A Review of Audit Research Using the Belief-Adjustment Model
The basic audit process requires that the auditor search for and evaluate evidence related to an audit assertion (e.g., Cushing & Loebbecke, 1986; Felix & Kinney, 1982; Gibbins, 1984; Knechel & Messier, 1990). In evaluating an audit assertion (e.g., validity of accounts receivable balance), the auditor begins with an initial belief and then revises that belief upward or downward depending on whether each new piece of evidence is positive or negative, respectively. While this process has been modeled using Bayes rule (Kinney, 1975), a number of behavioral studies in auditing (e.g., Joyce & Biddle, 1981a, b; Kinney & Uecker, 1982; and others) indicate that Bayes rule is not a good descriptor of the auditor’s judgment process. In particular, the research results have indicated that an auditor’s judgment process is sensitive to normatively irrelevant variables such as hypothesis frame (Kida, 1984b), temporal sequence (Joyce & Biddle, 1981a), and more generally that auditors employ heuristics in their judgments of probability. These findings are consistent with similar research in psychology (e.g., Slovic & Lichtenstein, 1971; Tversky &Kahneman, 1974). Accordingly, decision theorists have recently focused on the effects of task variables on information-processing strategies.
KeywordsRecency Effect Belief Revision Likelihood Rating Initial Belief Negative Evidence
Unable to display preview. Download preview PDF.