Cognitive dissonance, selective perception and the self-fulfilling prophecy in the examination of audit evidence

The problem addressed by this study was one aspect of the audit process: how accurately auditors perceive evidential matter. The study involved an experiment with practicing auditors which used a case concerning dispositions of confinnations of accounts payable. The experiment was designed to test for the possible effects of selective perception, cognitive dissonance and the Self-Fulfilling Prophecy on the auditors perception of audit evidence. These three theories imply, generally, that investigators, perhaps even auditors, tend to find what they expect to find. Distinctions were made between auditor perceptions and auditor judgments because the experiment was directly related to auditor perceptions and only secondarily related to auditor judgments. A case containing identical information concerning dispositions of exceptions to confirmations of accounts payable was administered to two hundred and fourteen practicing auditors from thirteen offices representing seven of the "Big Eight" firms, one national firm, one regional firm, and two local firms. The subjects were randomly divided into three experimental groups and, prior to working the actual case, were given different information concerning internal control related to the case. Group 1 received no information on internal control; Group 2 received information indicating weak internal control; and Group 3 received information indicating strong internal control. The experiment was to determine if the expectations built by the prior information on internal control influenced the subjects perceptions of the evidential matter in the case. The study explained that prior information can be properly used in judgments, but prior information should not influence perceptions. This experiment involved auditor perceptions, not judgments. Selective perception, cognitive dissonance and the Self-Fulfilling Prophecy implied that, although all subjects had exactly the same evidential material in the case. Group 2 (weak internal control) should have perceived significantly more unsatisfactory dispositions than Group 1 (no internal control information), and Group 1 should have perceived more unsatisfactory dispositions than Group 3 (strong internal control) if the effects related to these theories existed in the experiment. If expectations did not affect the subjects perceptions, no significant differences should have existed. Statistical analysis of the results of the experiment indicated that the subjects in Group 2 did perceive significantly more unsatisfactory dispositions than Group 3. Further analysis suggested that the confidence level of the subjects evaluation of internal control (a surrogate for the strength of the subjects expectations) was related to significant differences in subject group responses. These results supported the existence of selective perception in the experiment. Initial and final evaluations of internal control by the subject groups were examined to determine if cognitive dissonance and the Self- Fulfilling Prophecy existed in the experiment. The changes in measures of variability supported the existence of cognitive dissonance, but results related to the Self-Fulfilling Prophecy were ambiguous and the existence of the Self-Fulfilling Prophecy was not determinable for this experiment. Tests for randomness, subject consistency in internal control evaluation, and principal investigator bias all strongly supported the internal validity of the experiment.