Auditing, Audit experience, Memory errors

This study is an examination of auditors' ability to integrate audit evidence. Integration is defined here as identification of the meaningful relationships that exist between separate pieces of information. Failure to integrate is defined as failure to identify such relationships. Auditor subjects at four levels of experience simulated a supervisor's review of audit workpapers; the workpapers contained eight target contradictory pairs of evidence. The nature of the evidence was manipulated so that the contradicting and contradicted pieces in each pair were either in the same or different audit tests, and either used repeated wording or did not. Integration was counted when subjects wrote review notes indicating that the evidence covered by the contradictory documentation needed to be reconsidered. Failure to integrate was counted when subjects failed to point out any problem with the contradictory evidence. The results indicate that experienced auditors integrated significantly more often than inexperienced auditors. It was expected that this superior performance would carry over to contradictory items that were not proximate when encountered and that were not semantically similar. Contrary to expectations, experienced as well as inexperienced subjects failed to integrate more often when the contradictory items were in different audit tests or did not repeat wording. Some implications of these results for the design of decision aids and future research are discussed. Résumé. L'auteur examine la capacité des vérificateurs d'intégrer l'information probante. L' intégration est définie comme étant la capacité de déterminer les relations significatives qui existent entre différents éléments d'information. L' incapacité d'intégrer est définie, elle, comme étant l'impossibilité de déterminer ces relations. Les vérificateurs qui ont participé à l'étude possédaient quatre niveaux d'expérience; ils ont simulé l'examen par le chef d'équipe des feuilles de travail d'un dossier de vérification. Ces feuilles de travail contenaient huit parties cibles d'informations probantes contradictoires. La nature de l'information probante a été manipulée de telle sorte que l'information contradictoire et l'information contredite de chaque paire étaient issues soit des mêmes sondages de vérification, soit de sondages différents. Les cas d'intégration dénombrés correspondaient à la situation où les sujets indiquaient dans leurs notes de révision que l'information probante touchée par la documentation contradictoire devait être réexaminée. Les cas de non‐intégration dénombrés correspondaient à la situation où les sujets ne parvenaient pas à cerner un problème quelconque à partir de l'information probante contradictoire. Les résultats indiquent que les cas d'intégration sont beaucoup plus fréquents chez les vérificateurs expérimentés que chez les vérificateurs inexpérimentés. On aurait pu s'attendre à ce que cette performance supérieure se transpose aux éléments contradictoires dont l'occurrence était éloignée et qui ne se ressemblaient pas sur le plan sémantique. Or, constatation inattendue, ni les sujets expérimentés ni les sujets inexpérimentés ne sont parvenus à intégrer plus fréquemment l'information lorsque les éléments contradictoires étaient issus de sondages de vérification différents ou que leur formulation était différente. L'auteur élabore sur les conséquences de ces résultats pour la conception d'aides à la décision et l'orientation des recherches à venir.

Accountants who presume that documentation of observations will be available whenever needed may not process information thoroughly enough to internalize it. If they rely on access to documentation, they may later fail to recognize the need to refer to that documentation, and may then overlook important pieces of information. As a result, they may be less effective when asked to evaluate inferences than accountants who rely on more careful initial processing. This paper reports experiment results that indicate that accountants may make more accurate tests of inferences when they rely on their memories than when they rely on free access to source materials. Résumé. Les experts‐comptables qui supposent que les pièces justificatives relatives aux observations seront à leur disposition dès qu'ils en auront besoin risquent de ne pas traiter l'information de façon suffisamment approfondie pour en garder souvenance. S'ils comptent sur l'accès aux pièces justificatives, il se peut qu'ils ne reconnaissent pas, le moment venu, la nécessité de se reporter à ces pièces justificatives et qu'ils négligent ainsi des éléments d'information importants. Par conséquent, ils risquent d'être moins efficaces, si on leur demande d'évaluer des inductions, que les experts‐comptables qui s'appuient sur un traitement initial de l'information plus rigoureux. Les auteurs livrent les résultats d'une experience qui révèle que les experts‐comptables peuvent parvenir à des sondages plus précis des inductions lorsqu'ils comptent sur leur mémoire que lorsqu'ils s'appuient sur le libre accès aux pièces justificatives.

Reviews the book `Accounting for the Environment,' by Rob Gray, Jan Bebbington and Diane Walters.

ABSTRACT: Auditors' reliance on their memories vis-a-vis their reliance on external information sources such as workpapers is of practical as well as research interest. This study examined auditors' relative confidence in relying on accurate and Inaccurate recognitions of audit evidence. Each of 85 practicing auditors was asked to review a set of hypothetical audit workpapers, then after one day's delay, to respond to a recognition test and rate their confidence in their recognitions. They were asked to use the confidence rating to reflect their willingness to rely on their recognitions rather than on reviewing the workpapers or source documents again. Auditors were most confident in two cases: when they accurately recognized evidence that they had previously seen and when they confused their own Inferences with actually observed evidence. Auditors are often as confident in their Incomplete and inaccurate recognitions as they are in their accurate recognitions.

[Auditors' reliance on their memories vis-�-vis their reliance on external information sources such as workpapers is of practical as well as research interest. This study examined auditors' relative confidence in relying on accurate and inaccurate recognitions of audit evidence. Each of 85 practicing auditors was asked to review a set of hypothetical audit workpapers, then after one day's delay, to respond to a recognition test and rate their confidence in their recognitions. They were asked to use the confidence rating to reflect their willingness to rely on their recognitions rather than on reviewing the workpapers or source documents again. Auditors were most confident in two cases: when they accurately recognized evidence that they had previously seen and when they confused their own inferences with actually observed evidence. Auditors are often as confident in their incomplete and inaccurate recognitions as they are in their accurate recognitions.]