Academic journal article Auditing: A Journal of Practice & Theory

Debiasing the Outcome Effect: The Role of Instructions in an Audit Litigation Setting

Academic journal article Auditing: A Journal of Practice & Theory

Debiasing the Outcome Effect: The Role of Instructions in an Audit Litigation Setting

Article excerpt


The outcome effect occurs where knowledge of the occurrence of an event outcome negatively influences individuals' evaluations of the skill and/or competence of a judge who did not foresee at outcome prior to its occurrence, even where it is not clear that the event outcome was foreseeable. (1) This paper reports the results of an experiment in an applied auditing setting (a court action for auditor negligence) that tests the efficacy of two different sets of instructions as potential mitigators of the biasing effect of outcome knowledge.

The first set of instructions, which we have labeled Outcome Effect Awareness Instructions (AWARE), stress the potential biasing effect of outcome knowledge. The second set of instructions, labeled Consequences (CONSEQ), stress the potential negative effects to the auditor should attribution be established. The study also provides evidence on the way in which outcome knowledge affects ex post judgments--through highlighting (downplaying) evidence consistent (inconsistent) with the known outcome or by affecting the manner in which the evidence is combined.

In the context represented by this experiment, the outcome effect may be highly consequential. In auditing, a fundamental judgment that the incumbent auditor must make is whether a client will be able to remain a going concern. If the auditor issues an unqualified audit report and the client subsequently fails, then the auditor may be sued for negligence for not recognizing the signs of impending business failure and informing the shareholders. A negative evaluation by jurors can have profound negative financial and professional consequences for the auditor. Such evaluations of the competence of the auditors and the quality of their judgment(s) are made with the outcome knowledge of insolvency or bankruptcy. Lowe and Reckers (1994) and Kadous (2000, 2001) found outcome effects in the decisions of mock jurors asked to assess auditor negligence in hypothetical cases of audit failure followed by the bankruptcy of the auditee; Anderson et al. (1993) and Anderson et al. (1997) found the same outcome effects in the decisions of trial judges.


Initial identification of the existence of an outcome knowledge bias was made by Fischhoff (1975). His research, and the studies of contemporary researchers, related to a phenomenon he termed "hindsight bias." The hindsight bias is "direct" in that knowledge of an event outcome causes ex post observers to overstate the probability with which they would have been able to predict that outcome before it occurred. An alternative phenomenon, the outcome effect, is indirect in that it relates to evaluation of the judge who made a decision based on what appears, after the event, to have been an erroneous estimation of the probability of an outcome. The outcome effect has been shown to exist in performance evaluation in medicine (Baron and Hershey 1988; Dawson et al. 1988; Mitchell and Kalb 1981); business management (Brown and Solomon 1987, 1993); military decision making (Lipshitz 1989); financial reporting (Buchman 1985); financial analysis (Fisher and Selling 1993); and strategic business decision making (Bukszar and Connolly 1988; Tan and Lipe 1997). As will be discussed in the next section, the outcome effect is separate and distinct from the notion of hindsight bias (Hawkins and Hastie 1990; Christensen-Szalanski and Fobian-Willham 1991).

The explanation for the biasing effect of outcome knowledge that has the strongest support (the explanation initially proposed by Fischhoff [1975]) is cognitive and involves reconstruction and assimilation of outcome knowledge to antecedent events. Individuals tend to "make sense" of a situation by reconciling their existing knowledge structure to include the effects of the known outcome. Causal links are formed in a backward processing mode from the outcome to antecedent factors. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.