BFA302 Assessment 1 Article

Accounting & Finance 57 (2017) 1177–1202
The effects of tone at the top and coordination with external

auditors on internal auditors’ fraud risk assessments
Isabel Z. Wanga, Neil Fargherb

a
The University of Western Australia, Crawley, WA, Australia
b
The Australian National University, Canberra, ACT, Australia
Abstract
Prior research suggests that internal auditors’ judgements are subject to

management influence resulting in compromised risk assessments. This study
investigates the effects of the tone at the top and coordination with external
auditors on internal auditors’ fraud risk assessments. Results of an experiment
involving 64 internal auditors indicate that when the tone at the top is poor,
rather than favouring management, internal auditors report a higher risk of
intentional misstatements and that coordination with external auditors can
further reduce expectations of the incidence of intentional misstatements.
Key words: Tone at the top; Coordination between internal and external
auditors; Fraud risk assessment
JEL classification: M41, M42
doi: 10.1111/acfi.12191
1. Introduction
This study investigates the effects of the quality of the tone at the top of an
organisation and coordination with external auditors on internal auditors’ fraud
risk assessments. The internal audit profession expects internal auditors to play
an active role in preventing and detecting fraudulent reporting (IIA, 2013c).1 If
1
Internal auditors perform multiple roles including a variety assurance-related task and
services (Ahlawat and Lowe, 2004; Gramling et al., 2004; Beasley et al., 2005; Gramling
and Myers, 2006; Coram et al., 2008; Trotman and Trotman, 2014). From an internal
audit perspective, an assurance service is ‘an objective examination of evidence for the
purpose of providing an independent assessment on governance, risk management and
control processes for the organization’ (IIA, 2013f). Internal auditing standards
increasingly require internal auditors to be able to identify and appropriately respond to
the potential for the occurrence of fraud and how the organization manages fraud risk
(e.g. IIA, 2013c).
© 2015 AFAANZ
1178 I. Z. Wang, N. Fargher/Accounting & Finance 57 (2017) 1177–1202
internal auditors are to play an active role in the prevention and detection of
fraudulent financial reporting, adequate attention must be paid to assessing the
risks arising from fraud perpetrated by senior management.2 Prior research
suggests that senior management has substantial influence over internal auditors
through compensation, performance evaluations and job security (Christopher
et al., 2009; Rose et al., 2013); as a result, internal auditors have been found to
tend to assume the role of advocates for management (Ahlawat and Lowe, 2004).
If internal auditors are subject to management influence, then they would be
expected to strategically bias their judgements on the risk of senior management-
related fraud, resulting in compromised fraud risk assessments. However, despite
their roles in the organisation as employees, internal auditors are also part of the
auditing profession and therefore expected to adhere to professional and ethical
responsibilities which should mitigate bias arising from management influence.
Prior research finds that the independence of internal auditors is threatened
by the power that senior management holds through the functional authority
over internal auditors (e.g. Norman et al., 2010; Rose et al., 2013). However,
research has not empirically examined how the quality of ethical leadership of
senior management affects internal auditors’ judgements. COSO (1987, 2013)
states that one of the most important factors contributing to the integrity of the
financial reporting process is the tone set by top management (Committee of
Sponsoring Organizations of the Treadway Commission (COSO), 1987, p. 11;
2013, p. 4).3 Similarly, ASA 315.A76 states that the tone at the top set by top
management influences the control consciousness of all personnel and is the
foundation for the other components of internal controls. Research also
suggests that an organisation’s ethical culture and leadership can affect
employees’ ability to rationalise fraud (e.g. Trompeter et al., 2013). Specifically,
Bartlett and Preston (2000) argue that employees can be expected to attain high
ethical standards only when the expectations of an organisation are commu-
nicated strongly through a corporate culture that respects ethical standards and
fosters ethical actions.4 While standard setters posit that internal auditors serve
as a key resource to the audit committee for monitoring top management
(COSO, 2012; IIA, 2013c), prior research on corporate culture raises a question
2
Senior management has been identified as one of the groups of concern in relation to
financial statement fraud (e.g. Dechow et al., 2011).
3
In organisational settings, the set of values and motives imposed by an organisation’s
leadership is typically referred to as the ‘tone at the top’. Tone at the top is used to refer
to the entity wide attitudes of integrity and control consciousness, as exhibited by the
most senior executives of an organisation (Association of Certified Fraud Examiners
[ACFE], 2006). Consistent with the usage among practitioners (e.g. ACFE, IIA), we use
the phrase ‘poor tone at the top’, as distinct from ‘bad tone at the top’, as the contrast to
‘good tone at the top’.
4
Patelli and Pedrini (2015) similarly find that leadership traits can craft an environment
characterised by either integrity or unethical practices.
© 2015 AFAANZ
I. Z. Wang, N. Fargher/Accounting & Finance 57 (2017) 1177–1202 1179
as to whether internal auditors who work within a poor ethical culture are able
to maintain an unbiased view when assessing the risk of senior management-
related fraud. This study therefore extends this research to consider the effect of
the quality of the tone at the top on internal auditors’ assessments of fraud risk
in an experimental setting.5
IIA (2013d) suggests that having access to the external auditors’ techniques
and methods enables internal auditors to improve their assurance-related work.
The second objective of this study is therefore to investigate whether
coordination between the internal and external auditors can mitigate any
negative management influence on internal auditors’ fraud risk assessments,
especially when the tone at the top is poor. Following Gramling et al. (2004),
which highlights the importance of a high-quality internal audit function
having a functional relationship with external auditors, we predict that the
internal auditor is more likely to conform to the expectations of the external
auditor by coordinating assurance-related work with the external auditor when
making judgements. When the internal auditor is vulnerable to biased
judgements in the face of conflicting interests as a result of perceived power
from the top management, we expect that the coordination between the internal
and external auditors can potentially mitigate negative management influence
on the internal auditor’s assessment of senior management-related fraud risk.
Using a 2 9 2 between-subjects factorial experiment, consistent with poor
tone at the top being a red-flag to fraudulent financial reporting and internal
auditors’ professional obligations, we find that internal auditors report a higher
likelihood of intentional misstatements under a poor tone at the top than under
a good tone at the top. This finding runs counter to concerns of bias in internal
auditors’ judgements from the influence of management. Our results also
indicate that internal auditors perceive that coordination with external auditors
is associated with expectations of fewer financial misstatements.
This study advances research on internal auditing in four ways. First, prior
research finds that the independence of internal auditors can be influenced by
senior management power (e.g. Norman et al., 2010; Rose et al., 2013). Our
study contributes to this area of literature by examining another source of
management influence, specifically the influence of poor tone at the top. Second,
our study responds to a call by Everett and Tremblay for research on how the
field’s ethical narratives are actually translated into internal audit practice
(Everett and Tremblay, 2014). Our findings suggest that under a poor tone at the
top internal auditors, risk assessments are consistent with the auditors being more
skeptical. This finding is consistent with the benefits of continuous training on
ethics and risk management undertaken by the internal audit profession. Third,
our study addresses calls to better understand the benefits of coordination
5
From a survey, Hansen et al. (2009) find that internal auditors who work in
organisations that assess tone at the top tend to report higher tone at the top, but the
link to auditor judgment is unclear.
© 2015 AFAANZ
between internal and external auditors from the internal auditors’ perspective
(e.g. Gramling et al., 2004; Lenz and Hahn, 2015). Our findings complement
prior research by providing evidence that the coordination between the two audit
groups helps internal auditors improve their assurance-related assessments.
Finally, our evidence of fraud risk assessments that indicate greater skepticism
under a poor tone at the top can potentially increase the external auditors’
reliance on internal auditors’ work even in situations where the tone at the top is
poor (Gramling et al., 2004; Munro and Stewart, 2010).
The following section reviews prior literature and develops the hypotheses.
This is followed by a description of the research methods and presentation of
the results. The paper concludes with a summary of the findings, limitations
and opportunities for future research.
2. Theoretical background and hypotheses
2.1 Tone at the top on internal auditors’ fraud risk assessments
We develop two competing hypotheses regarding the association between

tone at the top and internal auditors’ assessments of fraud risk. If employees,
including internal auditors, pay close attention to the behaviour of top
management, their judgements and decisions are influenced by the tone at the
top created in the workplace by top management (Schwarts et al., 2005). Prior
studies of internal auditors’ attitudes towards management or client suggest
that maintaining a true objective view may not be possible for internal auditors
because internal auditors tend to advocate for management (Brody and Lowe,
2000; Ahlawat and Lowe, 2004; Messier et al., 2011; Burton et al., 2015). When
internal auditors act as advocates for management, they will be more likely to
make judgements that align with management expectations (Harrell et al.,
1989; Mason and Levy, 2001; Roussy, 2013). However, despite their roles in the
organisation as employees, internal auditors are also part of the auditing
profession with professional and ethical responsibilities to the rest of society
(Gaa, 1992). As suggested by Lord and DeZoort (2001), internal auditors’
professional commitments should transcend their organisational commitments.
Given these competing commitments faced by internal auditors, in this study,
we develop two competing hypotheses to test the effect of tone at the top on
internal auditors’ fraud risk assessments.
First, we incorporate social power theory6 into the advocacy viewpoint. As
suggested by Mast et al. (2010), a manager has the ability to influence,
persuade, and motivate followers’ decision-making from the power the
manager is perceived to possess. Internal auditors are assumed the role of
management advocates because they are influenced by the managerial power.
6
For further discussion of social power theory from the psychology literature, refer to
Lasswell and Kaplan (1950) and French and Raven (1959).
© 2015 AFAANZ
Prior literature on internal auditors suggests that they feel vulnerable to and
dependent on senior management and the audit committee (Zain and
Subramaniam, 2007; Norman et al., 2010; Rodgers and Al Fayi, 2015).
Roussy (2015) finds that some internal auditors admitted to ‘turn a blind eye’ to
dubious situations if the top manager or an audited manager who exercises
significant power within the organisation requested it.
As explained by Lasswell and Kaplan (1950), a manager influences a
follower’s judgements by expressing certain values or motives. In organisational
settings, the ethical tone set by the upper management expresses management’s
attitudes towards integrity and corporate ethics. If upper management appears
concerned with ethics and integrity, and communicates a positive value to
employees (a good tone), then we would expect internal auditors are more likely
to make judgements in line with the management concern for ethics and
integrity. By contrast, if upper management does not uphold ethics and
integrity, and communicates a negative value to employees (a poor tone), and
internal auditors favour management views, then internal auditors would be less
concerned with professional ethics and integrity in their judgements. If advocacy
is of concern, then we would expect that internal auditors would be less
concerned with potential financial problems, either intentionally or uninten-
tionally, under a poor tone at the top. This leads to the first hypothesis:
H1a: The internal auditor’s assessment on fraud risk is lower when the tone
at the top is poor, compared to when the tone at the top is good.
Conversely, from the viewpoint of professional commitment, the commitment

to the internal audit profession should orient internal auditors towards
behaviour that is in the public interest and away from behaviour that has the
potential to damage the profession. In the context of fraud risk assessments,
internal auditors are expected to comply with the professional standards
requiring internal auditors to exercise due professional care by considering the
probability of significant errors, fraud or noncompliance (IIA, 2013b). Further,
empirical research in internal auditing examining the perceived effectiveness of
red flags as outlined in SAS No. 99 suggests that internal auditors are aware of red
flags regarding management characteristics (Apostolou et al., 2001; Church
et al., 2001). As opposed to the advocacy perspective, which suggests internal
auditors will bias their fraud risk assessments in favour of management due to
management influence, evidence from the red flag literature suggests that internal
auditors will be more skeptical when the tone at the top is poor. That is, internal
auditors are more likely to assess the risk of having financial misstatements to be
even higher when they consider the poor tone at the top as a signal to potential
fraud. The above arguments lead to a competing hypothesis to H1a:
H1b: The internal auditor’s assessment on fraud risk is higher when the tone
© 2015 AFAANZ
2.2 Coordination with external auditors on internal auditors’ fraud risk

assessments
Both internal and external auditing standards encourage internal and

external auditors to coordinate efforts on audit matters (e.g. ASA, 315:
Auditing and Assurance Standards Board, 2013; AU Section 322: PCAOB,
2007; IIA, 2013d). ASA 315 suggests that effective communication between the
internal and external auditors creates an environment in which the external
auditor can be informed of significant matters that may affect the external
auditor’s work. IIA (2013d) suggests the chief auditing executive should share
information and coordinate activities with other internal and external auditors
to ensure sufficient and correct coverage and minimise the duplication of
efforts.
Prior literature that examines internal–external auditor coordination is
mainly from the external auditors’ perspective. The coordination between
internal auditors and external auditors on audit matters can efficiently reduce
budgeted external audit hours, external audit efforts, external audit fees and
audit delay (e.g. Gramling, 1999; Felix et al., 2001; Goodwin-Stewart and
Kent, 2006; Bame-Aldred et al., 2013; Pizzini et al., 2015). The coordination
on the evaluation of internal controls can enhance the overall effectiveness of
the external auditors’ internal control evaluation, leading to greater detection
and disclosure of material control weaknesses (Lin et al., 2011).
We extend prior research by investigating the benefits of coordination to
internal auditors. Effective coordination between internal and external auditors
is viewed as a potentially effective mechanism to mitigate the effect of a poor
tone at the top on assessed fraud risks by internal auditors. We predict that the
influence of management on internal auditors’ fraud risk assessments under
poor tone at the top is mitigated by having the internal audit function
coordinate with external auditors on audit matters7 . The coordination allows
internal auditors to access the external auditors’ techniques, methods and
terminology within the scope of internal auditing. Thus, the coordination is
expected to help internal auditors improve their assurance-related assessments.
This study relies on the theory of informational social influence to derive
predictions on how the coordination with external auditors can mitigate
management influence on internal auditors’ judgements. According to Deutsch
and Gerard (1955), informational social influence is a psychological need to be
7
To our knowledge, there is no research that explicitly examines the impact of tone at
the top on the external auditors’ reliance decision. Desai et al. (2010) hypothesise that
providing external auditors with two of the three primary quality characteristics of the
internal audit function (competence, work performance and objectivity) at a high level,
the external auditors’ perception of the internal audit function remains high. Based upon
these findings, we expect that external auditors would still coordinate with internal
auditors to some extent even when their objectivity is compromised under a poor tone at
the top, providing that competence and work performance remain at a high level.
© 2015 AFAANZ
right when an individual makes a judgement to conform to the expectation of

others. Following this theory, individuals will often look to others for cues
concerning the correct judgements and behaviour when they are in a situation
where they are unsure of the correct way to behave (Aronson et al., 2005).
When the results from the internal audit are unfavourable to the management,
the internal auditors’ confidence in reporting such unfavourable results is
expected to be bolstered by coordinating assurance-related work with the
external auditors. In addition, knowing that their work will be checked and
used by the external auditor, the internal auditors are more likely to conform to
the expectations of the external auditors when making judgements. Therefore,
it is expected that the internal–external auditor coordination mitigates
management’s influence on internal auditors’ judgements when internal
auditors are vulnerable to biased judgements in the face of conflicting interests
as a result of perceived power from the top management. With focus on
internal auditors’ fraud risk assessments, we predict that the existence of
coordination between internal and external auditors will reduce the likelihood
of financial misstatements. This leads to the second hypothesis:
H2: The internal auditor’s assessment on fraud risk is lower when the internal
audit function coordinates with external auditors, compared to when the
internal audit function does not coordinate with external auditors.
The internal auditing standards on coordination suggest that having access to

the external auditors’ techniques, methods and terminology enables the internal
auditors to improve their assurance-related work (IIA, 2013d). We expect that
under a negative tone at the top condition, having internal auditors coordinate
with external auditors will mitigate the impact of a poor management tone on
fraud risk assessment. However, under a good tone at the top, we expect that
internal auditors’ fraud risk assessments will not differ significantly by having
coordination with external auditors. Consequently, we predict that the existence
of coordination between internal and external auditors mitigates the effect of a
poor tone at the top on internal auditors’ fraud risk assessments more than the
effect of a good tone at the top. This leads to the third hypothesis:
H3: The existence of coordination between internal and external auditors has a
larger effect on the internal auditor’s fraud risk assessment when the tone
3. Research method
This study employed a 2 9 2 between-subjects factorial experimental design

with independent variables: (i) tone at the top (good versus poor) and (ii)
coordination between internal and external auditors (exist versus not exist).
© 2015 AFAANZ
3.1 Participants and administration
Participants were internal auditors from various organisations in Aus-

tralia. The experiment instrument was distributed using two methods: a
paper-based mail survey and an online survey8 . Both mail and online
instruments have the same materials and questions. With the paper-based
mail survey, the contact person from each participating organisation was
mailed the different versions of the experimental instrument. The contact
person distributed the different versions to his/her colleagues, collected the
completed responses and then mailed the responses back to the researchers.
With the online survey, through the contact person, participants were
identified and emailed an invitation requesting participation and a web link
that directed them to the online survey. The online system automatically
allocated participants randomly into different groups to show them one of
the four versions of the instruments. Financial incentives were not provided
in this experiment.
The experiment was conducted from May to October 2012. We obtained
responses from 72 of 113 internal auditors contacted via email, yielding a
response rate of 64 percent. With these 72 responses, 16 were from the mail-
based survey and 56 were from the online survey.9 Eight responses are
eliminated from the sample based on a failed manipulation check. All data and
statistics reported are based on the remaining 64 participants.10
8
The instrument was distributed by mail and online because some participating
organisations prefer an online survey to a paper-based survey. For both administration
methods, the instrument was distributed to participants for completion and return
without the presence of the researchers. Although Trotman (1996, p. 91) suggests that a
noncontrolled setting fails to control for factors related to the independent completion
of the instrument and some extraneous variables (e.g. time taken to complete and
surrounding), we used this method to increase the likelihood of gaining participation of
internal auditors in different organisations and locations.
9
The response rate for the mail-based survey is 39 percent (16 responses of 41 mail-
based surveys). The response rate for the online survey is 78 percent (56 responses of 72
online surveys completed). We tested for potential differences in responses collected
from the mail-based and online surveys. The results do not systematically vary between
the two administration methods.
10
Another three participants did not complete the demographic questions. Their
responses to the case questions are determined by the researcher as useable and are
included in the data analyses. Participants were randomly assigned to one of the four
experimental conditions for both mail and online responses. The 72 responses we
obtained were not equally distributed among the four conditions. After excluding the
eight responses that failed manipulation check, each condition contains an equal
number of responses.
© 2015 AFAANZ
3.2 Task
The experimental instrument employs and modifies the experimental case

from Asare et al. (2008), where participants need to make judgements on the
risk of fraudulent financial reporting. According to Hermanson et al. (2008), a
large percentage of companies that were alleged to have senior management-
related weaknesses are manufacturing companies for the period of 2005–2008.
The use of the client description on a hypothetical manufacturing company in
Asare et al. (2008) as a basis for the case description in the current experiment
reflects the current context of internal auditors’ assessments on the likelihood
of senior management-related financial misstatements. The client description
was further modified with red-flag cues that are related to senior management-
related accounting fraud11 .
The experiment consisted of three parts. Part 1 presented a description of a
hypothetical manufacturing company, including the company background, a
description of key personnel, potential risks in credit sales12 and internal
controls in the hypothetical company, and a brief description of the external
audit history. The participants were instructed to assume the role of an in-
house internal auditor in the company. The participants were told that the
internal audit team’s focus was on the reliability and integrity of the financial
information and the internal audit team was planning the audit engagement of
the area of accountants receivable and related accounts. The Chief Audit
Executive reported functionally to the audit committee and administratively to
both the audit committee and the Chief Financial Officer (CFO). The CFO and
the audit committee had joint responsibilities to appoint, reward, evaluate and
dismiss members of the internal audit department.
Part 2 of the experiment consisted of information including a description of
the ethical culture and the coordination between the internal audit function and
external auditors. The order of the ethical culture and the coordination was
pilot-tested to ensure that order of the manipulations did not influence the
results of the experiment. Subsequent to this information, the participants were
asked to make assessment on the likelihood of intentional misstatements
existing in the net accounts receivable balance and the likelihood of
11
The experimental instrument was pilot-tested with 16 academics with a mean
experience of three years and a range of one to eight years of experience in accounting
research and education. We made minor changes to the client description to make
management-related red-flag cues more salient. The instrument was further modified
based on the feedback received from eight internal auditors who have different levels of
experience from a participating organisation. After making minor changes to the client
description and the extent of the coordination between internal and external auditors, all
agreed that the instrument was realistic.
12
The potential risks in credit sales of the case company imply that the senior
management would manipulate the accounts receivable to remain solvent.
© 2015 AFAANZ
management overriding internal controls to manipulate the net accounts

receivable balance.
After providing the above judgements relating to the fraud risk assessment,
the participants responded to manipulation checks and other debriefing
questions in Part 3. These questions were used to assess the understandability
of the case material and the effectiveness of the manipulations. In the end of the
experiment, the participants were asked a number of demographic questions,
including the participants’ work affiliation, work experience, education
background, professional certifications and current position in the employed
organisation.
3.3 Dependent variable
The dependent variable is the internal auditor’s assessment of fraud risk.

Two measures are used to test whether internal auditors’ judgements on fraud
risk are influenced by the tone at the top and the coordination between internal
and external auditors. First, the participants were asked to assess the likelihood
of intentional misstatements existing in the account balances. Given the current
context of the assessment of the likelihood of senior management-related
accounting fraud, a higher assessed likelihood of intentional misstatements
would suggest the participants are more concerned with the risks of senior
management-related accounting fraud. In our task, the participants were asked
to indicate their assessment of the likelihood of an intentional misstatement
existing in their unaudited net accounts receivable balance using a 100-point
Likert scale where ‘0’ indicated that minimum likelihood was assessed, and
‘100’ indicated that maximum likelihood was assessed.
The second measure of the dependent variable is the internal auditor’s
assessment of the likelihood of management overriding controls to manipulate
the account balances. Overriding internal controls is one of the ways senior
management perpetrate financial statement fraud. The internal audit activities
must assist the organisation in maintaining effective controls by evaluating
their effectiveness and efficiency (IIA, 2013e). Overriding internal controls is
considered as a significant control issue (AICPA, 2003), and therefore, internal
auditors should be sensitive to the risk that is posed by management override of
controls.13 Similar to the first measure, the participants were asked to indicate
their assessment of the likelihood of management overriding internal controls
to manipulate their unaudited net accounts receivable account on a 100-point
Likert scale where ‘0’ indicated that minimum likelihood was assessed, and
‘100’ indicated that maximum likelihood was assessed.
13
Interestingly, prior reviews of internal audit research (e.g. Cohen et al., 2004;
Gramling et al., 2004; Stewart and Subramaniam, 2010) consider many aspects of
internal audit but do not specifically consider internal auditors’ need to assess the risk of
management override of controls.
© 2015 AFAANZ
3.4 Independent variables
The first independent variable is the quality of the tone at the top in the
case company. The manipulated tone at the top conditions specifically
focuses on the behaviour of top managers in the workplace. The behaviour
of top managers in the workplace is designed to differentiate a good tone
from a poor tone. According to the Ethics Resource Center’s (2005)
National Business Ethics Survey, employees are much less likely to commit
misconducts when they feel that top management acts ethically in four
important ways. In designing the good tone condition, the experiment
combines four ways of top management’s ethical action reported in the
National Business Ethics Survey (2005): management talks about the
importance of ethics, informs employees about the code of conduct, keeps
promises about rewarding good conduct and models ethical behaviour by
setting examples of good conduct. Specifically, the participants in the good
tone condition were told:
The company has policy statements and a code of conduct which explicitly tell
employees how they should behave in the company. The code of conduct applies to
all employees, including top management. The importance of ethical behavior is
frequently highlighted by management through regular staff meetings. Employees
are encouraged to communicate to their supervisors both “good news” and “poor
news”. Good job performance is well recognised. Top management always rewards
appropriate behavior and addresses inappropriate behavior.
In contrast, in the poor tone condition, the four ways of top manage-
ment’s ethical action are designed to be exactly the opposite of the good
tone condition. Specifically, the participants in the poor tone condition were
told:
The company has policy statements and a code of conduct which include general
guidance of business ethics. The code of conduct applies to all employees, though
top management seems not to be bound by the code of conduct. Employees read
the code of conduct on the first day of their employment and seldom review it
afterwards. The management team is autocratic. Employees are always afraid of
delivering “poor news” to their supervisors. Good job performance is not always
well recognised. Top management does not seem to care about or reward
appropriate behavior and address misbehavior.
The second independent variable is the extent of the coordination between
the internal and external auditors. The internal–external auditor coordination
conditions specifically focus on the existence of coordination of internal and
external audit work. The coordination is designed to encourage internal
auditors to recognise and disclose any material misstatement in the financial
statements. Methods of the coordination are sourced from IIA (2013d) and
Lin et al. (2011), including conducting joint risk or planning sessions,
performing audits of specific processes or accounts and loaning internal audit
© 2015 AFAANZ
staff to the external auditors. Specifically, the participants in the condition where
the coordination between internal and external auditors exists were told:
For years, the internal audit department has coordinated audit work with the external
auditors. The internal audit department shares the information about internal
controls, operations, and financial reporting process with the external auditors. The
external auditors have access to the internal auditors’ programs and working papers.
Joint risk and planning sessions have been scheduled during the audit process to
ensure coordination of audit work and efficient and timely completion of audit
services, and to determine whether observations and recommendations from work
performed to date require that the scope of planned work be adjusted. The external
auditors rely heavily on the work of the internal auditors.
In contrast, the participants in the condition where the coordination between
internal and external auditors does not exist were told:
For years, the internal audit department has not coordinated audit work with the
external auditors. The external auditors have access to the internal auditors’
programs and working papers, yet they put little reliance on the work of the
internal audit activity in performing their work.
3.5 Manipulation and other checks
Manipulation checks were conducted to ensure that participants understand

the tone at the top and the coordination between internal and external auditors’
conditions as intended. For the tone at the top conditions, the participants were
asked to indicate the level of the tone at the top in the case company on an 11-
point Likert scale where ‘0’ indicated that the tone at the top is not good at all, and
‘10’ indicated that the tone at the top is very good. Most participants correctly
identified the level of the tone at the top for their respective condition except for
one participant14 in the poor tone condition who rated the tone at the top above
the mid-point of the tone at the top scale. As expected, the average rating of those
participants in the good tone condition is significantly higher than those in the
poor tone condition (means = 7.37 and 2.53, respectively; p < 0.00115 ).To check
the manipulation on the coordination between internal and external auditors, the
participants were asked to indicate the level of coordination between the internal
and external auditors in the case company on an 11-point Likert scale where ‘0’
represented being no coordination at all, and ‘10’ represented being very good
coordination. Seven participants failed this manipulation check16 . The average
14
Participants who failed the manipulation checks are not included in the data analyses.
15
Results on manipulation and other checks are not tabulated.
16
Two participants who were in the condition where the coordination exists rated the
level of coordination below the mid-point of the scale, while the five participants who
were in the condition where the coordination does not exist rated the level of
coordination above the mid-point of the scale.
© 2015 AFAANZ
rating of those participants in the condition which the coordination exists

between internal and external auditors is significantly higher than those in the
condition where the coordination does not exist (means = 7.15 and 1.94,
respectively; p = 0.004). In total, eight participants were excluded from hypoth-
esis testing.17
Subsequent to the manipulation check questions, the participants answered
additional questions related to the task on 11-point scales. These questions
assess whether the participants correctly interpret the experimental task. The
responses to these questions show that the internal auditors believed that
management will have more opportunities to override internal controls when
the tone at the top is poor and that the internal audit function is more likely to
effectively oversight the financial reporting process when the tone at the top is
good. The participants did not, however, rate the difficulty of remaining
objective differently under the different tone at the top conditions.
4. Results
4.1 Demographics
Demographic data were collected subsequent to the experimental task. The

participants had an average of 5.35 years of internal audit experience. Of the 64
participants, 28 percent indicated they were employed by a nonaccounting firm
that provides outsourced internal audit services, 22 percent by a Big 4
accounting firm, 23 percent by a governmental organisation, and the remaining
by other organisations.18 Among the 64 participants, 23 participants indicated
they had external audit experience with a mean of 5.31 years.
4.2 Test of hypotheses
Prior to undertaking the statistical analyses with each of the two measures of
internal auditors’ fraud risk assessments, we conduct a MANOVA to prevent
possible inflation of the experiment-wise error rate (Hummel and Sligo, 1971).
The overall MANOVA is estimated with the internal auditors’ assessments of
the risk of intentional misstatements and the internal auditors’ assessments of
17
To test the robustness of the results to the exclusion of the responses that failed the
manipulation check, two ANOVAs are conducted by including these responses in the
analysis (N = 72) across the two measures of the dependent variable. The results
consistently show significant main effects of the tone and coordination on the internal
auditors’ fraud risk assessments across the two measures. The interaction effect,
however, is not significant for either measure of the dependent variable.
18
There were 13 percent of the participants employed by for-profit publicly held
companies, 5 percent by a non-Big 4 accounting firm, and 9 percent by non-for-profit
companies.
© 2015 AFAANZ
the risk of management override of controls as the dependent variables. The

tone at the top (good tone versus poor tone) and the coordination between
internal and external auditors (exist versus not exist) are the independent
variables. The results of the overall MANOVA are presented in Table 1.
The MANOVA results indicate a significant main effect for the tone of the
top condition on the combined dependent variables (Wilks’ Lambda19
= 0.428, F = 39.491, p = 0.000). Similarly, there is significant difference
between the coordination-existing condition and the coordination not-existing
condition (Wilks’ Lambda = 0.643, F = 16.355, p = 0.000), suggesting that,
coordination has a significant effect on both measures of the internal auditors’
fraud risk assessments. However, the interaction of the tone at the top and
coordination is not significant in the MANOVA (Wilks’ Lambda = 0.939,
Table 1
Results for a MANOVA of internal auditors’ fraud risk assessments
Effect Wilks’ Lambda F-value df p-value
Intercept 0.067 408.697 2 0.000

Tone at the top 0.428 39.491 2 0.000***
Coordination 0.643 16.355 2 0.000***
Tone at the top 9 coordination 0.939 1.931 2 0.154
Box’s test of equality of covariance matrices
Box’s M 19.500
F-value 2.036
df1 9
df2 41,255.305
Significance 0.032
N 64
Dependent variables: The internal auditor’s assessment of the likelihood of intentional

misstatements is measured on a 100-point scale. The internal auditor’s assessment of the
likelihood of management overriding internal controls is measured on a 100-point scale.
Independent variables: Tone at the top: the tone at the top in the company is manipulated at
two levels: good tone v. poor tone.
Coordination: the existence of coordination between internal and external auditors is
manipulated at two levels: exist v. not exist.
*; **; *** indicate significance at the 10, 5 and 1 per cent levels, respectively.
19
There are a number of multivariate tests of statistically significant differences between
groups. We report the Wilks’ Lambdas because the F-tests for other statistics (e.g.
Pillai’s Trace and Hotelling’s Trace) are identical to Wilks’ Lambdas and according to
Tabachnick and Fidell (2001) Wilks’ Lambda is recommended for general use.
© 2015 AFAANZ
F = 1.931, p = 0.154),20 suggesting that the moderating effect of the coordi-

nation does not vary with the quality of the tone at the top. Further, the
analyses below indicate that this result is dependent upon the measure of fraud
risk assessment chosen.
To separately examine the potential effects of the tone at the top and
coordination on the two measures of internal auditors’ fraud risk assessments,
two separate two-way analyses of variance (ANOVAs) are conducted on
internal auditors’ assessments of the likelihood of intentional misstatements
and the likelihood of management overriding controls, respectively. The results
are reported in the following sections.
4.2.1 Internal auditors’ assessments of the likelihood of intentional misstatements
The primary measurement of internal auditors’ fraud risk assessments is the

assessed likelihood of intentional misstatements. Descriptive statistics across
treatment conditions are reported in Panel A of Table 2 and the results for an
ANOVA are reported in Panel B.
On average, the internal auditors assessed the likelihood of having
intentional misstatements in the task company is 43.59 percent across all
experimental conditions. The pattern of cell means in Panel A indicates
potential main effects of the tone at the top and the coordination between
internal and external auditors. Specifically, the participants in the poor tone
condition assessed the likelihood of intentional misstatements higher
(mean = 57.50) than the participants in the good tone condition
(mean = 29.69). The internal auditors in the coordination not-existing condi-
tion assessed the likelihood of intentional misstatements higher (mean = 52.81)
than the internal auditors in the coordination-existing condition
(mean = 34.38).21
The ANOVA results indicate a significant main effect of the tone at the top
(F = 68.96, p = 0.000) on the internal auditors’ assessed likelihood of inten-
tional misstatements. This finding supports Hypothesis 1b; that is, the internal
auditors are more concerned with the integrity of financial statements when the
tone at the top is poor, compared to when the tone at the top is good. Instead
of being influenced by the poor tone from the top, the internal auditors are
20
The F-tests in MANOVA and ANOVA below are two-tailed. As Hypothesis 3 is
directional, the planned contrast tests on the interaction effect are one-tailed tests.
21
One sample t-tests are conducted to compare cell means for each of the four
treatments (results are not tabulated). The mean difference on the internal auditors’
assessed likelihood of intentional misstatements is significant (t = 6.74, p < 0.001).
The mean difference on assessed likelihood of intentional misstatements is also highly
significant between the two conditions (t = 3.76, p < 0.001). Overall, the results on the
t-tests are consistent with the predictions in Hypothesis 1b and 2.
© 2015 AFAANZ
Table 2
Internal auditors’ assessments on the likelihood of intentional misstatements
Panel A: Mean of assessments of the risk of intentional misstatements (standard deviations) by

treatment conditions
Good tone Poor tone Overall
Coordination exists 23.75 45.00 34.38

(11.47) (16.73) (17.77)
N = 16 N = 16 N = 32
Coordination 35.63 70.00 52.81
does not exist (15.04) (8.94) (21.29)
N = 16 N = 16 N = 32
Overall 29.69 57.50 43.59
(14.48) (18.32) (21.56)
N = 32 N = 32 N = 64
Panel B: Test of hypotheses by treatment conditions
Source Type III sum of squares df Mean square F-value p-value
Corrected model 18,504.687 3 6168.229 34.367 0.000

Intercept 121,626.563 1 121,626.563 677.664 0.000
Tone at the top 12,376.563 1 12,376.563 68.958 0.000***
Coordination 5439.063 1 5439.063 30.305 0.000***
Tone at the top 689.063 1 689.063 3.839 0.045**
9 coordination
Error 10,768.750 60
Total 150,900.000 64
Corrected total 29,273.437 63
Panel C: Ordinal interaction contrast coding test
Contrast Weights
Good tone Poor tone

Tone at the top:
Hypothesis Coordination: Exist Not exist Exist Not exist T statistics p-value#
H3 1 1 0 2 9.828 0.000 ***
Dependent variable: The internal auditor’s assessment of the likelihood of intentional

misstatements is measured on a 100-point scale.
two levels: good tone versus poor tone.
manipulated at two levels: exist versus not exist.
#
One-tailed probability levels.
© 2015 AFAANZ
more skeptical in assessing fraud risk by considering the poor tone at the top as
a signal to potential fraud.
The significant main effect of the coordination between internal and external
auditors in Panel B (F = 30.30, p = 0.000) and the associated means in Panel A
of Table 2 suggest that the coordination has a significant impact on the internal
auditors’ assessments on the likelihood of intentional misstatements. Consis-
tent with Hypothesis 2, the finding indicates that the internal auditors’
suspicion about the existence of intentional misstatements is higher when there
is no coordination between internal and external auditors, compared to when
there is such coordination. That is, having the internal audit function to
coordinate assurance-related activities with external auditors is perceived to be
associated with fewer financial misstatements.
The interaction of tone at the top and coordination is significant at the five
percent level in the ANOVA (F = 3.84, p = 0.045). To examine this interaction
effect further, we conduct planned contrasts tests. Hypothesis 3 predicts that
the existence of coordination between internal and external auditors can
moderate the effect of a poor tone at the top on internal auditors’ fraud risk
assessments more than the effect of a good tone at the top. We expect that the
mean difference of the internal auditors’ fraud risk assessments between the
poor tone-coordination condition and poor tone-no coordination condition
will be greater than the mean difference between good tone-coordination
condition and good tone-no coordination condition. Given that we also expect
that the internal auditors’ fraud risk assessments will not differ significantly
under a good tone at the top by having coordination with external auditors, we
assign the same weight to the two means of fraud risk assessments under the
good tone condition. Therefore, we use the following contrast codes (in
parentheses) to examine the ordinal interaction: Good tone-Coordination ( 1);
Good tone-No coordination ( 1); poor tone-coordination (0); poor tone-no
coordination (2)22 .
The results using planned contrasts are shown in Panel C of Table 2. An
ordinal interaction contrast coding which reflects our specific directional
predictions shows a significant result, supporting Hypothesis 3 (t = 9.828,
p = 0.000) on the moderating effect of the existence of coordination. That is,
the existence of the coordination between internal and external auditors has a
larger downward effect on the internal auditors’ assessments on the likelihood
of intentional misstatements when the tone at the top is poor than when the
tone at the top is good.
We present a plot of the mean for each treatment group as Figure 1. As
shown in Figure 1, the difference in the mean assessments on the likelihood of
intentional misstatements across the coordination conditions is larger when the
tone is poor, compared to when the tone is good. Based upon the statistical
22
The statistical inference remains unchanged when we employ the alternative contrast
weights ( 2, 2, 1, 3).
© 2015 AFAANZ
80
The likelihood of intentional misstatements
70
60
50
Coordination does not
40 exist
Coordination exists
30
20
10
0
Poor tone Good tone
Figure 1 Interaction plot on the assessed likelihood of intentional misstatements. Dependent

variable: The internal auditor’s assessment of the likelihood of intentional misstatements is
measured on a 100-point scale. Independent variables: Tone at the top: the tone at the top in the
company is manipulated at two levels: good tone v. poor tone. Coordination: the existence of
coordination between internal and external auditors is manipulated at two levels: exist v. not exist.
significant interaction and the pattern of means in Figure 1, our results suggest
that effective coordination between internal and external auditors is viewed as
an effective mechanism to mitigate the effect of poor tone at the top on assessed
risk of material misstatements by the internal auditors.
4.2.2 Internal auditors’ assessments of the likelihood of management override of

controls
As a secondary measure of the dependent variable, we also examine the

internal auditors’ assessments on the risk of management overriding internal
controls as a subset of the internal auditors’ fraud risk assessments. The results
are reported in Table 3 with the descriptive statistics reported in Panel A and
ANOVA results reported in Panel B. The results from the descriptive statistics,
and t-tests on the mean differences, show that the participants in the poor tone
condition assessed the likelihood of management overriding internal controls
higher than the participants in the good tone condition (t = 6.85, p < 0.001).
The internal auditors in the coordination not-existing condition assessed the
risk of management overriding internal controls higher than the internal
auditors in the coordination-existing condition (t = 3.49, p < 0.001). Panel B
of Table 3 shows a significant main effect for the tone at the top (F = 66.403,
p = 0.000), and the existence of coordination (F = 25.281, p = 0.000). The
results are consistent with the primary measure, supporting Hypothesis 1b and
© 2015 AFAANZ
2. That is, a poor tone at the top is considered as an important signal to the risk
of management overriding by the internal auditors because a poor tone may
provide more opportunities for the management to override internal controls.
The internal auditors are more concerned with the risk of management
overriding when there is no coordination between internal and external
auditors.
The interaction effect of the tone at the top and coordination on the internal
auditors’ assessed likelihood of management overriding controls is not
statistically significant (F = 2.43, p = 0.124). Buckless and Ravenscroft (1990)
suggests that contrast coding provides greater statistical power than the
conventional ANOVA when examining ordinal interactions. As we predict an
ordinal interaction effect in Hypothesis 3, we use contrast coding to further
examine this interaction effect. Similar to the primary measure of the dependent
variable, we employ the same contrast coding.23 Planned contrasts in Panel C
of Table 3 highlight significant ordinal interactive effect between the tone at the
top and coordination with external auditors on the judgement of likelihood of
management overriding controls. Using the same contrast weights to each
treatment as of the primary measure, the results on the planned contrasts test
reveal that our specific directional prediction is significant (t = 9.343,
p = 0.000). That is, the existence of the coordination moderates the impact
of a poor tone at the top on the internal auditors’ assessments on the likelihood
of management overriding controls.
4.3 Additional analysis
Using the responses to the manipulation check questions as covariates24 , we

conduct three separate ANCOVAs to determine whether the perceptions of
tone at the top, coordination, and the overall risk level in the case company
would influence the hypothesis tests. We find that the results from the main test
are robust after controlling overall risk level in the case company. Neither the
perception of tone nor coordination significantly affects the internal auditors’
fraud risk assessments.
We also conduct an ANCOVA to determine whether the participants’
accounting and auditing experience would influence the hypothesis tests. The
dependent variable is the internal auditors’ fraud risk assessments. The
independent variables represent the manipulations of the tone at the top and
the coordination between internal and external auditors. Covariates are
included for the participants’ working experience in external auditing, internal
23
We employ an alternative contrast weights ( 2, 2, 1, 3). The statistical inference
remains unchanged.
24
The participants were asked to rate their perception of the quality of the tone at the
top, the level of coordination with external auditors and the overall fraud risk in the case
company. The ratings on these questions are used as covariates in the ANCOVAs.
© 2015 AFAANZ
Table 3
Internal auditors’ assessments on the likelihood of management overriding controls
Panel A: Mean of assessments of the likelihood of management overriding internal controls

(standard deviations) by treatment conditions
Good tone Poor tone Overall

exists (14.38) (19.62) (20.78)
N = 16 N = 16 N = 32
does not exist (12.65) (8.94) (20.79)
N = 16 N = 16 N = 32
Overall 33.75 63.13 48.43
(14.76) (19.25) (22.55)
N = 32 N = 32 N = 64
Panel B: Test of hypotheses by treatment conditions
Source Type III sum of squares df Mean square F-value p-value
Corrected model 19,568.750 3 6522.917 31.373 0.000

Intercept 150,156.250 1 150,156.250 722.194 0.000
Tone at the top 13,806.250 1 13,806.250 66.403 0.000***
Coordination 5256.250 1 5256.250 25.281 0.000***
Tone at the top 9 506.250 1 506.250 2.435 0.124
Coordination
Error 12,475.000 60 207.917
Total 182,200.000 64
Corrected total 32,043.750 63
Panel C: Ordinal interaction contrast coding test
Contrast Weights
Good tone Poor tone

Tone at the top:
Hypothesis Coordination: Exist Not exist Exist Not exist T statistics p-value#
H3 1 1 0 2 9.343 0.000***
Dependent variable: The internal auditor’s assessment of the likelihood of management

overriding the internal controls is measured on a 100-point scale.
two levels: good tone versus poor tone.
manipulated at two levels: exist versus not exist.
#
One-tailed probability levels.
© 2015 AFAANZ
auditing, public companies and manufacturing companies. None of these

experience factors are significant in this analysis and affect the results reported
earlier.
The participants’ education level, professional licences and current position
in the organisation are also individually included in three one-way ANOVAs as
an independent variable. The results show no significant differences in
education levels and the current positions that the participants were holding
across the two measures of the internal auditors’ fraud risk assessments. The
participants’ professional licence is found to have a significant effect on their
assessed likelihood of management overriding controls. The results from the
multiple comparisons on simple main effects indicate that the participants who
were holding Certificate in Control Self-Assessment (CCSA)25 assessed a higher
likelihood of management overriding controls than participants holding other
accounting and auditing certifications26 . Consistent with Bedard and Graham
(2011) which posit that outside consultants have greater expertise in control
assessments that improves the detection and disclosure of control problems,
our finding suggests that the internal auditors who have specific knowledge
about the risks and controls that generally apply to business processes are more
sensitive to the risk of management overriding controls.
5. Conclusion
If top management has significant influence over internal auditors, then

internal auditors may compromise their judgements to appease management
(Harrell et al., 1989). This study investigates whether top management’s
attitudes towards ethics and integrity influences internal auditors’ fraud risk
assessments. We specifically examine whether internal auditors’ assessments on
the likelihood of senior management-related financial misstatements are biased
towards the favour of the management under a poor tone at the top in which
the top management does not uphold ethics and integrity and communicates
negative values to their employees.
We find that the internal auditors’ assessed fraud risk is higher when the tone
at the top is poor, relative to when the tone at the top is good. This finding runs
counter to any bias in internal auditors’ judgements from management
influence. Our results reveal that internal auditors are more concerned with the
integrity of the financial statements under a poor tone at the top, suggesting
25
Certificate in Control Self-Assessment (CCSA) is a specialty certification for
practitioners of control self-assessment (CSA) by the Institute of Internal Auditors.
CCAS candidates must obtain one year of control-related business experience, such as
CSA, auditing, quality assurance, risk management or environmental auditing.
26
These certifications include Certified Practicing Accountant (CPA), Chartered
Accountant (CA) and Certified Internal Auditor (CIA).
© 2015 AFAANZ
that they perceive a poor tone at the top as an important signal to potential
financial misstatements.
We also investigate the benefit of having internal auditors coordinate
assurance-related activities with external auditors on internal auditors’ fraud
risk assessments. We find that the internal auditors’ assessed fraud risk is lower
when there is coordination with external auditors, compared to when such
coordination does not exist. The finding suggests that internal auditors perceive
the coordination with external auditors as a means to reduce potential financial
problems and to enhance the quality of financial statements.
This study is subject to several limitations. First, while we follow previous
research methodology, the concept of management influence over internal
auditors is difficult to operationalise to an experimental setting. In the light of
recent corporate frauds, internal auditors may be more wary of advocating for
management positions when making judgements in an experimental setting.
Second, while our experiment uses an in-house internal audit setting, prior
studies on the source of the internal audit service suggest that an outsourced
internal audit function is perceived to be more independent and competent
(DeZoort et al., 2001; Del Vecchio and Clinton, 2003; Carey et al., 2006;
Glover et al., 2008; Desai et al., 2011; Prawitt et al., 2012; Koch et al., 2015).
Although the results from the sensitivity analyses suggest no significant
differences between participants classified by employer, future research could
consider differences in perceptions arising from the location of the internal
audit function. Third, the current experiment does not measure the confidence
level of the internal auditors when they are making fraud risk assessments.
Finally, the manipulation of tone at the top is designed to be relatively strong.
While having such a strong manipulation helps maintain a high level of internal
validity, the design of the manipulation in the current experiment does not
capture the many subtle characteristics of tone at the top in an actual
organisational setting. Much further research is needed from a variety of
methodological perspectives to better understand the influence of tone at the
top on the role and decisions of the internal auditor, and indeed how the
internal audit function can serve to improve the tone at the top.
Acknowledgments
We are grateful for helpful suggestions from participants of the Australian

National Centre for Audit and Assurance Research Forum 2012, the Forensic
and Investigation Accounting Conference 2013, the Accounting and Finance
Association of Australian and New Zealand Conference 2013, and the
International Symposium on Audit Research 2014. We thank Gary Monroe
for his comments. This research is derived from a doctoral thesis at The
Australian National University. We gratefully acknowledge comments from
the thesis examiners.
© 2015 AFAANZ
References
Ahlawat, S. S., and D. J. Lowe, 2004, An examination of internal auditor objectivity: in-
house versus outsourcing, Auditing: A Journal of Practice and Theory 23(2), 147–158.
American Institution of Certified Public Accountants (AICPA), 2003, Consideration of
Fraud in a Financial Statement Audit (Supersedes SAS No. 82), Statement on Auditing
Standard (SAS) No. 99. (AICPA, New York, NY).
Apostolou, B. A., J. M. Hassell, S. A. Webber, and G. E. Sumners, 2001, The relative
importance of management fraud risk factors, Behavioral Research in Accounting 13
(1), 1–24.
Aronson, E., T. D. Wilson, and A. M. Akert, 2005, Social Psychology, 5th edn. (Prentice
Hall, Upper Saddle River, NJ).
Asare, S. K., R. A. Davidson, and A. A. Gramling, 2008, Internal auditors’ evaluation
of fraud factors in planning an audit: the importance of audit committee quality and
management incentives, International Journal of Auditing 12(3), 181–203.
Association of Certified Fraud Examiners (ACFE), 2006, Tone at the top: How
management can prevent fraud in the workplace. Available at: www.acfe.com/doc-
uments/tone-at-the-top-research.pdf
Auditing and Assurance Standards Board, 2013, Australian Auditing Standard (ASA)
315: Identifying and Assessing the Risks of Material Misstatement through
Understanding the Entity and Its Environment (Compiled). Available at http://
www.auasb.gov.au/Pronouncements/Australian-Auditing-Standards.aspx
Bame-Aldred, C. W., D. M. Brandon, M. F. Messier, L. E. Rittenberg, and C. M.
Stefaniak, 2013, A summary of research on external auditor reliance on the internal
audit function, Auditing: A Journal of Practice and Theory 25(Supplement 1),
251–286.
Bartlett, A., and D. Preston, 2000, Can ethical behavior really exist in business?, Journal
of Business Ethics 23, 199–209.
Beasley, M. S., R. Clune, and D. R. Hermanson, 2005, ERM: a status report, Internal
Auditing 20(1), 67–72.
Bedard, J. C., and L. Graham, 2011, Detection and severity classifications of
Sarbanes-Oxley Section 404 internal control deficiencies, The Accounting Review 86
(3), 825–855.
Brody, R. G., and D. J. Lowe, 2000, The new role of the internal auditor: implications
for internal auditor objectivity, International Journal of Auditing 4(2), 169–176.
Buckless, F. A., and S. P. Ravenscroft, 1990, Contrast coding: a refinement of ANOVA
in behavioral analysis, The Accounting Review 65(4), 933–945.
Burton, F. G., M. W. Starliper, S. L. Summer, and D. A. Wood, 2015, The effects of
using the internal audit function as a management training ground or as a consulting
services provider in enhancing the recruitment of internal auditors, Accounting
Horizons 29(1), 115–140.
Carey, P., N. Subramaniam, and K. C. W. Ching, 2006, Internal audit outsourcing in
Australia, Accounting and Finance 46(1), 11–30.
Christopher, J., G. Sarens, and P. Leung, 2009, A critical analysis of the independence of
the internal audit function: evidence from Australia, Accounting, Auditing and
Accountability Journal 22(2), 200–220.
Church, B. K., J. J. McMillan, and A. Schneider, 2001, Factors affecting internal
auditors’ consideration of fraudulent financial reporting during analytical procedures,
Auditing: A Journal of Practice and Theory 20(1), 65–80.
Cohen, J. R., G. Krishnamoorthy, and A. Wright, 2004, The corporate governance
mosaic and financial reporting quality, Journal of Accounting Literature 23, 87–152.
© 2015 AFAANZ
Committee of Sponsoring Organizations of the Treadway Commission (COSO), 1987,

Report of the National Commission on Fraudulent Financial Reporting (NCFFR). (US
Government Printing Office, Washington, DC).
Coram, P., C. Ferguson, and R. Moroney, 2008, Internal audit, alternative internal
audit structures and the level of misappropriation of assets fraud, Accounting and
Finance 48(4), 543–559.
COSO, 2012, Internal Control – Integrated Framework, Exposure Draft. Available at:
http://www.ic.coso.org.
COSO, 2013, Internal Control – Integrated Framework, Executive Summary. (AICPA,
New York, NY).
Dechow, P. M., W. Ge, C. R. Larson, and R. G. Sloan, 2011, Predicting material
accounting manipulations, Contemporary Accounting Research 28(1), 17–82.
Del Vecchio, S., and B. D. Clinton, 2003, Co-sourcing and other alternatives in
acquiring internal audit services, Internal Auditing 18(May/June), 33–39.
Desai, V. R., W. Roberts, and R. Srivastava, 2010, An analytical model for external
auditor evaluation of the internal audit function using belief functions, Contemporary
Accounting Research 27(2), 537–575.
Desai, N. K., G. J. Gerard, and A. Tripathy, 2011, Internal audit sourcing arrangements
and reliance by external auditors, Auditing: A Journal of Practice and Theory 30(1),
149–171.
Deutsch, M., and H. B. Gerard, 1955, A study of normative and informational social
influences upon individual judgment, Journal of Abnormal and Social Psychology 51,
629–636.
DeZoort, F. T., R. W. Houston, and M. F. Peters, 2001, The impact of internal auditor
compensation and role on external auditors’ planning judgments and decisions,
Contemporary Accounting Research 18(2), 257–281.
Ethics Resource Center, 2005, National Business Ethics Survey: how employees view
ethics in their organizations 1994–2005. Available at http://www.ethics.org/
nbes/download.html
Everett, J., and M. Tremblay, 2014, Ethics and internal audit: moral will and moral skill
in a heteronomous field, Critical Perspectives on Accounting 25(3), 181–196.
Felix, W. L., A. A. Gramling, and M. J. Maletta, 2001, The contribution of internal
audit as a determinant of external audit fees and factors influencing this contribution,
Journal of Accounting Research 39(3), 513–534.
French, J., and B. Raven, 1959, The bases of social power, in: D. Cartwright, A. Zander,
eds. Group Dynamics (Row, Petersen, Evanston, IL).
Gaa, J. C., 1992, The auditor’s role: The philosophy and psychology of indepen-
dence and objectivity, Proceedings of the 1992 Deloitte and Touche/University of
Kansas Symposium on Auditing Problems. (University of Kansas, Lawrence, KS),
7–43.
Glover, S. M., D. F. Prawitt, and D. A. Wood, 2008, Internal audit sourcing
arrangement and the external auditor’s reliance decision, Contemporary Accounting
Research 25(1), 193–213.
Goodwin-Stewart, J., and P. Kent, 2006, Relation between external audit fees,
audit committee characteristics and internal audit, Accounting and Finance 46(3),
387–404.
Gramling, A. A., 1999, External auditors’ reliance on work performed by internal
auditors: the influence of fee pressure on this reliance decision, Auditing: A Journal of
Practice and Theory 18(Supplement), 117–135.
Gramling, A. A., and P. M. Myers, 2006, Internal auditing’s role in ERM, Internal
Auditing 63(April), 52–58.
© 2015 AFAANZ
Gramling, A. A., M. J. Maletta, A. Schneider, and B. K. Church, 2004, The role of the
internal audit function in corporate governance: a synthesis of the extant internal
auditing literature and directions for future research, Journal of Accounting Literature
23, 194–244.
Hansen, J., N. M. Stephens, and D. A. Wood, 2009, Entity-level controls: the internal
auditor’s assessment of management tone at the top, Current Issues in Auditing 3(1),
1–13.
Harrell, A., M. Taylor, and E. Chewning, 1989, An examination of management’s
ability to bias the professional objectivity of internal auditors, Accounting, Organi-
zations and Society 14(3), 259–269.
Hermanson, D. R., D. M. Ivancevich, and S. H. Ivancevich, 2008, Tone at the top:
insights from Section 404, Strategic Finance 90(5), 39–45.
Hummel, T. J., and J. R. Sligo, 1971, Empirical comparison of univariate and
multivariate analysis of variance procedures, Psychological Bulletin 76, 46–57.
Institute of Internal Auditors (IIA), 2013a, Attribution Standard 1120: Individual
Objectivity, The International Standards for the Professional Practice of Internal
Auditing. (Institute of Internal Auditors Research Foundation, Altamonte Springs,
FL).
IIA, 2013b, Attribute Standard 1220: Due Professional Care, The International Standards
for the Professional Practice of Internal Auditing. (Institute of Internal Auditors
Research Foundation, Altamonte Springs, FL).
IIA, 2013c, Performance Standard 2120: Risk Management, The International Standards
for the Professional Practice of Internal Auditing. (Institute of Internal Auditors
Research Foundation, Altamonte Springs, FL).
IIA, 2013d, Performance Standard 2050: Coordination, The International Standards for
the Professional Practice of Internal Auditing. (Institute of Internal Auditors Research
Foundation, Altamonte Springs, FL).
IIA, 2013e, Performance Standard 2130: Control, The International Standards for the
Professional Practice of Internal Auditing. (Institute of Internal Auditors Research
Foundation, Altamonte Springs, FL).
IIA, 2013f, Glossary, The International Standards for the Professional Practice of Internal
Auditing. (Institute of Internal Auditors Research Foundation, Altamonte Springs,
FL).
Koch, S., M. Gros, and C. Wallek, 2015, Internal audit function quality and the financial
reporting process: results of a survey on German listed companies, Paper presented at
the European Accounting Association 38th Annual Congress, Glasgow UK.
Lasswell, H. D., and A. Kaplan, 1950, Power and Society: A Framework for Political
Inquiry, Vol. 2, (Yale University Press, New Haven, CT).
Lenz, R., and U. Hahn, 2015, A synthesis of empirical internal audit effectiveness
literature pointing to new research opportunities, Managerial Auditing Journal 30(1),
5–33.
Lin, S., M. Pizzini, M. Vargus, and I. R. Bardhan, 2011, The role of the internal audit
function in the disclosure of material weaknesses, The Accounting Review 86(1), 287–
323.
Lord, A. T., and T. F. DeZoort, 2001, The impact of commitment and moral reasoning
on auditors’ responses to social influence pressure, Accounting, Organizations and
Society 26(3), 215–235.
Mason, G. S., and L. G. Levy, 2001, The use of the latent constructs method in
behavioral accounting research: the measurement of client advocacy, Advances in
Taxation 13, 123–129.
© 2015 AFAANZ
Mast, M. S., J. A. Hall, and P. C. Schmid, 2010, Wanting to be boss and wanting to be
subordinate: effects on performance motivation, Journal of Applied Social Psychology
40(2), 458–472.
Messier, W. F., J. K. Reynolds, C. A. Simon, and D. A. Wood, 2011, The effect of using
the internal audit function as a management training ground on the external auditor’s
reliance decision, The Accounting Review 86(6), 2131–2154.
Munro, L., and J. Stewart, 2010, External auditors’ reliance on internal audit: the
impact of sourcing arrangements and consulting activities, Accounting and Finance 50
(2), 371–387.
Norman, C. S., A. M. Rose, and J. M. Rose, 2010, Internal audit reporting lines, fraud
risk decomposition, and assessments of fraud risk, Accounting, Organizations and
Society 35(5), 546–557.
Patelli, L., and M. Pedrini, 2015, Is tone at the top associated with financial reporting
aggressiveness?, Journal of Business Ethics 126, 3–19.
PCAOB, 2007, The auditor’s consideration of the internal audit function in an audit of
financial statements (AU Section 322). Available at: http://pcaobus.org/Standards/
Auditing/Pages/AU322.aspx
Pizzini, M., S. Lin, and D. E. Ziegenfuss, 2015, The impact of internal audit function
quality and contribution on audit delay, Auditing: A Journal of Practice and Theory 34
(1), 25–58.
Prawitt, D. F., N. Y. Sharp, and D. A. Wood, 2012, Internal audit outsourcing and the
risk of misleading or fraudulent financial reporting: did Sarbanes-Oxley get it wrong?,
Contemporary Accounting Research 29(4), 1109–1136.
Rodgers, W., and S. M. Al Fayi, 2015, Ethical positions influencing boundary span
between board/CEO and internal audit reporting, Paper presented at the European
Accounting Association 38th Annual Congress, Glasgow UK.
Rose, A. M., J. M. Rose, and C. S. Norman, 2013, Is the objectivity of internal audit
compromised when the internal audit function is a management training ground?,
Accounting and Finance 53(4), 1001–1019.
Roussy, M., 2013, Internal auditors’ roles: from watchdogs to helpers and protectors of
the top manager, Critical Perspectives on Accounting 24, 550–571.
Roussy, M., 2015, Welcome to the day-to-day of internal auditors: how do they cope
with conflicts? Auditing: A Journal of Practice and Theory 34(2), 237–264.
Schwarts, M. S., T. W. Dunfee, and M. J. Kline, 2005, Tone at the top: an ethics code
for directors?, Journal of Business Ethics 58(1–3), 79–100.
Stewart, J., and N. Subramaniam, 2010, Internal audit independence and objectivity:
emerging research opportunities, Managerial Auditing Journal 25(4), 328–360.
Tabachnick, B. G., and L. S. Fidell, 2001, Using Multivariate Statistics, 4th edn.
(HarperCollins, New York, NY).
Trompeter, G. M., T. D. Carpenter, N. Desai, K. L. Jones, and R. A. Riley, 2013, A
synthesis of fraud-related research, Auditing: A Journal of Practice and Theory 32
(Supplement), 287–321.
Trotman, K. T., 1996, Research Methods for Judgment and Decision making Studies in
Auditing (Coopers and Lybrand and Accounting Association of Australia and New
Zealand, Melbourne, Australia).
Trotman, A. J., and K. T. Trotman, 2014, Internal audit’s role in GHG emissions and
energy reporting: evidence from audit committees, senior accountants and internal
auditors, Auditing: A Journal of Practice and Theory 34(1), 199–230.
Zain, M. M., and N. Subramaniam, 2007, Internal auditor perceptions on audit
committee interactions: a qualitative study in Malaysian public corporations,
Corporate Governance: An International Review 15(5), 894–908.
© 2015 AFAANZ

BFA302 Assessment 1 Article

Încărcat de

Informații document

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

BFA302 Assessment 1 Article

Încărcat de

Drepturi de autor:

Formate disponibile

Accounting & Finance 57 (2017) 1177–1202

The eﬀects of tone at the top and coordination with external

Isabel Z. Wanga, Neil Fargherb

Prior research suggests that internal auditors’ judgements are subject to

JEL classification: M41, M42

2. Theoretical background and hypotheses

2.1 Tone at the top on internal auditors’ fraud risk assessments

We develop two competing hypotheses regarding the association between

Conversely, from the viewpoint of professional commitment, the commitment

2.2 Coordination with external auditors on internal auditors’ fraud risk

Both internal and external auditing standards encourage internal and

right when an individual makes a judgement to conform to the expectation of

The internal auditing standards on coordination suggest that having access to

This study employed a 2 9 2 between-subjects factorial experimental design

3.1 Participants and administration

Participants were internal auditors from various organisations in Aus-

The experimental instrument employs and modiﬁes the experimental case

management overriding internal controls to manipulate the net accounts

3.3 Dependent variable

The dependent variable is the internal auditor’s assessment of fraud risk.

3.4 Independent variables

3.5 Manipulation and other checks

Manipulation checks were conducted to ensure that participants understand

rating of those participants in the condition which the coordination exists

Demographic data were collected subsequent to the experimental task. The

4.2 Test of hypotheses

the risk of management override of controls as the dependent variables. The

Eﬀect Wilks’ Lambda F-value df p-value

Intercept 0.067 408.697 2 0.000

Box’s test of equality of covariance matrices

Dependent variables: The internal auditor’s assessment of the likelihood of intentional

F = 1.931, p = 0.154),20 suggesting that the moderating eﬀect of the coordi-

4.2.1 Internal auditors’ assessments of the likelihood of intentional misstatements

The primary measurement of internal auditors’ fraud risk assessments is the

Panel A: Mean of assessments of the risk of intentional misstatements (standard deviations) by

Good tone Poor tone Overall

Coordination exists 23.75 45.00 34.38

Panel B: Test of hypotheses by treatment conditions

Source Type III sum of squares df Mean square F-value p-value

Corrected model 18,504.687 3 6168.229 34.367 0.000

Good tone Poor tone

H3 1 1 0 2 9.828 0.000 ***

Dependent variable: The internal auditor’s assessment of the likelihood of intentional

Figure 1 Interaction plot on the assessed likelihood of intentional misstatements. Dependent

4.2.2 Internal auditors’ assessments of the likelihood of management override of

As a secondary measure of the dependent variable, we also examine the

4.3 Additional analysis

Using the responses to the manipulation check questions as covariates24 , we

Panel A: Mean of assessments of the likelihood of management overriding internal controls

Good tone Poor tone Overall

Coordination 27.50 51.25 39.38

Panel B: Test of hypotheses by treatment conditions

Source Type III sum of squares df Mean square F-value p-value

Corrected model 19,568.750 3 6522.917 31.373 0.000

Good tone Poor tone

Dependent variable: The internal auditor’s assessment of the likelihood of management

auditing, public companies and manufacturing companies. None of these

If top management has signiﬁcant inﬂuence over internal auditors, then

We are grateful for helpful suggestions from participants of the Australian

Committee of Sponsoring Organizations of the Treadway Commission (COSO), 1987,

S-ar putea să vă placă și