Information Saliency, Auditors’ Analytical Assessments and Learning
Minwoo Lee, Hyun S. Hwang

This study performed an experimental investigation using expert subjects on how different information conditions affect auditors’ probability judgments regarding customer account default in a setting of analytical procedure. This study achieved significant improvements in experimental design issues identified in previous studies in this area, and also addressed unexplored issues. The study found the saliency level of prior probability of the account default negatively affected the magnitude of probability assessment errors. These results based on expert subjects are different from those with non-expert subjects in the previous literature. Previously, the saliency of prior probability was not found to significantly affect the magnitude of probability assessment error. Despite that the subjects were found to generate bias in probability judgments, the magnitude of the error decreased over time with the subjects’ learning from feedback. This learning effect was significant in the condition where the prior probability of default was more salient. This significant learning effect resulted in improvement in probability assessments over time with a reduction of the probability judgment error by a substantial amount.

Full Text: PDF     DOI: 10.15640/ijat.v5n1a5