Auditor decision-making; an analysis of the effects of ordering and tests of the predictive validity of the Contrast/Surprise Model

Research in both psychology and accounting indicates that decision makers frequently make judgments which deviate from normative decision models, especially as the level of complexity and uncertainty increases. Many of these deviations appear to result from the use of "heuristics" by the decision maker. One of the heuristics that has been identified as used in the belief revision process is "anchoring and adjustment". This theory is founded upon the belief that individuals are sequential information processors with limited information processing capacity. Thus, a very important question must be asked--since individuals are sequential information processors unable to process a great deal of information simultaneously, how does the order in which information is processed affect the decision outcome?;This study involved an experiment in which 206 practicing auditors were asked to render an audit decision on the financial statements of a company with going concern problems. A case scenario was utilized to present background, general information and current financial position of a company. Both evidence that supports an unqualified audit opinion and evidence that supports a qualified audit opinion was presented. The ordering of the evidence in the case was manipulated to test for any differences in decisions that occurred as a result of the differences in order.;As expected, the ordering of the information did have an impact on the auditor's decisions, but the results did not fully support the predictions of the Contrast/Surprise Model. While the differences due to order were significant, the predicted recency effect did not prove to be significant in all cases. Auditor's appeared more prone to use the contrary evidence (evidence supporting a qualified opinion) but the extent of reliance appeared to be affected by the order in which the evidence was encountered. The findings suggest that audit efficiency and effectiveness may be affected due to the differences in decisions that may be caused by the ordering of audit evidence.;Einhorn and Hogarth (1985) have developed the Contrast/Surprise Model for updating beliefs, which is based upon the theory that individuals are sequential processors and will use an "anchoring and adjustment" strategy to incorporate new information into the decision process. The model predicts strong recency effects, i.e., overreliance on the information processed last, when both consistent and inconsistent evidence is processed. The purpose of this research was to test the predictive validity of the Contrast/Surprise model in a highly uncertain, complex accounting decision, thus testing the effect of order on the professional decision of the accountant.