Auditors' Generation of Diagnostic Hypotheses in Response to a Superior's Suggestion: Interference Effects*

Abstract

Abstract. This study investigates the effects of a supedor's suggestion on auditors' generation of additional hypotheses from long-term memory (LTM). It investigates whether an inherited hypothesis affects the category membership of additional hypotheses retdeved from LTM, in particular when hypotheses from few alternative categodes are considered. Based on research in cognitive psychology, the authors predicted that providing auditors with a supedor's suggestion would interfere with their generation of additional hypotheses from the same transaction cycle as that of the supedor's suggestion. The expedmental results are consistent with the expected interference effect. Auditors who inhedted a supedor's suggestion from a particular transaction cycle generated fewer additional hypotheses from the same transaction cycle than did auditors who were not provided with a supedor's suggestion. Moreover, the effect came into play immediately when auditors inhedted a hypothesis involving a typical problem: the first hypothesis generated tended to come from a different transaction cycle than that of the supedor's suggestion.Risumi. Dans l'artide qui suit, les auteurs se penchent sur Tinddence que peut avoir la suggestion d'un supedeur sur les hypotheses suppiementaires que formulent les vedficateurs en puisant dans la « m^moire k long terme ». Ds cherchent 4 determiner si une hypothSse transmise exerce une influence sur la categode h laquelle appartiennent les hypotheses suppiementaires tires de la memoire h long terme, en particulier lorsque les hypotheses envisag^es proviennent d'un petit nombre de categodes possibles. En s'appuyant sur la recherche en psychologie cognitive, les auteuts supposent que la suggestion communiqu^e par un supedeur aux vedficateurs perturbera chez ces derniers la formulation d'hypoth^ses suppiementaires tirees du meme cycle d'op^rations que celui dont provient la suggestion du supedeur. Les resultats expedmentaux des chercheurs confirment cet effet perturbateur. Les vedficateurs h qui un supedeur a transmis une suggestion d'hypothSse provenant d'un cycle d'operations donne ont formuie moins d'hypothSses suppiementaires provenant du meme cycle d'operations que les vedficateurs h qui aucune suggestion n'a ete faite par un supedeur. En outre, l'effet perturba-* The authors gratefully acknowledge the helpful comments of Steve Allen, Bdan Gaber, Tom McKee, Frededck Phillips, Claude Pilote, Rick Tubbs, two anonymous referees, and participants in workshops at

An examination of auditor hypothesis testing strategies in ‘tone at the top’ evaluations: Evidence of diagnostic knowledge structures

“…Although it is not surprising that multiple hypothesis testing strategies are present (diagnostic and confirmatory), the study was designed to disentangle diagnostic from conservative hypothesis testing. By design, the study controls for inherited hypotheses given its likelihood in the auditing context (Bonner et al, 2018; Church & Schneider, 1993) and documents that the indirect support for diagnostic testing (i.e., decision aid main effects) is found to generalize across inherited hypothesis conditions.…”

“…In the ethicality (competency) trait dimension, participants were asked to test whether the new CFO was ethical (competent) or unethical (incompetent) as it relates to his financial reporting and internal control over financial reporting responsibilities. Prior research suggests that auditors inherit such hypotheses—for example, from superior's suggestions (Church & Schneider, 1993) and/or from prior year workpapers (Bonner et al, 2018). The purpose of providing an inherited hypothesis was twofold.…”

An examination of auditor hypothesis testing strategies in ‘tone at the top’ evaluations: Evidence of diagnostic knowledge structures

“…Although it is not surprising that multiple hypothesis testing strategies are present (diagnostic and confirmatory), the study was designed to disentangle diagnostic from conservative hypothesis testing. By design, the study controls for inherited hypotheses given its likelihood in the auditing context (Bonner et al, 2018; Church & Schneider, 1993) and documents that the indirect support for diagnostic testing (i.e., decision aid main effects) is found to generalize across inherited hypothesis conditions.…”

“…In the ethicality (competency) trait dimension, participants were asked to test whether the new CFO was ethical (competent) or unethical (incompetent) as it relates to his financial reporting and internal control over financial reporting responsibilities. Prior research suggests that auditors inherit such hypotheses—for example, from superior's suggestions (Church & Schneider, 1993) and/or from prior year workpapers (Bonner et al, 2018). The purpose of providing an inherited hypothesis was twofold.…”

The Impact of an Auditor's Initial Hypothesis on Subsequent Performance at Identifying Actual Errors*

“…In an auditing study, Frederick (1991) found that experienced auditors who were provided with some items from a list of intemat controls recalled fewer items from the complete list than those who were not provided with any items at all. Similarly, Church and Schneider (1993) found that auditors who were provided with a superior's suggestion (i.e., an inherited hypothesis) from a specific transaction cycle generated fewer additional hypotheses from that cycle than auditors who were not provided with the superior's suggestion.…”

Impact of the timing of receipt of an inherited explanation on auditors’ analytical procedures judgements

“…Contrary to expectations, Libby (1985) found the receipt of an explanation from a particular transaction cycle led to fewer explanations generated from the same transaction cycle. While Libby and Frederick (1990) noted that this result was because of the typicality of the received explanation, Church and Schneider (1993) found the result held, despite explanation typicality. In terms of error/non-error explanation generation, while Anderson et al .…”

An examination of auditor hypothesis testing strategies in ‘tone at the top’ evaluations: Evidence of diagnostic knowledge structures

“…Although it is not surprising that multiple hypothesis testing strategies are present (diagnostic and confirmatory), the study was designed to disentangle diagnostic from conservative hypothesis testing. By design, the study controls for inherited hypotheses given its likelihood in the auditing context (Bonner et al, 2018; Church & Schneider, 1993) and documents that the indirect support for diagnostic testing (i.e., decision aid main effects) is found to generalize across inherited hypothesis conditions.…”

“…In the ethicality (competency) trait dimension, participants were asked to test whether the new CFO was ethical (competent) or unethical (incompetent) as it relates to his financial reporting and internal control over financial reporting responsibilities. Prior research suggests that auditors inherit such hypotheses—for example, from superior's suggestions (Church & Schneider, 1993) and/or from prior year workpapers (Bonner et al, 2018). The purpose of providing an inherited hypothesis was twofold.…”

The Impact of an Auditor's Initial Hypothesis on Subsequent Performance at Identifying Actual Errors*

“…In an auditing study, Frederick (1991) found that experienced auditors who were provided with some items from a list of intemat controls recalled fewer items from the complete list than those who were not provided with any items at all. Similarly, Church and Schneider (1993) found that auditors who were provided with a superior's suggestion (i.e., an inherited hypothesis) from a specific transaction cycle generated fewer additional hypotheses from that cycle than auditors who were not provided with the superior's suggestion.…”

Impact of the timing of receipt of an inherited explanation on auditors’ analytical procedures judgements

“…Contrary to expectations, Libby (1985) found the receipt of an explanation from a particular transaction cycle led to fewer explanations generated from the same transaction cycle. While Libby and Frederick (1990) noted that this result was because of the typicality of the received explanation, Church and Schneider (1993) found the result held, despite explanation typicality. In terms of error/non-error explanation generation, while Anderson et al .…”

An examination of auditor hypothesis testing strategies in ‘tone at the top’ evaluations: Evidence of diagnostic knowledge structures

“…Although it is not surprising that multiple hypothesis testing strategies are present (diagnostic and confirmatory), the study was designed to disentangle diagnostic from conservative hypothesis testing. By design, the study controls for inherited hypotheses given its likelihood in the auditing context (Bonner et al, 2018; Church & Schneider, 1993) and documents that the indirect support for diagnostic testing (i.e., decision aid main effects) is found to generalize across inherited hypothesis conditions.…”

“…In the ethicality (competency) trait dimension, participants were asked to test whether the new CFO was ethical (competent) or unethical (incompetent) as it relates to his financial reporting and internal control over financial reporting responsibilities. Prior research suggests that auditors inherit such hypotheses—for example, from superior's suggestions (Church & Schneider, 1993) and/or from prior year workpapers (Bonner et al, 2018). The purpose of providing an inherited hypothesis was twofold.…”

The Impact of an Auditor's Initial Hypothesis on Subsequent Performance at Identifying Actual Errors*

“…In an auditing study, Frederick (1991) found that experienced auditors who were provided with some items from a list of intemat controls recalled fewer items from the complete list than those who were not provided with any items at all. Similarly, Church and Schneider (1993) found that auditors who were provided with a superior's suggestion (i.e., an inherited hypothesis) from a specific transaction cycle generated fewer additional hypotheses from that cycle than auditors who were not provided with the superior's suggestion.…”

Impact of the timing of receipt of an inherited explanation on auditors’ analytical procedures judgements

“…Contrary to expectations, Libby (1985) found the receipt of an explanation from a particular transaction cycle led to fewer explanations generated from the same transaction cycle. While Libby and Frederick (1990) noted that this result was because of the typicality of the received explanation, Church and Schneider (1993) found the result held, despite explanation typicality. In terms of error/non-error explanation generation, while Anderson et al .…”