The influence of peer attitude and inherent scepticism on auditors’ sceptical judgments^†

† Paper accepted by Mark Clatworthy.

Abstract

Our study examines whether auditors’ sceptical judgments are influenced by peer attitude in the context of a peer providing informal advice to auditors in China. We hypothesise that when peer attitude reflects a high (low) emphasis on professional scepticism, auditors will be more (less) sceptical in their judgments. Our results from a between-subjects experiment support this hypothesis. We further examine whether the effect of peer attitude on sceptical judgments differs between auditors with higher or lower levels of inherent scepticism. Our results show that the effect of peer attitude is stronger among auditors with higher levels of inherent scepticism. The results provide evidence on the potential boundaries of informal peer advice as a mechanism to elevate auditors’ professional scepticism. Our findings demonstrate the importance of understanding how peer attitude and inherent scepticism jointly influence their auditors’ sceptical judgments.

Keywords:

• professional scepticism
• auditors’ judgments
• peer attitude
• inherent scepticism
• Chinese auditors

JEL Classifications:

• M40
• M41
• M42

Acknowledgements

We sincerely thank the editor Mark Clatworthy and the two anonymous reviewers for their constructive and helpful comments. We are also grateful to Graeme Harrison, Nonna Martinov-Bennie, Elaine Evans, and Parmod Chand for their helpful suggestions on earlier versions of this paper. We are also grateful to constructive feedback that we received when the paper was presented at the 2017 Macquarie University Conference on Contemporary Accounting Research in Sydney.

Disclosure statement

No potential conflict of interest was reported by the authors.

Data availability

Data are available from the first author.

ORCID

Sammy X. Ying http://orcid.org/0000-0003-0158-8515

Peipei Pan http://orcid.org/0000-0002-2680-7369

Notes

† Paper accepted by Mark Clatworthy.

1 Nolder and Kadous (2018, 3) propose a theoretical framework which consolidates determinants of auditors’ PS “into two categories, individual and social/situational factors.” Nolder and Kadous (2018, 3) note that “individual factors include auditors’ personality traits, knowledge, ability, and motivation,” and social/situational factors include firm culture, the preferences of an auditor's peer group, client pressures, auditing standards, and firm methodologies.

2 In organisational behaviour studies, peer attitude is also referred to as co-worker’s attitude.

3 Formal advice refers to information, recommendations, and alternative perspectives from peers that are documented and offered in a formal setting including meetings (Ng and Shankar 2010, Dalal and Bonaccio 2010, Wright and Bhattacharjee 2018). In contrast, informal advice includes discussions and communication among peers that are not formally documented (Ng and Shankar 2010, Dalal and Bonaccio 2010, Wright and Bhattacharjee 2018).

4 Kadous et al. (2013) manipulate the strength of social bonds as either strong or weak and find that auditors with stronger social bonds with peers heighten their receptivity to peers’ advice. In contrast, we do not manipulate but control for the strength of social bonds with peers by providing consistent information to all participants on their relationship with the peer who provides the advice. We manipulate peer attitude as either a high or low emphasis on PS. Additional details are provided in the research method section.

5 Following the recommendation by Shafer (2009), a local audit firm is the one with operations solely based in mainland China. Shanghai and Shenzhen, being two of the top four Chinese cities classified as first-tier cities, are the most important commercial centres in China (PwC China 2018). Shanghai ranked the first with a GDP of over 3 trillion RMB and Shenzhen ranked the third with a GDP of 2.2 trillion RMB in 2017 (China Daily 2018).

6 Research suggests that social conformity to others often occurs in situations involving either group or peer interactions (Levine and Tindale 2015, Gommans et al. 2017). In a group situation, social conformity is likely to occur when there is a certain degree of consensus in group members’ responses. Alternatively, when individuals interact with a peer, social conformity is likely to occur when the peer is perceived to be more experienced or knowledgeable in the task on hand. We are interested in individual auditors’ judgments when interacting with a peer, rather than in a group situation. Therefore, a relatively more experienced peer was used in our scenario.

7 In our manipulation of peer attitude, the term “professional scepticism” is explicitly mentioned in the high peer emphasis condition, but not in the low peer emphasis condition. Caution may be needed in interpreting the results due to a possible concern of demand effects. To address this possible concern, we pilot tested our manipulation of peer attitude. The feedback from our pilot testing suggested that this manipulation is realistic. The underlying reason is that the fundamental importance of professional scepticism in auditing has been explicitly stressed in the current regulatory environment.

8 Two different perspectives of PS, namely neutral and presumptive doubt views, have emerged in the literature (Nelson 2009, Kang et al. 2015). The neutral view refers “ … a perspective in which an auditor neither assumes that management is dishonest nor assumes unquestioned honesty” (Nelson 2009, 3). The presumptive doubt view “ … refers to an auditors’ assumption of some level of dishonesty unless data indicate otherwise” (Nelson 2009, 3). Researchers suggest a shift from a neutral towards presumptive doubt perspective of PS (Bell et al. 2005, Nelson 2009, Quadackers et al. 2014, Kang et al. 2015). Moreover, Nelson (2009, 3) suggests that, “regulators appear to take more of the presumptive doubt perspective, as they typically refer to PS as something that was missing when an audit failure has occurred”. We adopt the presumptive doubt perspective, which is reflected in the measures of sceptical judgments used.

9 We pilot tested the research instrument in the following stages. First, we pilot tested the instrument in English among 22 Master’s accounting students, five accounting professors, and two practising auditors, with the objective of assessing the readability and understandability of the instrument materials, and whether the scenario was realistic. We received feedback that the scenario was realistic and some ambiguous expressions were identified and corrected. Second, we developed a Chinese version of the research instrument using translation-and-back-translation following prior research (Brislin 1980). Any discrepancies between the final and original versions were discussed and resolved. Finally, the research instrument in Chinese was evaluated by two Chinese audit partners, three auditing professors, and then we pre-tested the instrument in Chinese among 25 Master’s accounting students in a Chinese university. We made some editorial changes based on the feedback to improve the understandability and readability of the research instrument.

10 This research was approved by the “Human Research Ethics Committee” of the University.

11 Responses also included 16 participants who had no experience in auditing trade receivable. We conducted additional tests, which excluded these responses, and the results are consistent with our results obtained from the main analyses. Therefore, we only report the main results in the paper.

12 As discussed earlier, auditors’ sceptical judgments were measured by three items. For Item 1: “What is the likelihood that the explanation provided by the CFO is reliable?”, higher scores indicate lower scepticism. For the other two items, higher scores indicate higher scepticism. To be consistent with the other two items, scores on Item 1 were reverse coded in the subsequent data analyses. This reverse-coding enables us to conduct comparisons between three items when testing the hypothesis and the research question.

13 Consistent with prior audit judgment studies (Boyle et al. 2015, Ying and Patel 2016, DeZoort et al. 2019), we also conducted MANOVA to test the effect of peer attitude on the three measures of sceptical judgments taken together. The rationale is that these three dependent variables are used to measure a single construct, namely sceptical judgments, although each of these measures focuses on different aspects of the construct. Moreover, as an important preliminary test for MANOVA, we examined the correlations of the three items measuring sceptical judgments. The correlations show that the three items were significantly correlated with each other (p < 0.01) and therefore it is appropriate to use MANOVA.

14 This is consistent with the suggestion of prior research that “ … professional scepticism, as measured by this scale, may be a trait that is fixed by the time an auditor commences audit training and practice and is not malleable.” (Carpenter and Reimers 2013, 64) Our results show that while a peer’s attitude influences auditors’ sceptical judgments, it does not influence auditors’ inherent levels of scepticism as measured by the Hurtt’s (2010) scale.

15 We also computed multiple regression, controlling for gender, age, local versus international audit firm, current position, subjects’ confidence and familiarity with case-specific task. The results show that there is no significant influence of gender (p = 0.090), age (p = 0.916), local versus international audit firm (p = 0.972), current position (p = 0.228), subjects’ familiarity (p = 0.381) and confidence (p = 0.061) on auditors’ sceptical judgments at the (p = 0.05) level. The inferences are consistent with the main analyses.

16 Consistent with prior literature including Harding and Trotman (2017), we also computed Analysis of Covariance (ANCOVA) and Multivariate Analysis of Covariance (MANCOVA) using the z-standardized inherent scepticism score as a covariate, peer attitude as the independent variable, and three items measuring auditors’ sceptical judgments as the dependent variables. Our results show that there is no significant influence of inherent scepticism on sceptical judgments (ANVOVA results: p = 0.38 for Item 1, p = 0.118 for Item 2, p = 0.31 for Item 3; MANCOVA result: p = 0.416 for aggregating the three items together).

17 We also conducted additional tests (non-tabulated) to examine whether auditors with higher (lower) levels of inherent scepticism make different sceptical judgments in situations when peer attitude reflects either a low or high emphasis on PS respectively. Our results confirm that the differences of sceptical judgments between auditors with lower and higher levels of inherent scepticism are not significant in both situations.

Additional information

Funding

This paper has also benefited from the funding support by National Natural Science Foundation of China (NSFC) Project No. 71272227 and No. 71872136.