded to a plausibility survey). Experimental results provide evidence that type of explanation has a significant effect on participants' judgment performance in the form of their assessments of the likelihood of the inherited explanation. Explanations specifying a non-error cause were judged as more likely than explanations specifying an error cause. However, contrary to predictions, the source of the explanation did not affect participants' likelihood assessments. Findings of this study suggest that auditors are more likely to begin with a non-error explanation for an unexpected difference than an error explanation and that the source of an inherited explanation does not have a strong effect on auditors' selection of an initial preferred hypothesis.