Description
We examine the effect of perspective taking on auditors’ ability to evaluate managers’ reported earnings
and, in turn, contribute to high-quality financial reporting. Using an experimental-economics approach,
we design two experiments to investigate auditor – manager interactions. In our first experiment, we
manipulate auditors’ prior experience in the manager’s role. We predict and find that role-taking experience
stimulates perspective taking, which allows auditors to more readily put themselves ‘‘in the manager’s
shoes,’’ benefitting financial-reporting quality. In our second experiment, we examine dispositional
perspective taking, focusing on individuals’ propensity to spontaneously take the viewpoint of another,
as a dimension of personality.
Perspective taking in auditor–manager interactions: An experimental
investigation of auditor behavior
q
Bryan K. Church
a,1
, Marietta Peytcheva
b,?,1
, Wei Yu
c,1
, Ong-Ard Singtokul
d
a
Scheller College of Business, Georgia Institute of Technology, United States
b
Department of Accounting, College of Business and Economics, Lehigh University, United States
c
Department of Economics, Hunter College, City University of New York, United States
d
Chulalongkorn Business School, Chulalongkorn University, Charoen Pokphand Group Company Limited, Thailand
a r t i c l e i n f o
Article history:
Received 17 September 2012
Revised 20 April 2015
Accepted 7 July 2015
Available online 25 July 2015
Keywords:
Perspective taking
Role-taking experience
Individual difference
Experimental economics
a b s t r a c t
We examine the effect of perspective taking on auditors’ ability to evaluate managers’ reported earnings
and, in turn, contribute to high-quality ?nancial reporting. Using an experimental-economics approach,
we design two experiments to investigate auditor – manager interactions. In our ?rst experiment, we
manipulate auditors’ prior experience in the manager’s role. We predict and ?nd that role-taking expe-
rience stimulates perspective taking, which allows auditors to more readily put themselves ‘‘in the man-
ager’s shoes,’’ bene?tting ?nancial-reporting quality. In our second experiment, we examine dispositional
perspective taking, focusing on individuals’ propensity to spontaneously take the viewpoint of another,
as a dimension of personality. We predict and ?nd that auditors with high perspective-taking disposition
are better able to judge managers’ reported earnings than auditors with low perspective-taking disposi-
tion. Taken together, the results of our two experiments highlight the importance of perspective taking as
a means to enhance auditors’ performance in strategic interactions with managers.
Ó 2015 Elsevier Ltd. All rights reserved.
Introduction
This paper reports the results of two experiments designed to
examine the effect of perspective taking on auditors’ ability to pro-
mote high-quality ?nancial reporting. We de?ne perspective tak-
ing as the capacity to entertain the psychological point of view of
another (Davis, Conklin, Smith, & Luce, 1996). Our focus is on cog-
nitive perspective taking, which entails understanding, as accu-
rately as possible, another’s thoughts, attitudes, or concerns in a
speci?c situation (Epley, Caruso, & Bazerman, 2006). We use an
experimental-economics approach to examine auditor–manager
interactions, where players have con?icting interests. Speci?cally,
we investigate how perspective taking affects auditors’ assessment
of managers’ reporting choices, including auditors’ propensity to
identify and curtail reporting bias. We contend that successful per-
spective taking allows auditors to develop better mental models of
clients’ earnings, which leads to enhanced ?nancial-reporting
quality (Peecher, Schwartz, & Solomon, 2007).
Prior studies suggest that effective perspective taking improves
individuals’ judgments and decision making. Successfully taking
another’s perspective can reduce anchoring effects, con?rmation
bias, actor–observer bias, and in-group favoritism (Galinsky &
Mussweiler, 2001). Moreover, being able to take the perspective
of one’s counterpart leads to more bene?cial outcomes for self
(Galinsky & Mussweiler, 2001). In an auditing context, Altiero,
Kang, and Peecher (2014) document that auditors who are
prompted to take an investor’s perspective, by completing a series
of investor-minded tasks, provide higher-quality materiality judg-
ments than auditors who are not prompted. Our study comple-
ments Altiero et al. (2014) by examining the linkage between
perspective taking and auditor behavior in strategic interactions
with managers.
In our ?rst experiment, we examine whether role-taking
experience stimulates auditors’ perspective taking. We contendhttp://dx.doi.org/10.1016/j.aos.2015.07.001
0361-3682/Ó 2015 Elsevier Ltd. All rights reserved.
q
We gratefully acknowledge helpful comments from Jim Dearden, Jeffrey Hales,
Mark Peecher, Timothy Shields, T. Jeffrey Wilks, Ping Zhang, and two anonymous
reviewers, as well as participants in a workshop at Rutgers University and the 2010
Auditing Section Midyear Meeting in San Diego, CA. The authors acknowledge the
?nancial support by the Scheller College of Business, Georgia Institute of Technol-
ogy, a Faculty Research Grant and a grant from the Martindale Center at Lehigh
University, and a research grant from Haslam College of Business, the University of
Tennessee. We also appreciate the research assistance of John Castonguay, Lauren
Reid, and Jonathan Shipman.
?
Corresponding author.
E-mail addresses: [email protected] (B.K. Church), map608@
lehigh.edu (M. Peytcheva), [email protected] (W. Yu), [email protected]
(O.-A. Singtokul).
1
The names of the ?rst three authors are listed in alphabetical order.
Accounting, Organizations and Society 45 (2015) 40–51
Contents lists available at ScienceDirect
Accounting, Organizations and Society
j our nal homepage: www. el sevi er . com/ l ocat e/ aos
that role-taking experience enhances auditors’ understanding of
the manager’s viewpoint, which bene?ts auditors’ performance.
Audit ?rms, especially the Big Four, have spent increasing amounts
of resources to recruit former employees, commonly known as
boomerangs (Badal, 2006; Deloitte, 2011). Firms’ recruiting direc-
tors assert that auditors who return, after having spent time in
industry, bring back ‘‘stronger knowledge, a broader sense of expe-
rience, a broader skill set’’ to the ?rm (Hyland, 2006). Our experi-
ment sheds light on an advantage that accrues to audit ?rms by
hiring employees from industry.
We manipulate auditors’ prior experience in the manager’s role
(experience versus no experience), examining its effect on auditors’
behavior. We predict and ?nd that role-taking experience stimu-
lates perspective taking. Auditors with role-taking experience more
accurately estimate managers’ reported earnings, as compared to
auditors without such experience, and in turn make better report-
ing decisions, which promotes high-quality ?nancial reporting.
We design a second experiment to further investigate the effect
of perspective taking on auditors’ capacity to evaluate managers’
reported earnings (i.e., to assess whether earnings are materially
misstated). We focus on dispositional perspective taking, represent-
ing individuals’ natural ability to spontaneously take the viewpoint
of another. Our second experiment allows us to cleanly examine the
effect of perspective taking on auditor-participants’ task perfor-
mance, apart from the effect of situational factors that also might
in?uence perspective taking. We use an established personality
measure to appraise auditor-participants’ perspective-taking dispo-
sition (Davis, 1980, 1983). We ?nd that auditors with high
perspective-taking disposition are better able to judge managers’
reported earnings than auditors with lowperspective-taking dispo-
sition. This result suggests that perspective-taking disposition is an
important individual trait that underlies auditors’ performance.
Accordingly, our study complements other auditing-based research
that investigates individuals’ characteristics, including disposi-
tional tendencies (e.g., Bonner & Lewis, 1990; Hurtt, 2010; Majors,
Shefchik, & Vitalis, 2014; Rose, 2007; Sco?eld, Phillips, & Bailey,
2004).
Our study makes several contributions to the extant literature on
audit quality. First, we provide evidence that perspective taking
improves auditors’ ability to accurately assess managers’ reported
earnings. The takeaway is that effective perspective taking is bene?-
cial to audit quality. Second, we provide evidence that role-taking
experience stimulates perspective taking. Furthermore, some indi-
viduals quite naturally are better at perspective taking than others,
aside fromthe effects of role-taking experience. Those who are better
at perspective taking, in turn, may have an advantage in strategic
interactions as compared to others. Third, our ?ndings suggest that
auditors who possess better mental models of true earnings,
indeed, are able to make better decisions, which enhances
?nancial-reporting quality (Peecher et al., 2007). While this link often
is assumed in the literature, we are able to provide empirical evi-
dence of its validity. We suggest that auditors might bene?t by mak-
ing earnings estimates of reporting segments as a way to judge the
suf?ciency and accuracy of their mental models for a client’s overall
earnings process (see also Budescu, Peecher, & Solomon, 2012).
The results of our two experiments, taken together, have impor-
tant implications for audit practice. First, audit ?rms can bene?t
from hiring auditors with prior experience in the corporate world,
especially those involved in the ?nancial-reporting process. Audit
?rms’ efforts to recruit accountants from industry, including
boomerangs, appear to be prudent as industry experience may
aid perspective taking (i.e., anticipating managers’ behavior and
actions). Second, audit ?rms can bene?t from targeted training
programs that involve role taking. Such training programs poten-
tially represent an ef?cient means to boost audit quality. As
pointed out by Trotman, Wright, and Wright (2005), role-taking
training is a relatively low-cost technique, as it can be conducted
with other participants from the same ?rm. Third, audit ?rms
may want to consider dispositional perspective taking in staf?ng
assignments, for example, ensuring that some team members have
high perspective-taking disposition. The importance of including
such individuals on audit teams likely is magni?ed when the audi-
tor and client are involved in resolving signi?cant disagreements
over accounting matters. Lastly, audit ?rms are advised to consider
other ways to facilitate perspective taking (e.g., prompts in audit
programs) as a means to improve auditor performance and, ulti-
mately, audit quality.
The remainder of this paper is organized as follows. Section 2
reviews the related literature, which provides a basis to develop
our research hypotheses. Section 3 describes our ?rst experiment,
and Section 4 presents the experimental results, along with a dis-
cussion of the ?ndings. Section 5 describes our second experiment
and, subsequently, presents the results. Section 6 offers concluding
remarks.
Background, theory, and hypotheses
Background
The importance of perspective taking in cognitive processes is
examined extensively in Piaget’s (1932, 1950) early work.
Individuals take another’s perspective by simulating another’s
internal states, and the form of simulation can vary depending on
task demands (Niedenthal, Barsalou, Winkielman, Krauth-Gruber,
& Ric, 2005). The aim of perspective taking is to put oneself ‘‘in
another’s shoes’’ and to effectively take another’s vantage. Prior
studies document numerous cognitive bene?ts of successful per-
spective taking, which arise in various social interactions (e.g.,
Davis et al., 1996; Galinsky & Moskowitz, 2000; Galinsky &
Mussweiler, 2001; Galper, 1976; Johnson, 1967; Johnson &
Johnson, 1982; Regan & Totten, 1975; Sessa, 1996).
Altiero et al. (2014) investigate perspective taking in an audit-
ing context, noting that regulatory standards require auditors to
consider investors’ perspective when assessing materiality. The
authors conduct an experiment, with experienced auditors as par-
ticipants, and use a series of investor-minded tasks to stimulate
auditors to actively take an investor’s perspective. The ?ndings
indicate that prompting auditors to take the investor’s perspective,
by using investor-minded tasks, enables them to discriminate
between potential audit adjustments that are more or less likely
to be material for qualitative reasons. Otherwise, auditors fail to
make the distinction. The results hold for both specialist and
non-specialist auditors, but are more pronounced for specialists.
2
These ?ndings underscore the importance of successful perspective
taking as a means to reinforce audit quality.
As compared to Altiero et al. (2014), we examine perspective
taking in strategic interactions with managers. We maintain that
perspective taking improves auditors’ strategic reasoning – the
process whereby an agent reasons about the best strategy to adopt
in a multi-player scenario, taking into account the likely behavior
of the counterpart(s).
3
Without question, strategic reasoning is an
2
When prompted, specialist auditors draw upon rich, domain-speci?c knowledge,
permitting them to hone in on qualitative facets of materiality, which affects their
evaluation of potential audit adjustments (e.g., whether adjustments affect the trend
of key performance indicators).
3
In strategic interactions, zero-order reasoning occurs when agents consider their
own incentives, but not their counterparts’ incentives. Zero-order reasoning involves
‘‘no understanding of the desires, beliefs, or thoughts of others’’ (Hedden & Zhang,
2002) and can lead to myopic choices. In contrast, higher-order reasoning involves
strategic consideration of the incentives of one’s counterparts as well as consideration
of counterparts’ beliefs about, and anticipation of, one’s own incentives (Colman,
2003; Hedden & Zhang, 2002; Perner & Wimmer, 1985).
B.K. Church et al. / Accounting, Organizations and Society 45 (2015) 40–51 41
important part of auditing (Bloom?eld, 1995; Bowlin, 2011), and
prior ?ndings suggest that strategic thinking can improve auditors’
performance (e.g., Bowlin, 2011; Hoffman & Zimbelman, 2009;
Zimbelman & Waller, 1999).
We claim that successful perspective taking facilitates strategic
reasoning in that it allows auditors to better understand managers’
problem space, including their reporting choices and incentives.
Notwithstanding, successfully taking another’s perspective can be
dif?cult. People tend to overestimate the extent to which others
share their point of view, and they often over-rely on their own
experiences when interpreting or predicting the behavior of others
(e.g., Epley, Keysar, Boven, & Gilovich, 2004; Van Boven, Dunning, &
Loewenstein, 2000; Van Boven, Loewenstein, & Dunning, 2005).
We are interested in ways to enhance perspective taking, help-
ing auditors put themselves in the manager’s shoes. We examine
whether gaining experience in the counterpart’s role enables audi-
tors to make better judgments and decisions in strategic interac-
tions with managers. Next, we consider role-taking experience
and its effect on perspective taking and performance.
Role-taking experience, perspective taking, and performance
Research in cognitive and social psychology emphasizes the
importance of role taking in the process of social learning and pro-
poses that successful interactions require the ability to ‘‘take the
role of the other’’ (Biddle, 1986; Mead, 1934; Moreno, 1934,
1946). This line of research recognizes that the same individual
might reason and act quite differently in different roles and, like-
wise, that different individuals may behave similarly in similar
roles (Turner, 2006). Work grounded in the tradition of Piaget
(1932, 1950) highlights role taking as an antecedent of successful
perspective taking (e.g., Feffer, 1959).
Previous studies provide evidence that role-taking experience
improves perspective taking (e.g., Burns & Brainerd, 1979;
Chandler, 1973; Iannotti, 1978; Staub, 1971).
4
Chalmers and
Townsend (1990) ?nd that role-taking experience facilitates under-
standing others’ problems and incentives. Sessa (1996) shows that
role-taking experience in?uences individuals’ perceptions of con?ict,
primarily because it enables better insight into others’ perspective.
In auditing, Trotman et al. (2005) ?nd that role-taking experience
helps auditors develop successful strategies when negotiating with
clients.
Prior research indicates that role-taking experience helps indi-
viduals combat the tendency to overestimate the similarities
between self and others, referred to as social-projection bias.
5
Johnson (1967) ?nds that role-taking experience leads opponents
to gain a clearer understanding of others’ position. Van Boven et al.
(2000) show that social-projection bias, attributable to the endow-
ment effect (Kahneman, Knetsch, & Thaler, 1991), can be overcome
by equipping individuals with experience in another’s role. In con-
trast, the same study ?nds that providing information about the
endowment effect through lecturing alone does not reduce partici-
pants’ social-projection bias. These ?ndings highlight the importance
of role-taking experience to improve perspective taking and under-
score that role-taking experience cannot necessarily be substituted
by secondhand encounters or common knowledge of incentives.
The linkage between role-taking experience and enhanced per-
spective taking is, by no means, a tautology or foregone conclusion.
For example, Bowlin, Hales, and Kachelmeier (2009) provide
evidence that, in strategic interactions, role-taking experience does
not eliminate social-projection bias. Their ?ndings suggest that the
effect of such bias on participants’ performance can be for better or
worse. By comparison, we maintain that effective perspective tak-
ing improves individuals’ performance.
6
We assert that role-taking
experience allows auditors to form more accurate mental represen-
tations of the manager’s incentives, options, and alternatives.
Role-taking experience can enhance auditors’ perspective taking by
sensitizing them to situational factors that affect managers’ report-
ing choices, most notably managers’ incentives. In strategic interac-
tions, successful perspective taking serves to amplify the salience of
the counterpart’s incentives (e.g., Fiske, Taylor, Etcoff, & Laufer,
1979; Wyer, Srull, Gordon, & Hartwick, 1982). The increased salience
directs auditors’ attention to managers’ incentives.
7
In our ?rst experiment, we manipulate auditors’ prior experi-
ence in the manager’s role, with the aim of stimulating auditors’
perspective taking. We do so guided by prior research in cognitive
and social psychology, which documents a link between
role-taking experience and perspective taking (Burns & Brainerd,
1979; Chalmers & Townsend, 1990; Chandler, 1973; Iannotti,
1978; Staub, 1971). As explained by Epley and Caruso (2009), three
conditions are necessary for successful perspective taking:
(a) actively attempting to place oneself in another’s role or posi-
tion; (b) reasoning about the other’s perspective in a deliberate
and effortful way in order to avoid dwelling on one’s point of view
and to infer the perceptions of the other; and (c) using diagnostic
information about the other being evaluated. In strategic interac-
tions, role-taking experience facilitates the realization of all three
conditions. When the auditor has prior experience as a manager,
the auditor has been in the other’s role and has unique insight into
how the other might reason and act.
Auditors with role-taking experience likely are attuned to the
manager’s incentives, and this experience provides a basis for them
to understand and gauge the manager’s behavior. This point is
intuited by experimental economists, who often make participants
switch roles in signaling experiments, primarily to improve partic-
ipants’ understanding of different parties’ incentives (e.g., Brandts
& Holt, 1992, 1993; Cooper, Garvin, & Kagel, 1997a, 1997b; Kübler,
Müller, & Normann, 2008; Potters & van Winden, 1996). In fact,
Anderson and Camerer (2000, 701) explicitly discuss this issue.
‘‘There is a strong intuition among experimentalists that players do
learn faster when they switch roles [. . .]. This is easily testable, by
4
Role-taking experience has been used across various ?elds to aid professionals in
acquiring the perspectives of important others (see Sogunro, 2004 for a review).
5
Social-projection bias is known under other labels, including ‘‘egocentric bias’’
(Epley et al., 2004), ‘‘false consensus’’ (Krueger & Clement, 1994; West, 1996),
‘‘attributive projection’’ (Holmes, 1968), and ‘‘egocentric attribution’’ (Bazinger &
Kühberger, 2012; Heider, 1958).
6
Role-taking experience may not eliminate social-projection bias in Bowlin et al.
(2009) for a variety of reasons. The experimental setting in Bowlin et al. (2009)
focuses solely on players’ payoffs. Each player makes a dichotomous choice: the
manager chooses to report cautiously or aggressively and the auditor chooses to be
diligent or lax. The combination of players’ choices determines their respective
payoffs. By comparison, our setting permits more variation in players’ responses. The
manager chooses a reported amount (i.e., reported earnings). The auditor then
estimates earnings, using the manager’s report. The auditor also decides whether to
accept or reject the manager’s report. By incorporating earnings in our design (i.e., a
?ner measure of managers’ strategic intentions and auditors’ expectations), we may
be able to better capture the bene?ts of role-taking experience, that being
improvements in perspective taking.
7
Substantial literature in cognitive psychology provides evidence on the relation-
ship between attention and cognitive performance (e.g., James, 1890; Kahneman,
1973; Posner & Petersen, 1990). Prior ?ndings suggest that focusing attention on a
speci?c item serves as a selection mechanism for the importance of the item: it
increases the accessibility of knowledge in working memory that is relevant to the
item as compared to other items (Oberauer, 2002, 2009; Rerko & Oberauer, 2013).
Focused attention increases the likelihood that an item will be ef?ciently transferred
into working memory and leads to better encoding of stimuli (Dulas & Duarte, 2013;
Smith & Sewell, 2013) and, thus, superior cognitive performance (Daneman &
Carpenter, 1980). Research in education also provides evidence of an association
between focused attention and improvements in cognitive performance (Hidi, 1995;
Musso, Kyndt, Cascallar, & Dochy, 2012; Wilson, Will, Schoen?eld-McNeill, &
Montague, 2013).
42 B.K. Church et al. / Accounting, Organizations and Society 45 (2015) 40–51
comparing experiments with different degrees of role-switching,
but we know of no such experiments.’’
Because role-taking experience leads to a heightened under-
standing of situational incentives, it provides a basis for auditors
to carefully assess managers’ reporting choices, including the
chance of reporting biases. Auditors who have been in the man-
ager’s shoes have ?rsthand experience making reporting choices
and seeing how others react to such choices. Simply put,
role-taking experience stimulates auditors’ perspective taking,
which dampens their propensity to anchor on managers’ reports
and, further, allows them to better evaluate whether managers’
earnings reports are materially misstated (Galinsky &
Mussweiler, 2001). Therefore, auditors are expected to bene?t
from prior experience in the manager’s role. Our research hypothe-
ses are expressed as follows.
H1. Auditors with role-taking experience (as a manager) are better
able to assess whether the manager’s reported earnings are
materially misstated than auditors without role-taking experience.
H2. Auditors’ perspective taking mediates the relationship
between role-taking experience and their assessment of the man-
ager’s reported earnings.
Research method
Overview
We recruit 58 students from a public university to participate in
our ?rst experiment. The average age of the participants is
21 years, and 33 are male. The vast majority are undergraduates
(83%) and most are business majors (88%). Our use of student par-
ticipants is appropriate in that we examine economic behavior in
an abstract setting, where players are rewarded based on their per-
formance. Other experimental-economics studies use a similar
approach (e.g., King, 2002; Magilke, Mayhew, & Pike, 2009).
Participants are randomly assigned to one of two groups: a
role-taking group and a non-role-taking group. The experiment
consists of six rounds and lasts approximately 90 min. In rounds
1 and 2, participants in the role-taking group are managers, and
those in the non-role-taking group are auditors.
8
In rounds 3–6,
all participants are auditors. Therefore, participants in the
role-taking group switch roles in round 3, whereas those in the
non-role-taking group remain in the same role throughout the
experiment. Participants’ role by round is shown in Table 1.
Task and procedures
Upon arrival, participants report to one of two rooms, where
they stay for the entirety of the experiment. Instructions are dis-
tributed and read aloud by a researcher. Next, the researcher pro-
vides an example of how each experimental round proceeds and
answers participants’ questions. Then, six experimental rounds
are administered.
At the beginning of each round, participants learn their roles,
and managers and auditors are paired. Each round proceeds as
follows.
1. The manager is endowed with a public and private signal of
earnings. The combined signals indicate earnings, such that
the manager knows actual earnings.
2. The auditor observes the public signal of earnings, but not the
private one.
3. The manager submits an earnings report, which the auditor
reviews.
9
The manager’s report does not have to be truthful
(e.g., it can be in?ated).
4. The auditor makes his/her own estimate of earnings, condi-
tioned on the manager’s reported amount.
5. The auditor decides whether to accept or reject the manager’s
earnings report.
6. Participants are informed of their payoff.
Procedurally, actual earnings are determined by summing two
numbers, denoted Integer I and Integer II. Integer I is the public sig-
nal, and Integer II is the private one. Both numbers are randomly
generated each round by drawing from a uniform distribution that
ranges from 0 to 50. Thus, earnings can range from 0 to 100.
The manager is incentivized to report higher earnings values, as
long as the auditor accepts the reported amounts. The auditor, on
the other hand, is incentivized to accept amounts that are close
to the actual value (i.e., within a speci?ed materiality threshold)
and otherwise reject reported amounts. Each player’s speci?c pay-
offs/incentives are common knowledge. The procedures repeat for
six rounds, where auditor–manager dyads are re-paired each
round.
In rounds 1 and 2, managers in the role-taking group submit
earnings reports to auditors in the non-role-taking group. In
rounds 3–6, such pairings are not possible because all participants
are auditors. We make this design choice to ensure that auditors
only receive earnings reports from managers who have not been
auditors previously: that is, to ensure that our design is not con-
founded by managers’ speci?c prior experience as an auditor. To
generate managers’ reports for rounds 3–6, we use amounts
reported, along with the corresponding actual earnings, froma pre-
vious experimental session that includes manager–auditor pairs
without role-taking experience.
10
We make participants aware of
this fact. After the six rounds are ?nished, participants complete a
post-experiment questionnaire. The questionnaire includes an item
to measure participants’ perspective taking. Speci?cally, we include
the following statement, ‘‘I understand the perspective of the sen-
der’’ (manager), where participants respond on a seven-point
Likert scale, anchored by 1 = strongly disagree and 7 = strongly
Table 1
Participants’ role by round: Experiment one.
Round 1 Round 2 Round
3
a
Round
4
Round
5
Round
6
Role-taking
group
Manager Manager Auditor Auditor Auditor Auditor
Non-role-
taking
group
Auditor Auditor Auditor Auditor Auditor Auditor
a
As shown, over rounds 3–6, all participants are auditors. We use managers’
reports, along with corresponding actual earnings, randomly selected from a pre-
vious experimental session that includes managers and auditors over all rounds.
8
In the experimental materials, we refer to the manager as the ‘‘sender’’ and the
auditor as the ‘‘receiver.’’ We use neutral labels to denote players’ roles in order to
guard against extraneous in?uences that are associated with the assigned role. For
expositional convenience, we use the manager/auditor terminology throughout the
paper.
9
In the experimental materials, the manager’s earnings report is referred to as a
report of commodity value.
10
In the previous session, the basic procedures and experimental parameters are
identical to those in experiment one. Prior to conducting experiment one, a research
assistant, with no knowledge of the experiment, randomly selected four managers’
reports, along with the corresponding actual earnings amounts, from the previous
session. As a consequence, we control for managers’ reports and clients’ actual
earnings over rounds 3–6.
B.K. Church et al. / Accounting, Organizations and Society 45 (2015) 40–51 43
agree. Participants’ responses represent a measure of their perceived
ability to take the counterpart’s perspective.
11
Managers’ payoff
The manager’s payoff per round is contingent on the auditor’s
decision to accept or reject the reported amount. If the auditor
accepts the manager’s earnings report, the manager’s payoff is
the reported amount. Otherwise, the manager’s payoff is 65% of
the actual earnings. The payoff choices derive from practical con-
siderations such that (1) higher reported amounts are preferable
(e.g., increasing the likelihood of meeting targets, increasing the
manager’s bonuses, and bene?tting the manager’s reputation)
and (2) auditor rejection is costly (e.g., delays in reporting, strained
relations with the auditor, and additional audit costs that, at some
point, are passed on to the client).
Auditors’ payoff
The auditor’s payoff is contingent on the auditor’s estimate of
earnings, the auditor’s accept/reject decision, and the manager’s
earnings report. The auditor relies on the manager’s report to
determine an estimated earnings amount. The auditor then consid-
ers whether to accept or reject the manager’s report. Relatedly,
Peecher et al. (2007) suggest that the accuracy of auditors’ earnings
estimates in?uences their reporting decisions, for the betterment
of ?nancial-reporting quality. Budescu et al. (2012) assert that
auditors’ expectations impact their assessment of material mis-
statement: auditors are more likely to conclude that reported
amounts are fairly stated when auditors’ evidence-based expecta-
tions are closer to managers’ reported amounts. For our purposes,
auditors’ earnings estimates are compared to managers’ reported
amounts and deviations are judged in light of the speci?ed materi-
ality threshold. Thus, auditors’ earnings estimates provide a basis
to accept or reject reported amounts, thereby impacting
?nancial-reporting quality.
The auditor’s payoff per round, depicted in Fig. 1, indicates that
the auditor’s payoff includes a ?xed component and a bonus com-
ponent. We discuss the two components below.
If the auditor accepts the manager’s report, the auditor’s ?xed
payoff is 50 when the reported amount is within ±10 of the actual
amount and 0 otherwise. Under these circumstances, the auditor
approves the client’s ?nancial statements without adjustment.
The ±10 represents materiality bounds, such that the manager’s
report is materially correct as long as it is within these bounds.
Accordingly, the auditor receives a high payoff for approving
reports that are materially correct and a low payoff for approving
reports that are materially misstated.
Next, we consider the auditor’s payoff when s/he rejects the
manager’s report. Now, the auditor’s ?xed payoff is 35 or 5, contin-
gent on the auditor’s estimate of earnings. Under these circum-
stances, the auditor’s estimate becomes the amount that is
re?ected in the client’s ?nancial statements. If the auditor’s esti-
mate is within the materiality bounds, and it is closer to the actual
earnings than the manager’s reported amount, the auditor’s ?xed
payoff is 35. In this case, the client’s ?nancial statements are mate-
rially correct. The payoff choice re?ects the practical reality that if
the auditor initially rejects the manager’s report, additional audit
work is required and client relations are strained, which reduces
the auditor’s ?xed payoff (as compared to a ?xed payoff of 50).
By comparison, if the auditor’s estimate is outside of the
materiality bounds or it is farther away from the actual earnings
than the manager’s reported amount, the auditor’s ?xed payoff is
5. In this case, the auditor’s estimate either is materially misstated
or inferior to the manager’s earnings report. Whichever the case,
the practical implications are that additional audit work is
required, client relations are strained, and outputs are inferior.
12
In addition, the auditor can earn a bonus each round based on
the earnings estimate. The bonus is computed as follows.
We provide a bonus to ensure that auditors are incentivized to be
accurate in their earnings estimate, irrespective of the accept/re-
ject decision.
In determining the experimental parameters (refer to Fig. 1), we
assume that the auditor’s objective is to approve materially correct
reports and otherwise reject reports. The auditor’s decision, in turn,
provides a basis for differences in the ?xed payoff component.
When the auditor rejects the manager’s report, we assume that
the auditor must determine an acceptable earnings amount to
report: that is, an amount within the materiality bounds. The audi-
tor contributes to high-quality ?nancial reporting as long as s/he
(1) approves materially correct reports or (2) rejects reports and
determines a value to report that is materially correct and superior
to that of managers’ reported amounts. For illustrative purposes,
Fig. 2 depicts when the auditor contributes to high-quality ?nancial
reporting: that is, the decision nodes where the auditor does and
does not bene?t ?nancial-reporting quality.
Results
We are interested in the effect of role-taking experience on
auditor-participants’ ability to determine whether managers’
reported earnings are materially misstated and, further, whether
perspective taking mediates the relationship. First, we consider
the accuracy of auditors’ estimates of the manager’s earnings. We
argue that auditors who make more accurate estimates are better
equipped to judge whether reported earnings are free of material
misstatement and, thus, to promote high-quality ?nancial report-
ing. Next, we scrutinize whether role-taking experience facilitates
auditors’ perspective taking, which allows us to get at the cognitive
process underlying auditors’ judgments. Lastly, we perform addi-
tional analyses and discuss our ?ndings.
Tests of H1
Our ?rst hypothesis predicts that auditors with role-taking
experience are better able to assess whether managers’ reported
Bonus ¼ ð100 À jAuditor’s Estimate of Earnings ÀClient’s Actual EarningsjÞ=10:
11
This measure emphasizes the cognitive aspect of perspective taking (Anderson &
Camerer, 2000; Epley & Caruso, 2009).
12
Economic theory predicts that the manager reports a value such that the auditor
is indifferent to accepting or rejecting the amount: the auditor accepts any amount
equal to or below the value and rejects any amount above it. Based on our
experimental parameters, the economic predictions are that (1) the manager reports
Integer I + 43 and (2) the auditor accepts the reported amount. Details are available
from the authors upon request.
44 B.K. Church et al. / Accounting, Organizations and Society 45 (2015) 40–51
earnings are materially misstated than auditors without such
experience. We examine auditor-participants’ estimates of
earnings over rounds 3–6: participants in both experimental
groups provide estimates of earnings in rounds 3–6. We compute
the average absolute estimation error per round for each partici-
pant as follows.
By computing an average error, we generate one independent
observation per participant, which allows us to avoid potential
problems that arise if participants’ estimation errors are correlated
across rounds.
Table 2 presents descriptive statistics for auditors’ average
absolute estimation error, partitioning the data by the two exper-
imental groups. The mean (median) estimation error of the
role-taking group is less than that of the non-role-taking group:
7.96 (7.00) versus 10.43 (9.75), respectively. A parametric t-test
and a nonparametric Mann–Whitney test show that the difference
Auditor’s Acceptance
of the Manager’s
Reported Value
|D| ? 10
|E| ? 10 and |E| ? |D|
0
points + Bonus
35
points + Bonus
5
points + Bonus
Yes
No
No
No
50
points + Bonus Yes
Yes
Fig. 1. Auditor’s payoff per round. Note: D = Manager’s Earnings Report À Actual Earnings, E = Auditor’s Estimate of Earnings À Actual Earnings, and Bonus = (100 À |Auditor’s
Estimate of Earnings À Actual Earnings|)/10.
Basis of
Financial
Reporting:
Is the Auditor
Contributing to
High-Quality
Financial
Reporting?
YES
Financial
Reporting is
based on the
Manager’s
Report
Is the
Manager’s
Report within
the materiality
bounds?
NO
Does the
Auditor Accept
the Manager’s
Report?
YES
Is the
Auditor’s
Estimate better
than the
Manager’s
Report? NO
Financial
Reporting is
based on the
Auditor’s
Estimate
Is the Auditor’s
Estimate within
the materiality
bounds?
NO
YES
YES
NO
YES
NO
YES
NO
NO
Fig. 2. The auditor’s role in high-quality ?nancial reporting.
Estimation Error ¼
X
jAuditor’s Estimate of Earnings ÀClient’s Actual Earningsj=4:
B.K. Church et al. / Accounting, Organizations and Society 45 (2015) 40–51 45
between the two groups is statistically signi?cant at conventional
levels: t(56) = À2.10, p = 0.02, one-tailed, and Z = À2.52, p = 0.006,
one-tailed, respectively. Inferences are unaffected if we perform a
repeated measures analysis of variance, using participants’ abso-
lute estimation error per round as the dependent variable. Our
?ndings are consistent with H1 and suggest that role-taking expe-
rience enhances auditors’ performance.
We further probe the accuracy of auditors’ earnings estimates.
Fig. 3 presents the frequency of participants’ estimation errors over
rounds 3–6, partitioned into categories based on the magnitude of
the error. The frequencies are presented separately for each exper-
imental group. Fisher’s exact test indicates that the proportion of
absolute estimation errors that falls within the materiality bounds
(i.e., |E| 6 10) is signi?cantly greater for the role-taking group than
the non-role-taking group at p = 0.003, one-tailed. These ?ndings
provide further support for H1.
High-quality ?nancial reporting and links to the accuracy of earnings
estimates
Next we consider the effect of role-taking experience on audi-
tors’ decisions. Panel A of Table 3 presents descriptive data, sum-
marizing the percentage of time that auditors’ decisions
contribute to ?nancial-reporting quality (refer to Fig. 2), parti-
tioned by role-taking group. We perform a parametric t-test and
a nonparametric Mann–Whitney test, collapsing participants’ deci-
sions across the four rounds. We ?nd that auditors in the
role-taking group are more likely to make decisions that contribute
to ?nancial-reporting quality than auditors in the non-role-taking
group: t(56) = 2.58, p = 0.007, one-tailed, and Z = 2.44, p = 0.007,
one-tailed, respectively. We also examine whether auditors’
accept/reject decisions are correct, irrespective of their
earnings estimates. That is, we examine the percentage of time
that auditors accept reported amounts that are within ±10 of the
actual amount and otherwise reject reported amounts.
Panel B of Table 3 presents descriptive data. Again, we perform
parametric and nonparametric tests and ?nd that auditors in the
role-taking group are more likely to make correct decisions than
auditors in the non-role-taking group: t(56) = 2.17, p = 0.017,
one-tailed, and Z = 1.85, p = 0.03, one-tailed, respectively. Our ?nd-
ings suggest that role-taking experience fosters ?nancial-reporting
quality.
We conduct further analyses to assess the link between the
accuracy of auditors’ earnings estimates and ?nancial-reporting
quality. As suggested earlier, auditors who formulate more accu-
rate estimates of actual earnings make sounder decisions, bene?t-
ting ?nancial-reporting quality (see also Peecher et al., 2007). To
formally test this relationship, we regress the number of times that
auditors’ decisions contribute to ?nancial-reporting quality (over
rounds 3–6) on their average absolute estimation error.
We ?nd that the coef?cient of the average estimation error is neg-
ative and statistically signi?cant (b = À0.18, t(56) = À11.25,
p < 0.001). We repeat the analysis substituting the number of times
that auditors make correct accept/reject decisions as the depen-
dent variable and results are similar (b = À0.13, t(56) = À6.77,
p < 0.001).
13
Our ?ndings suggest that the accuracy of
auditors’ earnings estimates are associated with better decisions.
Auditors who have smaller estimation errors, indeed, promote
?nancial-reporting quality.
Tests of H2
Our second hypothesis predicts that perspective taking medi-
ates the relationship between role-taking experience and auditors’
assessment of managers’ reporting choices. As part of the
post-experiment questionnaire, participants respond to a state-
ment that elicits perceived perspective taking, where higher values
suggest better self-assessed perspective-taking ability. We ?nd
that the average response of the role-taking group (M = 5.59,
SD = 1.50) is higher than that of the non-role-taking group
(M = 3.66, SD = 1.86), and the difference is statistically signi?cant
using parametric (t(56) = 4.36, p < 0.001, one-tailed) and nonpara-
metric (Z = 3.89, p < 0.001, one-tailed) tests. We investigate
whether participants’ perspective taking mediates the relationship
between auditors’ role-taking experience and the accuracy of their
earnings estimates.
Fig. 4 presents the results of a path analysis. We observe that
when auditors’ perspective taking is included in the model,
role-taking experience loses its statistical signi?cance. Further,
the path from role-taking experience to auditors’ perspective tak-
ing is statistically signi?cant (p < 0.001), as is the path from per-
spective taking to auditors’ estimation error (p = 0.019). We test
for mediation, as suggested by Preacher and Hayes (2008). Using
the bootstrapping macro, we document a signi?cant mediation
Table 2
The effect of role-taking experience on auditors’ accuracy in estimating earnings:
Experiment one.
a
Statistic Role-taking group Non-role-taking group Total
Mean 7.96 10.43 9.19
Median 7.00 9.75 8.63
Std. Deviation 4.47 4.48 4.61
Min 2.25 1.25 1.25
Max 22.75 20.25 22.75
Mean test: t(56) = À2.10, p = 0.02, one-tailed
Median test: Z = À2.54, p = 0.006, one-tailed
a
In experiment one, auditors in the role-taking group have prior experience in
the manager’s role. Accuracy in estimating earnings is measured by the auditor’s
average estimation error, calculated as
P
|Auditor’s Estimate of Earnings À Actual
Earnings|/4.
Fig. 3. Frequency of auditors’ absolute estimation error: Experiment one. Note: The
?gure shows the frequency of participants’ absolute estimation error, computed as
P
|Auditor’s Estimate of Earnings À Actual Earnings|/4, broken down into categories
based on the magnitude of the error I. The ?gure includes all observations from
rounds 3–6. Using Fisher’s exact test, the proportion of absolute estimation errors
that falls within the materiality bounds (i.e., |E| 6 10) is signi?cantly greater for the
role-taking group than the non-role-taking group at p = 0.003, one-tailed.
13
We re-perform the analyses using logistic regression, controlling for round
effects. The dependent variable is auditors’ decision per round: 1 = contributing to
?nancial-reporting quality (or making a correct decision) and 0 = otherwise. The
independent variable is auditors’ absolute estimation error per round. Inferences are
unaffected: that is, smaller estimation errors are associated with better decisions.
46 B.K. Church et al. / Accounting, Organizations and Society 45 (2015) 40–51
effect (p = 0.031, one-tailed), and we ?nd that the 95% bootstrap
con?dence interval for indirect effects excludes zero. Our results
provide support for H2.
To gain further insight, we consider the effect of perspective
taking on auditors’ performance in each experimental group sepa-
rately. We do so because others (Davis, 1980, 1983; Davis et al.,
1996; Davis, Hull, Young, & Warren, 1987) document individual
differences in perspective-taking ability. The analyses presented
above are indicative of differences between the two experimental
groups (role-taking versus non-role-taking). We probe whether
differences also arise within groups. According to our theoretical
development, effective perspective taking improves individuals’
judgments (e.g., Altiero et al., 2014; Davis et al., 1996; Galinsky &
Moskowitz, 2000; Galinsky & Mussweiler, 2001; Galper, 1976;
Johnson, 1967; Johnson & Johnson, 1982; Regan & Totten, 1975;
Sessa, 1996). We examine whether this relationship holds, irre-
spective of participants’ prior experience.
We regress auditors’ average absolute estimation error on their
perceived perspective taking, performing the analysis separately
for each role-taking group. We ?nd that, in both groups, the coef-
?cient of auditors’ perspective taking is negative and statistically
signi?cant: b = À0.83, t(27) = À1.89, p = 0.035, one-tailed, for the
non-role-taking group, and b = À0.95, t(27) = À1.75, p = 0.046,
one-tailed, for the role-taking group.
14
Hence, participants who per-
ceive that they are better at perspective taking, indeed, are more
accurate in estimating earnings. Our ?ndings suggest that individual
differences in perspective taking in?uence auditor behavior, beyond
their role-taking experience.
Additional analyses
Prior ?ndings suggest that individuals are susceptible to
social-projection bias (e.g., Bowlin et al., 2009; Epley et al., 2004;
Rizzolatti & Craighero, 2004; Van Boven et al., 2000, 2005). To
determine whether social-projection bias is a viable explanation
for our ?ndings, we analyze the association between prior experi-
ence and auditors’ judgments. Social-projection bias predicts that,
in the role-taking group, prior experience (over rounds 1–2)
explains auditors’ estimates of earnings and their decisions to
accept/reject reported earnings (over rounds 3–6). More speci?-
cally, social-projection bias suggests that the degree to which par-
ticipants in?ate reported earnings as a manager affects (1) the
extent to which they discount reported earnings as an auditor
and (2) their propensity to reject reported earnings.
For participants in the role-taking group, we regress auditors’
average downward adjustment of earnings (over rounds 3–6) on
their average reporting bias (over rounds 1–2). Auditors’ down-
ward adjustment is de?ned as the difference between managers’
reported earnings and auditors’ estimated earnings. Reporting bias,
on the other hand, is de?ned as the difference between managers’
reported earnings and actual earnings. We ?nd that the coef?cient
of average reporting bias is not signi?cant (b = À0.06,
t(27) = À0.78, p = 0.44). We repeat the analysis using the number
of times that auditors reject managers’ reported earnings as the
dependent variable and again, the coef?cient of average reporting
bias is not signi?cant (b = 0.01, t(27) = 0.51, p = 0.61). Finally, we
re-perform the analyses substituting a dummy variable for average
reporting bias, de?ned as 1 if the average bias (over rounds 1–2) is
greater than 10 (the materiality threshold) and 0 otherwise, and
inferences are unaffected. Hence, we conclude that speci?c prior
experience does not account for auditors’ behavior, and we rule
out social-projection bias as an alternative explanation for our
results.
Discussion
Our results suggest that role-taking experience facilitates audi-
tors’ assessment of managers’ reporting choices (i.e., determining
whether reported earnings are materially misstated). Auditors
with role-taking experience (as a manager) make more accurate
estimates of earnings than auditors without such experience. In
terms of the underlying cognitive mechanism, we ?nd that
role-taking experience stimulates perspective taking, which
enables auditors to more effectively put themselves in the man-
ager’s shoes. In turn, they are better able to judge managers’
reported earnings. We rule out social-projection bias as an alterna-
tive explanation for our ?ndings, and we offer evidence that links
the accuracy of auditors’ earnings estimates with their decision
to accept/reject manager’s reports, thereby promoting
?nancial-reporting quality. Along these lines, Peecher et al.
(2007) provide a framework that characterizes the audit process
as evidence-driven, belief-based, risk assessment. As part of the
audit process, auditors formulate beliefs about reported amounts
and disclosures, which serve as a basis to evaluate the reasonable-
ness of managers’ reporting choices. Auditors who formulate
richer, more accurate beliefs are expected to provide higher quality
services, bene?tting ?nancial-reporting quality (see also Budescu
et al., 2012). Our ?ndings are consistent with this depiction of
the audit process.
Our results also provide evidence of individual differences in
perspective taking. We ?nd that, even in the non-role-taking
group, auditors’ perspective taking is associated with the accuracy
of their earnings estimates. This result suggests that
perspective-taking disposition can signi?cantly impact individuals’
judgments and decision making in strategic interactions.
Table 3
Auditors’ decisions and role-taking experience: Experiment one.
Round Role-taking group Non-role-taking group
Panel A: Percentage of auditors’ decisions that contribute to ?nancial-reporting
quality
a
3 0.59 0.45
4 0.90 0.69
5 0.76 0.55
6 0.41 0.31
Total M = 0.66 M = 0.50
SD = 0.21 SD = 0.27
Panel B: Percentage of auditors’ decisions that correctly accept/reject managers’
report
b
3 0.66 0.52
4 0.72 0.55
5 0.79 0.66
6 0.45 0.41
Total M = 0.66 M = 0.53
SD = 0.18 SD = 0.24
a
The cell entries in Panel A indicate the percentage of time that auditors’ decision
contribute to ?nancial-reporting quality. This result occurs when (1) the auditor
accepts reported earnings that are within ±10 of actual earnings or (2) the auditor
rejects the manager’s report, the auditor’s earnings estimate is acceptable (within
±10 of actual earnings, and the auditor’s estimate is superior to that of the manager
(closer to actual earnings than the manager’s reported amount). The speci?cs that
underlie high-quality ?nancial reporting are depicted in Fig. 2. The total at the
bottom of the Panel indicates the mean and standard deviation of the percentage of
time that auditor-participants’ decisions contribute to ?nancial-reporting quality.
b
The cell entries in Panel B indicate the percentage of time auditors make the
correct decision to accept or reject reported earnings, irrespective of auditors’
estimates of earnings. The auditor’s decision is correct as long as the auditor accepts
reported earnings that are within ±10 of actual earnings and rejects other amounts.
The total at the bottom of the Panel indicates the mean and standard deviation of
the percentage of time that auditor-participants’ decisions are correct.
14
We also put participants into high and low perspective-taking groups, using a
median split, and repeat the regression analysis for each role-taking group. Inferences
are unaffected.
B.K. Church et al. / Accounting, Organizations and Society 45 (2015) 40–51 47
We note that other research has focused on perspective taking
as a trait or personality dimension (Davis, 1980, 1983; Davis
et al. 1996). More speci?cally, Davis (1980, 1983) has developed
a measure to gauge individuals’ tendency to spontaneously take
the psychological point of view of others, focusing on cognitive
perspective taking.
15
The measure is established and designed to
capture individuals’ capacity for non-egocentric thought, which
allows them to anticipate the behavior and reactions of others. By
comparison, the measure we use in our experiment is
self-assessed. One advantage of our measure is that it is context
speci?c: that is, the measure is elicited based on participants’ consid-
eration of our experimental setting. A potential drawback, however,
is that our measure has not been validated and, as such, it may cap-
ture other dimensions of personality beyond the ability to put one-
self in another’s shoes. The fundamental concern is that our
?ndings might be driven by some other facet of personality, rather
than perspective taking. Indeed, prior research documents that
perspective-taking ability is linked to self-con?dence, self-esteem,
and perceived self-ef?cacy (Corcoran & Mallinckrodt, 2000; Davis,
1983). Further, previous studies document that these personality
dimensions are positively associated with task performance
(Bandura, 1993; Berry, 1987; Berry & West, 1993; Stankov, 2000;
Stankov & Lee, 2008; Wood & Bandura, 1989).
In an effort to appraise the validity of our perspective-taking
measure, we recruit 60 Amazon Mechanical Turk participants
and elicit their responses to the Davis measure and our measure
from experiment one. Participants’ responses indicate that the
two measures are signi?cantly correlated, q = 0.77, p < 0.001.
Notwithstanding, we acknowledge that our measure is a poten-
tially noisy proxy for perspective-taking ability.
To shed further light on the effect of perspective taking on
auditor-participants’ task performance, we design a second exper-
iment using the Davis measure, a validated measure of individual
perspective-taking disposition. Our focus is on individual differ-
ences, absent other situational factors that might stimulate per-
spective taking. The aim is to cleanly examine the effects of
perspective taking on task performance: that is, to determine
whether participants’ perspective-taking disposition affects their
judgments in a predictable way. As discussed previously, effective
perspective taking helps individuals to better judge the behavior of
others in social interactions (Davis et al., 1996; Galinsky &
Moskowitz, 2000; Galinsky & Mussweiler, 2001; Galper, 1976;
Johnson, 1967; Johnson & Johnson, 1982; Regan & Totten, 1975;
Sessa, 1996). Accordingly, we expect participants with high
perspective-taking disposition to have better insight into others’
behavior than those with low perspective-taking disposition. Our
second experiment is designed to provide insight into whether
auditor-participants’ disposition to take another’s perspective
impacts their ability to judge managers’ reporting behavior.
Experiment two
Task and procedures
We recruit 39 students from a public university to participate in
our second experiment. The average age of the participants is
21 years, and 19 are male. Almost all participants are undergradu-
ates (92%), and the majority are business or economics majors
(59%). The experimental task is the same as that used in our ?rst
experiment, except that all participants are assigned to the
non-role-taking condition, meaning that everyone is an auditor
for the entirety of the experiment. As a consequence, we introduce
two practice rounds (replacing rounds 1–2 from our earlier exper-
iment) in which the instructions walk participants through two
examples. The examples explain players’ potential actions and
how such actions map into players’ payoffs. After the two practice
rounds, four experimental rounds are conducted. We use the same
managers’ earnings reports (rounds 3–6) as in our ?rst experiment.
After ?nishing the experimental task, participants complete a
post-experiment questionnaire, which includes the Davis (1980)
scale. The scale measures participants’ spontaneous
perspective-taking disposition and is comprised of seven items
that elicit how well various statements describe respondents.
Each statement involves respondents’ cognitive ability to
empathize with others.
16
Participants’ mean score on the Davis
scale is 4.42 (SD = 0.85). Following Davis et al. (1987), we split them
at the median and put them into high and low perspective-taking
groups. The average response is 5.00 (SD = 0.52) for participants in
the high perspective-taking group and 3.69 (SD = 0.61) for those in
the low perspective-taking group.
Role-Taking
Experience
Perspective Taking
Accuracy in
Estimating Earnings
?
1
= 2.47
p = 0.016
?
1
= 1.93
p < 0.001
?
2
= 0.88
p = 0.019
?
1
= 0.78
p = 0.623
Fig. 4. Path analysis results with perspective taking as a mediator: Experiment one. Note: The ?gure shows the path analysis results linking role-taking experience, perceived
perspective taking, and auditors’ accuracy in estimating earnings. The coef?cient for the direct path from role-taking experience to auditor’s accuracy in estimating earnings is
represented by c
1
when the mediator is not included in the path model and d
1
when the mediator is included in the model. Using Preacher and Hayes’ (2008) macro, we
document a signi?cant mediation effect at p = 0.031, one-tailed and we also ?nd that the 95% bootstrap con?dence interval for indirect effects excludes zero.
15
Davis (1980, 1983) develops the Interpersonal Reactivity Index, designed to
measure dispositional empathy. The instrument includes four separate subscales that
delve into separate facets of empathy, including perspective taking, empathic
concern, personal distress, and fantasy. We are interested in the perspective-taking
subscale that deals with cognition and understanding others’ beliefs, behavior, and
actions.
16
The scale has seven items: (1) I sometimes ?nd it dif?cult to see things from the
‘‘other person’s’’ point of view (reverse-coded); (2) I try to look at everybody’s side of
a disagreement before I make a decision; (3) I sometimes try to understand my
friends better by imagining how things look from their perspective; (4) If I’m sure I’m
right about something, I don’t waste much time listening to other people’s arguments
(reverse-coded); (5) I believe that there are two sides to every question and try to look
at them both; (6) When I’m upset at someone, I usually try to ‘‘put myself in his
shoes’’ for a while; and (7) Before criticizing somebody, I try to imagine how I would
feel if I were in their place. For each item, participants respond on a seven-point Likert
scale, anchored by 1 = does not describe me well to 7 = describes me well.
Participants’ average response across the seven items is a measure of their
perspective-taking disposition. Based on participants’ responses, we obtain a
Cronbach’s alpha of 0.88.
48 B.K. Church et al. / Accounting, Organizations and Society 45 (2015) 40–51
Results
We examine whether perspective-taking disposition is associ-
ated with auditors’ evaluation of managers’ earnings reports. We
compute auditors’ average absolute estimation error over the four
rounds. Descriptive statistics are presented in Table 4, partitioned
by perspective-taking group. We ?nd that the mean (median) esti-
mation error of the high perspective-taking group is less than that
of the low perspective-taking group: 7.11 (7.00) versus 9.49 (9.00),
respectively. A parametric t-test and a nonparametric Mann–
Whitney test indicate that the difference between the two groups
is statistically signi?cant at conventional levels: t(37) = À1.69,
p = 0.05, one-tailed and Z = À1.67, p = 0.049, one-tailed, respec-
tively. Our ?nding suggests that participants’ natural
perspective-taking disposition improves task performance.
To further judge the accuracy of auditors’ earnings estimates,
we examine the frequency of participants’ estimation errors over
the four rounds, partitioned into categories based on whether the
magnitude of the error is within ±10 of actual earnings. Fisher’s
exact test indicates that the proportion of absolute estimation
errors that falls within the materiality bounds (i.e., |E| 6 10) is sig-
ni?cantly greater for the high perspective-taking group than the
low perspective-taking group at p = 0.043, one-tailed. This ?nding
provides additional evidence of the effect of perspective-taking
disposition on participants’ performance.
Next, we investigate the effect of perspective-taking disposition
on auditors’ decisions. Panel A of Table 5 presents descriptive data,
summarizing the percentage of time that auditors’ decisions con-
tribute to ?nancial-reporting quality (refer to Fig. 2), partitioned
by perspective-taking group. Examination of the data indicates
that the percentage is larger for the high perspective-taking group
in three of four rounds. However, collapsing the data across the
four rounds, parametric and nonparametric tests indicate that
the difference between the two groups is not statistically signi?-
cant at conventional levels: t(37) = 0.85, p = 0.20, one-tailed, and
Z = 0.73, p = 0.23, one-tailed, respectively. Panel B of Tables 5 pre-
sents descriptive data, summarizing the percentage of time that
auditors’ make the correct accept/reject decision, irrespective of
their earnings estimates. Again, collapsing the data across the four
rounds, we ?nd that difference between the two groups is not sta-
tistically signi?cant: t(37) = 0.03, p = 0.49, one-tailed, and Z = 0.00,
p = 0.50, one-tailed. We suggest that perspective-taking disposition
may have a more subtle effect on auditors’ decisions as compared
to other variables (e.g., role-taking experience) that are designed to
stimulate perspective taking. Our conjecture is consistent with
?ndings in psychology that behavior is in?uenced by both contex-
tual factors and personality, but contextual factors are likely a
stronger predictor of behavior in a speci?c situation, whereas
personality is a more consistent predictor of behavior over time
(Fleeson, 2001; Fleeson & Noftle, 2008; Ross & Nisbett, 1991).
Subsequently, we scrutinize whether participants with more
accurate estimates of earnings promote high-quality ?nancial
reporting. As before, we regress the number of times that auditors’
decisions contribute to ?nancial-reporting quality on their average
absolute estimation error. We ?nd that the coef?cient of the aver-
age estimation error is negative and statistically signi?cant
(b = À0.18, t(37) = À5.17, p < 0.001), which suggests that estima-
tion accuracy is associated with superior decision making. We
repeat the analysis using the number of times that auditors make
the correct accept/reject decision and inferences are unaffected
(b = À0.11, t(37) = À3.41, p = 0.002).
Finally, we perform a path analysis to assess the links between
perspective-taking group, estimation accuracy, and promoting
high-quality ?nancial reporting. We document a signi?cant path
from perspective taking to estimation accuracy (path coef?cient
of À2.37, p = 0.042) and from estimation accuracy to promoting
high-quality ?nancial reporting (path coef?cient of À0.17,
p < 0.001). Results are similar examining the links between per-
spective taking, estimation accuracy, and making correct accep-
t/reject decisions. All in all, the results suggest that dispositional
perspective taking impacts auditor-participants’ assessment of
managers’ reporting choices, aside from the effects of other factors
that might be introduced to stimulate perspective taking.
Conclusion
We report the results of two experiments designed to examine
the effect of perspective taking on auditors’ ability to determine
whether managers’ reported earnings are materially misstated
and, in turn, to contribute to high-quality ?nancial reporting. In
our ?rst experiment, we manipulate auditors’ prior experience in
the manager’s role (experience versus no experience). We predict
Table 4
The effect of perspective-taking disposition on auditors’ accuracy in estimating
earnings: Experiment two.
a
Statistic Perspective-taking group Total
High Low
Mean 7.11 9.49 8.15
Median 7.00 9.00 7.50
Std. Deviation 3.79 5.01 4.46
Min 1.75 2.50 1.75
Max 15.75 21.25 21.25
Mean test: t(37) = À1.69, p = 0.05, one-tailed
Median test: Z = À1.67, p = 0.049, one-tailed
a
In experiment two, participants respond to the Davis scale, which allows us to
measure their perspective-taking disposition. We use a median split to partition
participants into a high and low perspective-taking group. Accuracy in estimating
earnings is measured by the auditor’s average estimation error, calculated as
P
|Auditor’s Estimate of Earnings À Actual Earnings|/4.
Table 5
Auditors’ decisions and perspective-taking disposition: Experiment two.
Round Perspective-taking group
High Low
Panel A: percentage of auditors’ decisions that contribute to ?nancial-reporting
quality
a
3 0.45 0.41
4 0.86 0.53
5 0.77 0.65
6 0.41 0.59
Total M = 0.63 M = 0.54
SD = 0.29 SD = 0.31
Panel B: Percentage of auditors’ decisions that correctly accept/reject managers’
report
b
3 0.45 0.41
4 0.64 0.47
5 0.86 0.82
6 0.41 0.65
Total M = 0.59 M = 0.59
SD = 0.25 SD = 0.25.
a
The cell entries in Panel A indicate the percentage of time that auditors’ decision
contribute to ?nancial-reporting quality. This result occurs when (1) the auditor
accepts reported earnings that are within ±10 of actual earnings or (2) the auditor
rejects the manager’s report, the auditor’s earnings estimate is acceptable (within
±10 of actual earnings, and the auditor’s estimate is superior to that of the manager
(closer to actual earnings than the manager’s reported amount). The speci?cs that
underlie high-quality ?nancial reporting are depicted in Fig. 2. The total at the
bottom of the Panel indicates the mean and standard deviation of the percentage of
time that auditor-participants’ decisions contribute to ?nancial-reporting quality.
b
The cell entries in Panel B indicate the percentage of time auditors make the
correct decision to accept or reject reported earnings, irrespective of auditors’
estimates of earnings. The auditor’s decision is correct as long as the auditor accepts
reported earnings that are within ±10 of actual earnings and rejects other amounts.
The total at the bottom of the Panel indicates the mean and standard deviation of
the percentage of time that auditor-participants’ decisions are correct.
B.K. Church et al. / Accounting, Organizations and Society 45 (2015) 40–51 49
and ?nd that role-taking experience stimulates perspective taking,
which allows auditors to more accurately estimate managers’
reported earnings, ultimately bene?tting ?nancial-reporting qual-
ity. In our second experiment, we appraise auditor-participants’
ability to spontaneously take another’s psychological viewpoint,
beyond the effects of role-taking experience. As expected, we ?nd
that auditors with high perspective-taking disposition are better
able to judge managers’ reported earnings than auditors with
low perspective-taking disposition. On the whole, our ?ndings
highlight the importance of perspective taking as a means to pro-
mote audit quality.
We recognize that our study is subject to various limitations.
Our experimental setting abstracts away from a speci?c audit con-
text. However, the use of an abstract setting allows us to carefully
focus on the underlying constructs and, importantly, to directly
test our theoretical arguments. Indeed, our experimental ?ndings
support the underlying theory. In designing the experimental task,
we sought to incorporate the key features of auditor–manager
interactions, speci?cally both parties’ situational incentives.
Importantly, we are not aware of any audit content factor that
would interact with role-taking experience or perspective-taking
ability in such a way as to alter our ?ndings. We leave this issue
for future study.
In addition, our study focuses on the cognitive aspects of per-
spective taking – mentally understanding another’s point of view.
But, other aspects of perspective taking can come into play in
strategic interactions. For example, affective empathy may be
important in gauging others’ behavior (e.g., Batson, 2009; Davis,
1983; Davis et al., 1987). Auditors may bene?t from sensitivity to
their counterpart’s feelings, including situational stress and dis-
comfort that are triggered by the reporting environment, which
impact managers’ actions. In our experimental setting, the affective
aspects of perspective taking are suppressed (to avoid potential
confounds). Our experimental task is calculative and game-like,
and it precludes social interaction (e.g., auditors observe managers’
reports and nothing else). Accordingly, our study does not shed
light on the associations between affective empathy, perspective
taking, and auditor performance. We readily acknowledge that
the emotional aspects of perspective taking are important, and
we encourage others to study this area in the future.
Despite the limitations, our results directly point to the bene?ts
of enhanced perspective taking in promoting audit quality. Our
?ndings suggest that audit ?rms can bene?t from hiring individu-
als who have been on the corporate side of the ?nancial-reporting
process, including boomerangs. Audit ?rms also may be able to
capture the bene?ts of role-taking experience by implementing
training programs that include role taking. In addition, our ?ndings
suggest that audit ?rms may want to consider dispositional per-
spective taking in staf?ng assignments. The importance of includ-
ing some team members with high perspective taking likely is
magni?ed when the auditor and client are involved in resolving
signi?cant disagreements over accounting matters. Finally, we
encourage audit ?rms to consider other ways to facilitate perspec-
tive taking as a means to improve auditor performance.
References
Altiero, E., Kang, Y. J., & Peecher, M. E. (2014). The investor perspective and its
in?uence on auditor materiality judgments. Working paper. University of
Illinois.
Anderson, C. M., & Camerer, C. F. (2000). Experience-weighted attraction learning in
sender–receiver signaling games. Economic Theory, 16(3), 689–718.
Badal, J. (2006). On the job: More ’boomerangs’ return to old jobs. The Wall Street
Journal 15 October.
Bandura, A. (1993). Perceived self-ef?cacy in cognitive development and
functioning. Educational Psychologist, 28(2), 117–148.
Batson, C. D. (2009). These things called empathy: Eight related but distinct
phenomena. In J. Decety & W. Ickes (Eds.), The social neuroscience of empathy
(pp. 3–15). Cambridge, MA: MIT Press.
Bazinger, C., & Kühberger, A. (2012). Is social projection based on simulation or
theory? Why new methods are needed for differentiating. New Ideas in
Psychology, 30(3), 328–335.
Berry, J. M. (1987). A self-ef?cacy model of memory performance. In Paper presented
at the meeting of the American Psychological Association, New York.
Berry, J. M., & West, R. L. (1993). Cognitive self-ef?cacy in relation to personal
mastery and goal setting across the life span. International Journal of Behavioral
Development, 16(2), 351–379.
Biddle, B. J. (1986). Recent development in role theory. Annual Review of Sociology,
12, 67–92.
Bloom?eld, R. J. (1995). Strategic dependence and inherent risk assessments. The
Accounting Review, 70(1), 71–90.
Bonner, S., & Lewis, B. L. (1990). Determinants of auditor expertise. Journal of
Accounting Research, 28(3), 1–20.
Bowlin, K. (2011). Risk-based auditing, strategic prompts, and auditor
sensitivity to the strategic risk of fraud. The Accounting Review, 86(4),
1231–1253.
Bowlin, K. O., Hales, J., & Kachelmeier, S. J. (2009). Experimental evidence of how
prior experience as an auditor in?uences managers’ strategic decisions. Review
of Accounting Studies, 14(1), 63–87.
Brandts, J., & Holt, C. H. (1992). An experimental test of equilibrium dominance in
signaling games. American Economic Review, 82(5), 1350–1365.
Brandts, J., & Holt, C. H. (1993). Adjustment patterns and equilibrium selection in
experimental signaling games. International Journal of Game Theory, 22(3),
279–302.
Budescu, D. V., Peecher, M. E., & Solomon, I. (2012). The joint in?uence of the extent
and nature of audit evidence, materiality thresholds, and misstatement type on
achieved audit risk. Auditing: A Journal of Practice and Theory, 31(2), 19–41.
Burns, S. M., & Brainerd, C. J. (1979). Effects of constructive and dramatic play on
perspective taking in very young children. Developmental Psychology, 15(5),
512–521.
Chalmers, J. B., & Townsend, M. A. (1990). The effects of training in social
perspective taking on socially maladjusted girls. Child Development, 61(1),
178–190.
Chandler, M. J. (1973). Egocentrism and antisocial behavior: The assessment and
training of social perspective-taking skills. Developmental Psychology, 9(3),
326–332.
Colman, A. M. (2003). Depth of strategic reasoning in games. Trends in Cognitive
Sciences, 7(1), 2–4.
Cooper, D. J., Garvin, S., & Kagel, J. H. (1997a). Adaptive learning vs. equilibrium
re?nements in an entry limit pricing game. The Economic Journal, 107(442),
553–575.
Cooper, D. J., Garvin, S., & Kagel, J. H. (1997b). Signaling and adaptive learning in an
entry limit pricing game. RAND Journal of Economics, 28(4), 662–683.
Corcoran, K. O., & Mallinckrodt, B. (2000). Adult attachment, self-ef?cacy,
perspective taking, and con?ict resolution. Journal of Counseling and
Development, 78(4), 473–483.
Daneman, M., & Carpenter, P. A. (1980). Individual differences in working
memory and reading. Journal of Verbal Learning and Verbal Behavior, 19(4),
450–466.
Davis, M. H. (1980). A multidimensional approach to individual differences in
empathy. JSAS Catalog of Selected Documents in Psychology, 10, 85–104.
Davis, M. H. (1983). Measuring individual differences in empathy: Evidence for a
multidimensional approach. Journal of Personality and Social Psychology, 44(1),
113–126.
Davis, M. H., Conklin, L., Smith, A., & Luce, C. (1996). Effect of perspective taking on
the cognitive representation of persons: A merging of self and other. Journal of
Personality and Social Psychology, 70(4), 713–726.
Davis, M. H., Hull, J. G., Young, R. D., & Warren, G. G. (1987). Emotional reactions to
dramatic ?lm stimuli: The in?uence of cognitive and emotional empathy.
Journal of Personality and Social Psychology, 52(1), 126–133.
Deloitte (2011). Good riddance or good business – How should companies manage
alumni relationships? Deloitte debates. .
Dulas, M. R., & Duarte, A. (2013). The in?uence of directed attention at encoding on
source memory retrieval in the young and old: An ERP study. Brain Research,
1500, 55–71.
Epley, N., Caruso, E. M., & Bazerman, M. H. (2006). When perspective taking
increases taking: Reactive egoism in social interaction. Journal of Personality and
Social Psychology, 91(5), 872–889.
Epley, N., & Caruso, E. M. (2009). Perspective taking: Misstepping into others’ shoes.
In K. D. Markman, W. M. P. Klein, & J. A. Suhr (Eds.), Handbook of imagination and
mental simulation. New York, NY: Psychology Press.
Epley, N., Keysar, B., Boven, L. F., & Gilovich, T. (2004). Perspective taking as
egocentric anchoring and adjustment. Journal of Personality and Social
Psychology, 87(3), 327–339.
Feffer, M. H. (1959). The cognitive implications of role taking behavior. Journal of
Personality, 27(2), 152–168.
Fiske, S. T., Taylor, S. E., Etcoff, N. L., & Laufer, J. K. (1979). Imaging, empathy, and
causal attribution. Journal of Experimental Social Psychology, 15(4), 356–377.
Fleeson, W. (2001). Towards a structure- and process-integrated view of
personality: Traits as density distributions of states. Journal of Personality and
Social Psychology, 80(6), 1011–1027.
Fleeson, W., & Noftle, E. E. (2008). Where does personality have its in?uence? A
super-matrix of consistency concepts. Journal of Personality, 76(6), 1355–1386.
50 B.K. Church et al. / Accounting, Organizations and Society 45 (2015) 40–51
Galinsky, A. D., & Moskowitz, G. B. (2000). Perspective-taking: Decreasing
stereotype expression, stereotype accessibility, and in-group favoritism.
Journal of Personality and Social Psychology, 78(4), 708–724.
Galinsky, A. D., & Mussweiler, T. (2001). First offers as anchors: The role of
perspective-taking and negotiator focus. Journal of Personality and Social
Psychology, 81(4), 657–669.
Galper, R. E. (1976). Turning observers into actors: Differential causal attributions
as a function of ‘‘empathy’’. Journal of Research in Personality, 10(3), 328–
335.
Hedden, T., & Zhang, J. (2002). What do you think I think you think?: Strategic
reasoning in matrix games. Cognition, 85(1), 1–36.
Heider, F. (1958). The psychology of interpersonal relations. New York, NY: Wiley.
Hidi, S. (1995). A reexamination of the role of attention in learning from text.
Educational Psychology Review, 7(4), 323–350.
Hoffman, V. B., & Zimbelman, M. F. (2009). Do strategic reasoning and
brainstorming help auditors change their standard audit procedures in
response to fraud risk? The Accounting Review, 84(3), 811–837.
Holmes, D. S. (1968). Dimensions of projection. Psychological Bulletin, 69(4),
248–268.
Hurtt, R. K. (2010). Development of a scale to measure professional skepticism.
Auditing: A Journal of Practice & Theory, 29(1), 149–171.
Hyland, T. (2006). ‘Alumni’ programs aim to bring back former employees.
Philadelphia Business Journal (April 3, 2006).
Iannotti, R. J. (1978). Effect of role-taking experiences on role taking, empathy,
altruism, and aggression. Developmental Psychology, 14(2), 119–124.
James, W. (1890). The principles of psychology. New York: Holt.
Johnson, D. W. (1967). The use of role reversal in intergroup competition. Journal of
Personality and Social Psychology, 7(2), 135–141.
Johnson, D. W., & Johnson, R. T. (1982). Joining together: Group theory and group
skills. Englewood Cliffs, NJ: Prentice Hall.
Kahneman, D. (1973). Attention and effort. Englewood Cliffs, NJ: Prentice Hall.
Kahneman, D., Knetsch, J. L., & Thaler, R. H. (1991). Anomalies: The endowment
effect, loss aversion, and status quo bias. The Journal of Economic Perspectives,
5(1), 193–206.
King, R. R. (2002). An experimental investigation of self-serving biases in an
auditing trust game: The effect of group af?liation. The Accounting Review, 77(2),
265–284.
Krueger, J., & Clement, R. W. (1994). The truly false consensus effect: An
ineradicable and egocentric bias in social perception. Journal of Personality
and Social Psychology, 67(4), 596–610.
Kübler, D., Müller, W., & Normann, H.-T. (2008). Job-market signaling and
screening: An experimental comparison. Games and Economic Behavior, 64(1),
219–236.
Magilke, M., Mayhew, B., & Pike, J. (2009). Are independent audit committee
members objective? The Accounting Review, 84(6), 1959–1981.
Majors, T. M., Shefchik, L. B., & Vitalis, A. M. (2014). The interactive effect of ego
depletion and trait skepticism on auditor performance. Working paper.
University of Illinois at Urbana-Champaign, Indiana University, and Georgia
Tech.
Mead, G. H. (1934). Mind, self, and society. Chicago: University of Chicago Press.
Moreno, J. L. (1934). Who shall survive?: A new approach to the problem of human
interrelations. Nervous and mental disease monograph series (Vol. 58, pp. 22–66).
Washington, DC: Nervous and Mental Disease Publishing Co.
Moreno, J. L. (1946). Role theory and role practice. Psychodrama (Vol. 1). Beacon, NY:
Beacon House.
Musso, M., Kyndt, E., Cascallar, E., & Dochy, F. (2012). Predicting mathematical
performance: The effect of cognitive processes and self-regulation factors.
Education Research International, 1–13.
Niedenthal, P. M., Barsalou, L. W., Winkielman, P., Krauth-Gruber, S., & Ric, F. (2005).
Embodiment in attitudes, social perception, and emotion. Personality and Social
Psychology Review, 9(3), 184–211.
Oberauer, K. (2002). Access to information in working memory: Exploring the focus
of attention. Journal of Experimental Psychology: Learning, Memory, and Cognition,
28(3), 411–421.
Oberauer, K. (2009). Design for a working memory. Psychology of Learning and
Motivation, 51, 45–100.
Peecher, M. E., Schwartz, R., & Solomon, I. (2007). It’s all about audit quality:
Perspectives on strategic-systems auditing. Accounting, Organizations and
Society, 32(4–5), 463–485.
Perner, J., & Wimmer, H. (1985). John thinks that Mary thinks that. . . Attribution of
second-order beliefs by 5- to 10-year-old children. Journal of Experimental Child
Psychology, 39(3), 437–471.
Piaget, J. (1932). The moral judgment of the child. London: Kegan, Paul, Trench,
Trubner.
Piaget, J. (1950). The psychology of intelligence. London: Routledge and Kegan Paul.
Posner, M. I., & Petersen, S. E. (1990). The attention system of the human brain.
Annual Review of Neuroscience, 13, 25–42.
Potters, J., & van Winden, F. (1996). Comparative statics of a signaling game:
An experimental study. International Journal of Game Theory, 25(1), 329–
353.
Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling strategies for
assessing and comparing indirect effects in multiple mediator models. Behavior
Research Methods, 40(3), 879–891.
Regan, D. T., & Totten, J. (1975). Empathy and attribution: Turning observers into
actors. Journal of Personality and Social Psychology, 32(5), 850–856.
Rerko, L., & Oberauer, K. (2013). Focused, unfocused, and defocused information in
working memory. Journal of Experimental Psychology: Learning, Memory, and
Cognition, 39(4), 1075–1096.
Rizzolatti, G., & Craighero, L. (2004). The mirror-neuron system. Annual Review of
Neuroscience, 27, 169–192.
Rose, J. M. (2007). Attention to evidence of aggressive ?nancial reporting and
intentional misstatement judgments: Effects of experience and trust. Behavioral
Research in Accounting, 19, 215–229.
Ross, L., & Nisbett, R. E. (1991). The person and the situation: Perspectives of social
psychology. New York, NY: McGraw-Hill.
Sco?eld, S. B., Phillips, T. J., Jr., & Bailey, C. D. (2004). An empirical reanalysis of the
selection-socialization hypothesis: A research note. Accounting, Organizations
and Society, 29(5–6), 543–563.
Sessa, V. I. (1996). Using perspective taking to manage con?ict and affect in teams.
The Journal of Applied Behavioral Science, 321(1), 101–115.
Smith, P. L., & Sewell, D. K. (2013). A competitive interaction theory of attentional
selection and decision making in brief, multielement displays. Psychological
Review, 120(3), 589–627.
Sogunro, O. A. (2004). Ef?cacy of role-playing pedagogy in training leaders: Some
re?ections. Journal of Management Development, 23(4), 355–371.
Stankov, L. (2000). Complexity, metacognition, and ?uid intelligence. Intelligence,
28(2), 121–143.
Stankov, L., & Lee, J. (2008). Con?dence and cognitive test performance. Journal of
Educational Psychology, 100(4), 961–976.
Staub, E. (1971). The use of role playing and induction in children’s learning of
helping and sharing behavior. Child Development, 42(3), 805–816.
Trotman, K. T., Wright, A., & Wright, S. (2005). Auditor negotiations: An
examination of the ef?cacy of intervention methods. The Accounting Review,
80(1), 349–367.
Turner, R. (2006). Role theory. In J. H. Turner (Ed.), Handbook of Sociological Theory.
New York, NY: Springer.
Van Boven, L. F., Dunning, D., & Loewenstein, G. (2000). Egocentric empathy gaps
between owners and buyers: Misperceptions of the endowment effect. Journal
of Personality and Social Psychology, 7(1), 66–76.
Van Boven, L. F., Loewenstein, G., & Dunning, D. (2005). The illusion of courage in
social predictions: Underestimating the impact of fear of embarrassment in
other people. Organizational Behavior and Human Decision Processes, 96(2),
130–141.
West, P. M. (1996). Predicting preferences: An examination of agent learning.
Journal of Consumer Research, 23(1), 68–80.
Wilson, B. J., Will, M. N., Schoen?eld-McNeill, J., & Montague, R. (2013). The
attention skills of boys with and without developmental delays: Associations
with academic performance. Early Child Development & Care, 183(5), 651–666.
Wood, R., & Bandura, A. (1989). Impact of conceptions of ability on self-regulatory
mechanisms and complex decision making. Journal of Personality and Social
Psychology, 56(3), 407–415.
Wyer, R. S., Srull, T. K., Gordon, S. E., & Hartwick, J. (1982). Effects of processing
objectives on the recall of prose material. Journal of Personality and Social
Psychology, 43(4), 674–688.
Zimbelman, M. F., & Waller, W. S. (1999). An experimental investigation of auditor–
auditee interaction under ambiguity. Journal of Accounting Research, 37(3),
135–156.
B.K. Church et al. / Accounting, Organizations and Society 45 (2015) 40–51 51
doc_453482365.pdf
We examine the effect of perspective taking on auditors’ ability to evaluate managers’ reported earnings
and, in turn, contribute to high-quality financial reporting. Using an experimental-economics approach,
we design two experiments to investigate auditor – manager interactions. In our first experiment, we
manipulate auditors’ prior experience in the manager’s role. We predict and find that role-taking experience
stimulates perspective taking, which allows auditors to more readily put themselves ‘‘in the manager’s
shoes,’’ benefitting financial-reporting quality. In our second experiment, we examine dispositional
perspective taking, focusing on individuals’ propensity to spontaneously take the viewpoint of another,
as a dimension of personality.
Perspective taking in auditor–manager interactions: An experimental
investigation of auditor behavior
q
Bryan K. Church
a,1
, Marietta Peytcheva
b,?,1
, Wei Yu
c,1
, Ong-Ard Singtokul
d
a
Scheller College of Business, Georgia Institute of Technology, United States
b
Department of Accounting, College of Business and Economics, Lehigh University, United States
c
Department of Economics, Hunter College, City University of New York, United States
d
Chulalongkorn Business School, Chulalongkorn University, Charoen Pokphand Group Company Limited, Thailand
a r t i c l e i n f o
Article history:
Received 17 September 2012
Revised 20 April 2015
Accepted 7 July 2015
Available online 25 July 2015
Keywords:
Perspective taking
Role-taking experience
Individual difference
Experimental economics
a b s t r a c t
We examine the effect of perspective taking on auditors’ ability to evaluate managers’ reported earnings
and, in turn, contribute to high-quality ?nancial reporting. Using an experimental-economics approach,
we design two experiments to investigate auditor – manager interactions. In our ?rst experiment, we
manipulate auditors’ prior experience in the manager’s role. We predict and ?nd that role-taking expe-
rience stimulates perspective taking, which allows auditors to more readily put themselves ‘‘in the man-
ager’s shoes,’’ bene?tting ?nancial-reporting quality. In our second experiment, we examine dispositional
perspective taking, focusing on individuals’ propensity to spontaneously take the viewpoint of another,
as a dimension of personality. We predict and ?nd that auditors with high perspective-taking disposition
are better able to judge managers’ reported earnings than auditors with low perspective-taking disposi-
tion. Taken together, the results of our two experiments highlight the importance of perspective taking as
a means to enhance auditors’ performance in strategic interactions with managers.
Ó 2015 Elsevier Ltd. All rights reserved.
Introduction
This paper reports the results of two experiments designed to
examine the effect of perspective taking on auditors’ ability to pro-
mote high-quality ?nancial reporting. We de?ne perspective tak-
ing as the capacity to entertain the psychological point of view of
another (Davis, Conklin, Smith, & Luce, 1996). Our focus is on cog-
nitive perspective taking, which entails understanding, as accu-
rately as possible, another’s thoughts, attitudes, or concerns in a
speci?c situation (Epley, Caruso, & Bazerman, 2006). We use an
experimental-economics approach to examine auditor–manager
interactions, where players have con?icting interests. Speci?cally,
we investigate how perspective taking affects auditors’ assessment
of managers’ reporting choices, including auditors’ propensity to
identify and curtail reporting bias. We contend that successful per-
spective taking allows auditors to develop better mental models of
clients’ earnings, which leads to enhanced ?nancial-reporting
quality (Peecher, Schwartz, & Solomon, 2007).
Prior studies suggest that effective perspective taking improves
individuals’ judgments and decision making. Successfully taking
another’s perspective can reduce anchoring effects, con?rmation
bias, actor–observer bias, and in-group favoritism (Galinsky &
Mussweiler, 2001). Moreover, being able to take the perspective
of one’s counterpart leads to more bene?cial outcomes for self
(Galinsky & Mussweiler, 2001). In an auditing context, Altiero,
Kang, and Peecher (2014) document that auditors who are
prompted to take an investor’s perspective, by completing a series
of investor-minded tasks, provide higher-quality materiality judg-
ments than auditors who are not prompted. Our study comple-
ments Altiero et al. (2014) by examining the linkage between
perspective taking and auditor behavior in strategic interactions
with managers.
In our ?rst experiment, we examine whether role-taking
experience stimulates auditors’ perspective taking. We contendhttp://dx.doi.org/10.1016/j.aos.2015.07.001
0361-3682/Ó 2015 Elsevier Ltd. All rights reserved.
q
We gratefully acknowledge helpful comments from Jim Dearden, Jeffrey Hales,
Mark Peecher, Timothy Shields, T. Jeffrey Wilks, Ping Zhang, and two anonymous
reviewers, as well as participants in a workshop at Rutgers University and the 2010
Auditing Section Midyear Meeting in San Diego, CA. The authors acknowledge the
?nancial support by the Scheller College of Business, Georgia Institute of Technol-
ogy, a Faculty Research Grant and a grant from the Martindale Center at Lehigh
University, and a research grant from Haslam College of Business, the University of
Tennessee. We also appreciate the research assistance of John Castonguay, Lauren
Reid, and Jonathan Shipman.
?
Corresponding author.
E-mail addresses: [email protected] (B.K. Church), map608@
lehigh.edu (M. Peytcheva), [email protected] (W. Yu), [email protected]
(O.-A. Singtokul).
1
The names of the ?rst three authors are listed in alphabetical order.
Accounting, Organizations and Society 45 (2015) 40–51
Contents lists available at ScienceDirect
Accounting, Organizations and Society
j our nal homepage: www. el sevi er . com/ l ocat e/ aos
that role-taking experience enhances auditors’ understanding of
the manager’s viewpoint, which bene?ts auditors’ performance.
Audit ?rms, especially the Big Four, have spent increasing amounts
of resources to recruit former employees, commonly known as
boomerangs (Badal, 2006; Deloitte, 2011). Firms’ recruiting direc-
tors assert that auditors who return, after having spent time in
industry, bring back ‘‘stronger knowledge, a broader sense of expe-
rience, a broader skill set’’ to the ?rm (Hyland, 2006). Our experi-
ment sheds light on an advantage that accrues to audit ?rms by
hiring employees from industry.
We manipulate auditors’ prior experience in the manager’s role
(experience versus no experience), examining its effect on auditors’
behavior. We predict and ?nd that role-taking experience stimu-
lates perspective taking. Auditors with role-taking experience more
accurately estimate managers’ reported earnings, as compared to
auditors without such experience, and in turn make better report-
ing decisions, which promotes high-quality ?nancial reporting.
We design a second experiment to further investigate the effect
of perspective taking on auditors’ capacity to evaluate managers’
reported earnings (i.e., to assess whether earnings are materially
misstated). We focus on dispositional perspective taking, represent-
ing individuals’ natural ability to spontaneously take the viewpoint
of another. Our second experiment allows us to cleanly examine the
effect of perspective taking on auditor-participants’ task perfor-
mance, apart from the effect of situational factors that also might
in?uence perspective taking. We use an established personality
measure to appraise auditor-participants’ perspective-taking dispo-
sition (Davis, 1980, 1983). We ?nd that auditors with high
perspective-taking disposition are better able to judge managers’
reported earnings than auditors with lowperspective-taking dispo-
sition. This result suggests that perspective-taking disposition is an
important individual trait that underlies auditors’ performance.
Accordingly, our study complements other auditing-based research
that investigates individuals’ characteristics, including disposi-
tional tendencies (e.g., Bonner & Lewis, 1990; Hurtt, 2010; Majors,
Shefchik, & Vitalis, 2014; Rose, 2007; Sco?eld, Phillips, & Bailey,
2004).
Our study makes several contributions to the extant literature on
audit quality. First, we provide evidence that perspective taking
improves auditors’ ability to accurately assess managers’ reported
earnings. The takeaway is that effective perspective taking is bene?-
cial to audit quality. Second, we provide evidence that role-taking
experience stimulates perspective taking. Furthermore, some indi-
viduals quite naturally are better at perspective taking than others,
aside fromthe effects of role-taking experience. Those who are better
at perspective taking, in turn, may have an advantage in strategic
interactions as compared to others. Third, our ?ndings suggest that
auditors who possess better mental models of true earnings,
indeed, are able to make better decisions, which enhances
?nancial-reporting quality (Peecher et al., 2007). While this link often
is assumed in the literature, we are able to provide empirical evi-
dence of its validity. We suggest that auditors might bene?t by mak-
ing earnings estimates of reporting segments as a way to judge the
suf?ciency and accuracy of their mental models for a client’s overall
earnings process (see also Budescu, Peecher, & Solomon, 2012).
The results of our two experiments, taken together, have impor-
tant implications for audit practice. First, audit ?rms can bene?t
from hiring auditors with prior experience in the corporate world,
especially those involved in the ?nancial-reporting process. Audit
?rms’ efforts to recruit accountants from industry, including
boomerangs, appear to be prudent as industry experience may
aid perspective taking (i.e., anticipating managers’ behavior and
actions). Second, audit ?rms can bene?t from targeted training
programs that involve role taking. Such training programs poten-
tially represent an ef?cient means to boost audit quality. As
pointed out by Trotman, Wright, and Wright (2005), role-taking
training is a relatively low-cost technique, as it can be conducted
with other participants from the same ?rm. Third, audit ?rms
may want to consider dispositional perspective taking in staf?ng
assignments, for example, ensuring that some team members have
high perspective-taking disposition. The importance of including
such individuals on audit teams likely is magni?ed when the audi-
tor and client are involved in resolving signi?cant disagreements
over accounting matters. Lastly, audit ?rms are advised to consider
other ways to facilitate perspective taking (e.g., prompts in audit
programs) as a means to improve auditor performance and, ulti-
mately, audit quality.
The remainder of this paper is organized as follows. Section 2
reviews the related literature, which provides a basis to develop
our research hypotheses. Section 3 describes our ?rst experiment,
and Section 4 presents the experimental results, along with a dis-
cussion of the ?ndings. Section 5 describes our second experiment
and, subsequently, presents the results. Section 6 offers concluding
remarks.
Background, theory, and hypotheses
Background
The importance of perspective taking in cognitive processes is
examined extensively in Piaget’s (1932, 1950) early work.
Individuals take another’s perspective by simulating another’s
internal states, and the form of simulation can vary depending on
task demands (Niedenthal, Barsalou, Winkielman, Krauth-Gruber,
& Ric, 2005). The aim of perspective taking is to put oneself ‘‘in
another’s shoes’’ and to effectively take another’s vantage. Prior
studies document numerous cognitive bene?ts of successful per-
spective taking, which arise in various social interactions (e.g.,
Davis et al., 1996; Galinsky & Moskowitz, 2000; Galinsky &
Mussweiler, 2001; Galper, 1976; Johnson, 1967; Johnson &
Johnson, 1982; Regan & Totten, 1975; Sessa, 1996).
Altiero et al. (2014) investigate perspective taking in an audit-
ing context, noting that regulatory standards require auditors to
consider investors’ perspective when assessing materiality. The
authors conduct an experiment, with experienced auditors as par-
ticipants, and use a series of investor-minded tasks to stimulate
auditors to actively take an investor’s perspective. The ?ndings
indicate that prompting auditors to take the investor’s perspective,
by using investor-minded tasks, enables them to discriminate
between potential audit adjustments that are more or less likely
to be material for qualitative reasons. Otherwise, auditors fail to
make the distinction. The results hold for both specialist and
non-specialist auditors, but are more pronounced for specialists.
2
These ?ndings underscore the importance of successful perspective
taking as a means to reinforce audit quality.
As compared to Altiero et al. (2014), we examine perspective
taking in strategic interactions with managers. We maintain that
perspective taking improves auditors’ strategic reasoning – the
process whereby an agent reasons about the best strategy to adopt
in a multi-player scenario, taking into account the likely behavior
of the counterpart(s).
3
Without question, strategic reasoning is an
2
When prompted, specialist auditors draw upon rich, domain-speci?c knowledge,
permitting them to hone in on qualitative facets of materiality, which affects their
evaluation of potential audit adjustments (e.g., whether adjustments affect the trend
of key performance indicators).
3
In strategic interactions, zero-order reasoning occurs when agents consider their
own incentives, but not their counterparts’ incentives. Zero-order reasoning involves
‘‘no understanding of the desires, beliefs, or thoughts of others’’ (Hedden & Zhang,
2002) and can lead to myopic choices. In contrast, higher-order reasoning involves
strategic consideration of the incentives of one’s counterparts as well as consideration
of counterparts’ beliefs about, and anticipation of, one’s own incentives (Colman,
2003; Hedden & Zhang, 2002; Perner & Wimmer, 1985).
B.K. Church et al. / Accounting, Organizations and Society 45 (2015) 40–51 41
important part of auditing (Bloom?eld, 1995; Bowlin, 2011), and
prior ?ndings suggest that strategic thinking can improve auditors’
performance (e.g., Bowlin, 2011; Hoffman & Zimbelman, 2009;
Zimbelman & Waller, 1999).
We claim that successful perspective taking facilitates strategic
reasoning in that it allows auditors to better understand managers’
problem space, including their reporting choices and incentives.
Notwithstanding, successfully taking another’s perspective can be
dif?cult. People tend to overestimate the extent to which others
share their point of view, and they often over-rely on their own
experiences when interpreting or predicting the behavior of others
(e.g., Epley, Keysar, Boven, & Gilovich, 2004; Van Boven, Dunning, &
Loewenstein, 2000; Van Boven, Loewenstein, & Dunning, 2005).
We are interested in ways to enhance perspective taking, help-
ing auditors put themselves in the manager’s shoes. We examine
whether gaining experience in the counterpart’s role enables audi-
tors to make better judgments and decisions in strategic interac-
tions with managers. Next, we consider role-taking experience
and its effect on perspective taking and performance.
Role-taking experience, perspective taking, and performance
Research in cognitive and social psychology emphasizes the
importance of role taking in the process of social learning and pro-
poses that successful interactions require the ability to ‘‘take the
role of the other’’ (Biddle, 1986; Mead, 1934; Moreno, 1934,
1946). This line of research recognizes that the same individual
might reason and act quite differently in different roles and, like-
wise, that different individuals may behave similarly in similar
roles (Turner, 2006). Work grounded in the tradition of Piaget
(1932, 1950) highlights role taking as an antecedent of successful
perspective taking (e.g., Feffer, 1959).
Previous studies provide evidence that role-taking experience
improves perspective taking (e.g., Burns & Brainerd, 1979;
Chandler, 1973; Iannotti, 1978; Staub, 1971).
4
Chalmers and
Townsend (1990) ?nd that role-taking experience facilitates under-
standing others’ problems and incentives. Sessa (1996) shows that
role-taking experience in?uences individuals’ perceptions of con?ict,
primarily because it enables better insight into others’ perspective.
In auditing, Trotman et al. (2005) ?nd that role-taking experience
helps auditors develop successful strategies when negotiating with
clients.
Prior research indicates that role-taking experience helps indi-
viduals combat the tendency to overestimate the similarities
between self and others, referred to as social-projection bias.
5
Johnson (1967) ?nds that role-taking experience leads opponents
to gain a clearer understanding of others’ position. Van Boven et al.
(2000) show that social-projection bias, attributable to the endow-
ment effect (Kahneman, Knetsch, & Thaler, 1991), can be overcome
by equipping individuals with experience in another’s role. In con-
trast, the same study ?nds that providing information about the
endowment effect through lecturing alone does not reduce partici-
pants’ social-projection bias. These ?ndings highlight the importance
of role-taking experience to improve perspective taking and under-
score that role-taking experience cannot necessarily be substituted
by secondhand encounters or common knowledge of incentives.
The linkage between role-taking experience and enhanced per-
spective taking is, by no means, a tautology or foregone conclusion.
For example, Bowlin, Hales, and Kachelmeier (2009) provide
evidence that, in strategic interactions, role-taking experience does
not eliminate social-projection bias. Their ?ndings suggest that the
effect of such bias on participants’ performance can be for better or
worse. By comparison, we maintain that effective perspective tak-
ing improves individuals’ performance.
6
We assert that role-taking
experience allows auditors to form more accurate mental represen-
tations of the manager’s incentives, options, and alternatives.
Role-taking experience can enhance auditors’ perspective taking by
sensitizing them to situational factors that affect managers’ report-
ing choices, most notably managers’ incentives. In strategic interac-
tions, successful perspective taking serves to amplify the salience of
the counterpart’s incentives (e.g., Fiske, Taylor, Etcoff, & Laufer,
1979; Wyer, Srull, Gordon, & Hartwick, 1982). The increased salience
directs auditors’ attention to managers’ incentives.
7
In our ?rst experiment, we manipulate auditors’ prior experi-
ence in the manager’s role, with the aim of stimulating auditors’
perspective taking. We do so guided by prior research in cognitive
and social psychology, which documents a link between
role-taking experience and perspective taking (Burns & Brainerd,
1979; Chalmers & Townsend, 1990; Chandler, 1973; Iannotti,
1978; Staub, 1971). As explained by Epley and Caruso (2009), three
conditions are necessary for successful perspective taking:
(a) actively attempting to place oneself in another’s role or posi-
tion; (b) reasoning about the other’s perspective in a deliberate
and effortful way in order to avoid dwelling on one’s point of view
and to infer the perceptions of the other; and (c) using diagnostic
information about the other being evaluated. In strategic interac-
tions, role-taking experience facilitates the realization of all three
conditions. When the auditor has prior experience as a manager,
the auditor has been in the other’s role and has unique insight into
how the other might reason and act.
Auditors with role-taking experience likely are attuned to the
manager’s incentives, and this experience provides a basis for them
to understand and gauge the manager’s behavior. This point is
intuited by experimental economists, who often make participants
switch roles in signaling experiments, primarily to improve partic-
ipants’ understanding of different parties’ incentives (e.g., Brandts
& Holt, 1992, 1993; Cooper, Garvin, & Kagel, 1997a, 1997b; Kübler,
Müller, & Normann, 2008; Potters & van Winden, 1996). In fact,
Anderson and Camerer (2000, 701) explicitly discuss this issue.
‘‘There is a strong intuition among experimentalists that players do
learn faster when they switch roles [. . .]. This is easily testable, by
4
Role-taking experience has been used across various ?elds to aid professionals in
acquiring the perspectives of important others (see Sogunro, 2004 for a review).
5
Social-projection bias is known under other labels, including ‘‘egocentric bias’’
(Epley et al., 2004), ‘‘false consensus’’ (Krueger & Clement, 1994; West, 1996),
‘‘attributive projection’’ (Holmes, 1968), and ‘‘egocentric attribution’’ (Bazinger &
Kühberger, 2012; Heider, 1958).
6
Role-taking experience may not eliminate social-projection bias in Bowlin et al.
(2009) for a variety of reasons. The experimental setting in Bowlin et al. (2009)
focuses solely on players’ payoffs. Each player makes a dichotomous choice: the
manager chooses to report cautiously or aggressively and the auditor chooses to be
diligent or lax. The combination of players’ choices determines their respective
payoffs. By comparison, our setting permits more variation in players’ responses. The
manager chooses a reported amount (i.e., reported earnings). The auditor then
estimates earnings, using the manager’s report. The auditor also decides whether to
accept or reject the manager’s report. By incorporating earnings in our design (i.e., a
?ner measure of managers’ strategic intentions and auditors’ expectations), we may
be able to better capture the bene?ts of role-taking experience, that being
improvements in perspective taking.
7
Substantial literature in cognitive psychology provides evidence on the relation-
ship between attention and cognitive performance (e.g., James, 1890; Kahneman,
1973; Posner & Petersen, 1990). Prior ?ndings suggest that focusing attention on a
speci?c item serves as a selection mechanism for the importance of the item: it
increases the accessibility of knowledge in working memory that is relevant to the
item as compared to other items (Oberauer, 2002, 2009; Rerko & Oberauer, 2013).
Focused attention increases the likelihood that an item will be ef?ciently transferred
into working memory and leads to better encoding of stimuli (Dulas & Duarte, 2013;
Smith & Sewell, 2013) and, thus, superior cognitive performance (Daneman &
Carpenter, 1980). Research in education also provides evidence of an association
between focused attention and improvements in cognitive performance (Hidi, 1995;
Musso, Kyndt, Cascallar, & Dochy, 2012; Wilson, Will, Schoen?eld-McNeill, &
Montague, 2013).
42 B.K. Church et al. / Accounting, Organizations and Society 45 (2015) 40–51
comparing experiments with different degrees of role-switching,
but we know of no such experiments.’’
Because role-taking experience leads to a heightened under-
standing of situational incentives, it provides a basis for auditors
to carefully assess managers’ reporting choices, including the
chance of reporting biases. Auditors who have been in the man-
ager’s shoes have ?rsthand experience making reporting choices
and seeing how others react to such choices. Simply put,
role-taking experience stimulates auditors’ perspective taking,
which dampens their propensity to anchor on managers’ reports
and, further, allows them to better evaluate whether managers’
earnings reports are materially misstated (Galinsky &
Mussweiler, 2001). Therefore, auditors are expected to bene?t
from prior experience in the manager’s role. Our research hypothe-
ses are expressed as follows.
H1. Auditors with role-taking experience (as a manager) are better
able to assess whether the manager’s reported earnings are
materially misstated than auditors without role-taking experience.
H2. Auditors’ perspective taking mediates the relationship
between role-taking experience and their assessment of the man-
ager’s reported earnings.
Research method
Overview
We recruit 58 students from a public university to participate in
our ?rst experiment. The average age of the participants is
21 years, and 33 are male. The vast majority are undergraduates
(83%) and most are business majors (88%). Our use of student par-
ticipants is appropriate in that we examine economic behavior in
an abstract setting, where players are rewarded based on their per-
formance. Other experimental-economics studies use a similar
approach (e.g., King, 2002; Magilke, Mayhew, & Pike, 2009).
Participants are randomly assigned to one of two groups: a
role-taking group and a non-role-taking group. The experiment
consists of six rounds and lasts approximately 90 min. In rounds
1 and 2, participants in the role-taking group are managers, and
those in the non-role-taking group are auditors.
8
In rounds 3–6,
all participants are auditors. Therefore, participants in the
role-taking group switch roles in round 3, whereas those in the
non-role-taking group remain in the same role throughout the
experiment. Participants’ role by round is shown in Table 1.
Task and procedures
Upon arrival, participants report to one of two rooms, where
they stay for the entirety of the experiment. Instructions are dis-
tributed and read aloud by a researcher. Next, the researcher pro-
vides an example of how each experimental round proceeds and
answers participants’ questions. Then, six experimental rounds
are administered.
At the beginning of each round, participants learn their roles,
and managers and auditors are paired. Each round proceeds as
follows.
1. The manager is endowed with a public and private signal of
earnings. The combined signals indicate earnings, such that
the manager knows actual earnings.
2. The auditor observes the public signal of earnings, but not the
private one.
3. The manager submits an earnings report, which the auditor
reviews.
9
The manager’s report does not have to be truthful
(e.g., it can be in?ated).
4. The auditor makes his/her own estimate of earnings, condi-
tioned on the manager’s reported amount.
5. The auditor decides whether to accept or reject the manager’s
earnings report.
6. Participants are informed of their payoff.
Procedurally, actual earnings are determined by summing two
numbers, denoted Integer I and Integer II. Integer I is the public sig-
nal, and Integer II is the private one. Both numbers are randomly
generated each round by drawing from a uniform distribution that
ranges from 0 to 50. Thus, earnings can range from 0 to 100.
The manager is incentivized to report higher earnings values, as
long as the auditor accepts the reported amounts. The auditor, on
the other hand, is incentivized to accept amounts that are close
to the actual value (i.e., within a speci?ed materiality threshold)
and otherwise reject reported amounts. Each player’s speci?c pay-
offs/incentives are common knowledge. The procedures repeat for
six rounds, where auditor–manager dyads are re-paired each
round.
In rounds 1 and 2, managers in the role-taking group submit
earnings reports to auditors in the non-role-taking group. In
rounds 3–6, such pairings are not possible because all participants
are auditors. We make this design choice to ensure that auditors
only receive earnings reports from managers who have not been
auditors previously: that is, to ensure that our design is not con-
founded by managers’ speci?c prior experience as an auditor. To
generate managers’ reports for rounds 3–6, we use amounts
reported, along with the corresponding actual earnings, froma pre-
vious experimental session that includes manager–auditor pairs
without role-taking experience.
10
We make participants aware of
this fact. After the six rounds are ?nished, participants complete a
post-experiment questionnaire. The questionnaire includes an item
to measure participants’ perspective taking. Speci?cally, we include
the following statement, ‘‘I understand the perspective of the sen-
der’’ (manager), where participants respond on a seven-point
Likert scale, anchored by 1 = strongly disagree and 7 = strongly
Table 1
Participants’ role by round: Experiment one.
Round 1 Round 2 Round
3
a
Round
4
Round
5
Round
6
Role-taking
group
Manager Manager Auditor Auditor Auditor Auditor
Non-role-
taking
group
Auditor Auditor Auditor Auditor Auditor Auditor
a
As shown, over rounds 3–6, all participants are auditors. We use managers’
reports, along with corresponding actual earnings, randomly selected from a pre-
vious experimental session that includes managers and auditors over all rounds.
8
In the experimental materials, we refer to the manager as the ‘‘sender’’ and the
auditor as the ‘‘receiver.’’ We use neutral labels to denote players’ roles in order to
guard against extraneous in?uences that are associated with the assigned role. For
expositional convenience, we use the manager/auditor terminology throughout the
paper.
9
In the experimental materials, the manager’s earnings report is referred to as a
report of commodity value.
10
In the previous session, the basic procedures and experimental parameters are
identical to those in experiment one. Prior to conducting experiment one, a research
assistant, with no knowledge of the experiment, randomly selected four managers’
reports, along with the corresponding actual earnings amounts, from the previous
session. As a consequence, we control for managers’ reports and clients’ actual
earnings over rounds 3–6.
B.K. Church et al. / Accounting, Organizations and Society 45 (2015) 40–51 43
agree. Participants’ responses represent a measure of their perceived
ability to take the counterpart’s perspective.
11
Managers’ payoff
The manager’s payoff per round is contingent on the auditor’s
decision to accept or reject the reported amount. If the auditor
accepts the manager’s earnings report, the manager’s payoff is
the reported amount. Otherwise, the manager’s payoff is 65% of
the actual earnings. The payoff choices derive from practical con-
siderations such that (1) higher reported amounts are preferable
(e.g., increasing the likelihood of meeting targets, increasing the
manager’s bonuses, and bene?tting the manager’s reputation)
and (2) auditor rejection is costly (e.g., delays in reporting, strained
relations with the auditor, and additional audit costs that, at some
point, are passed on to the client).
Auditors’ payoff
The auditor’s payoff is contingent on the auditor’s estimate of
earnings, the auditor’s accept/reject decision, and the manager’s
earnings report. The auditor relies on the manager’s report to
determine an estimated earnings amount. The auditor then consid-
ers whether to accept or reject the manager’s report. Relatedly,
Peecher et al. (2007) suggest that the accuracy of auditors’ earnings
estimates in?uences their reporting decisions, for the betterment
of ?nancial-reporting quality. Budescu et al. (2012) assert that
auditors’ expectations impact their assessment of material mis-
statement: auditors are more likely to conclude that reported
amounts are fairly stated when auditors’ evidence-based expecta-
tions are closer to managers’ reported amounts. For our purposes,
auditors’ earnings estimates are compared to managers’ reported
amounts and deviations are judged in light of the speci?ed materi-
ality threshold. Thus, auditors’ earnings estimates provide a basis
to accept or reject reported amounts, thereby impacting
?nancial-reporting quality.
The auditor’s payoff per round, depicted in Fig. 1, indicates that
the auditor’s payoff includes a ?xed component and a bonus com-
ponent. We discuss the two components below.
If the auditor accepts the manager’s report, the auditor’s ?xed
payoff is 50 when the reported amount is within ±10 of the actual
amount and 0 otherwise. Under these circumstances, the auditor
approves the client’s ?nancial statements without adjustment.
The ±10 represents materiality bounds, such that the manager’s
report is materially correct as long as it is within these bounds.
Accordingly, the auditor receives a high payoff for approving
reports that are materially correct and a low payoff for approving
reports that are materially misstated.
Next, we consider the auditor’s payoff when s/he rejects the
manager’s report. Now, the auditor’s ?xed payoff is 35 or 5, contin-
gent on the auditor’s estimate of earnings. Under these circum-
stances, the auditor’s estimate becomes the amount that is
re?ected in the client’s ?nancial statements. If the auditor’s esti-
mate is within the materiality bounds, and it is closer to the actual
earnings than the manager’s reported amount, the auditor’s ?xed
payoff is 35. In this case, the client’s ?nancial statements are mate-
rially correct. The payoff choice re?ects the practical reality that if
the auditor initially rejects the manager’s report, additional audit
work is required and client relations are strained, which reduces
the auditor’s ?xed payoff (as compared to a ?xed payoff of 50).
By comparison, if the auditor’s estimate is outside of the
materiality bounds or it is farther away from the actual earnings
than the manager’s reported amount, the auditor’s ?xed payoff is
5. In this case, the auditor’s estimate either is materially misstated
or inferior to the manager’s earnings report. Whichever the case,
the practical implications are that additional audit work is
required, client relations are strained, and outputs are inferior.
12
In addition, the auditor can earn a bonus each round based on
the earnings estimate. The bonus is computed as follows.
We provide a bonus to ensure that auditors are incentivized to be
accurate in their earnings estimate, irrespective of the accept/re-
ject decision.
In determining the experimental parameters (refer to Fig. 1), we
assume that the auditor’s objective is to approve materially correct
reports and otherwise reject reports. The auditor’s decision, in turn,
provides a basis for differences in the ?xed payoff component.
When the auditor rejects the manager’s report, we assume that
the auditor must determine an acceptable earnings amount to
report: that is, an amount within the materiality bounds. The audi-
tor contributes to high-quality ?nancial reporting as long as s/he
(1) approves materially correct reports or (2) rejects reports and
determines a value to report that is materially correct and superior
to that of managers’ reported amounts. For illustrative purposes,
Fig. 2 depicts when the auditor contributes to high-quality ?nancial
reporting: that is, the decision nodes where the auditor does and
does not bene?t ?nancial-reporting quality.
Results
We are interested in the effect of role-taking experience on
auditor-participants’ ability to determine whether managers’
reported earnings are materially misstated and, further, whether
perspective taking mediates the relationship. First, we consider
the accuracy of auditors’ estimates of the manager’s earnings. We
argue that auditors who make more accurate estimates are better
equipped to judge whether reported earnings are free of material
misstatement and, thus, to promote high-quality ?nancial report-
ing. Next, we scrutinize whether role-taking experience facilitates
auditors’ perspective taking, which allows us to get at the cognitive
process underlying auditors’ judgments. Lastly, we perform addi-
tional analyses and discuss our ?ndings.
Tests of H1
Our ?rst hypothesis predicts that auditors with role-taking
experience are better able to assess whether managers’ reported
Bonus ¼ ð100 À jAuditor’s Estimate of Earnings ÀClient’s Actual EarningsjÞ=10:
11
This measure emphasizes the cognitive aspect of perspective taking (Anderson &
Camerer, 2000; Epley & Caruso, 2009).
12
Economic theory predicts that the manager reports a value such that the auditor
is indifferent to accepting or rejecting the amount: the auditor accepts any amount
equal to or below the value and rejects any amount above it. Based on our
experimental parameters, the economic predictions are that (1) the manager reports
Integer I + 43 and (2) the auditor accepts the reported amount. Details are available
from the authors upon request.
44 B.K. Church et al. / Accounting, Organizations and Society 45 (2015) 40–51
earnings are materially misstated than auditors without such
experience. We examine auditor-participants’ estimates of
earnings over rounds 3–6: participants in both experimental
groups provide estimates of earnings in rounds 3–6. We compute
the average absolute estimation error per round for each partici-
pant as follows.
By computing an average error, we generate one independent
observation per participant, which allows us to avoid potential
problems that arise if participants’ estimation errors are correlated
across rounds.
Table 2 presents descriptive statistics for auditors’ average
absolute estimation error, partitioning the data by the two exper-
imental groups. The mean (median) estimation error of the
role-taking group is less than that of the non-role-taking group:
7.96 (7.00) versus 10.43 (9.75), respectively. A parametric t-test
and a nonparametric Mann–Whitney test show that the difference
Auditor’s Acceptance
of the Manager’s
Reported Value
|D| ? 10
|E| ? 10 and |E| ? |D|
0
points + Bonus
35
points + Bonus
5
points + Bonus
Yes
No
No
No
50
points + Bonus Yes
Yes
Fig. 1. Auditor’s payoff per round. Note: D = Manager’s Earnings Report À Actual Earnings, E = Auditor’s Estimate of Earnings À Actual Earnings, and Bonus = (100 À |Auditor’s
Estimate of Earnings À Actual Earnings|)/10.
Basis of
Financial
Reporting:
Is the Auditor
Contributing to
High-Quality
Financial
Reporting?
YES
Financial
Reporting is
based on the
Manager’s
Report
Is the
Manager’s
Report within
the materiality
bounds?
NO
Does the
Auditor Accept
the Manager’s
Report?
YES
Is the
Auditor’s
Estimate better
than the
Manager’s
Report? NO
Financial
Reporting is
based on the
Auditor’s
Estimate
Is the Auditor’s
Estimate within
the materiality
bounds?
NO
YES
YES
NO
YES
NO
YES
NO
NO
Fig. 2. The auditor’s role in high-quality ?nancial reporting.
Estimation Error ¼
X
jAuditor’s Estimate of Earnings ÀClient’s Actual Earningsj=4:
B.K. Church et al. / Accounting, Organizations and Society 45 (2015) 40–51 45
between the two groups is statistically signi?cant at conventional
levels: t(56) = À2.10, p = 0.02, one-tailed, and Z = À2.52, p = 0.006,
one-tailed, respectively. Inferences are unaffected if we perform a
repeated measures analysis of variance, using participants’ abso-
lute estimation error per round as the dependent variable. Our
?ndings are consistent with H1 and suggest that role-taking expe-
rience enhances auditors’ performance.
We further probe the accuracy of auditors’ earnings estimates.
Fig. 3 presents the frequency of participants’ estimation errors over
rounds 3–6, partitioned into categories based on the magnitude of
the error. The frequencies are presented separately for each exper-
imental group. Fisher’s exact test indicates that the proportion of
absolute estimation errors that falls within the materiality bounds
(i.e., |E| 6 10) is signi?cantly greater for the role-taking group than
the non-role-taking group at p = 0.003, one-tailed. These ?ndings
provide further support for H1.
High-quality ?nancial reporting and links to the accuracy of earnings
estimates
Next we consider the effect of role-taking experience on audi-
tors’ decisions. Panel A of Table 3 presents descriptive data, sum-
marizing the percentage of time that auditors’ decisions
contribute to ?nancial-reporting quality (refer to Fig. 2), parti-
tioned by role-taking group. We perform a parametric t-test and
a nonparametric Mann–Whitney test, collapsing participants’ deci-
sions across the four rounds. We ?nd that auditors in the
role-taking group are more likely to make decisions that contribute
to ?nancial-reporting quality than auditors in the non-role-taking
group: t(56) = 2.58, p = 0.007, one-tailed, and Z = 2.44, p = 0.007,
one-tailed, respectively. We also examine whether auditors’
accept/reject decisions are correct, irrespective of their
earnings estimates. That is, we examine the percentage of time
that auditors accept reported amounts that are within ±10 of the
actual amount and otherwise reject reported amounts.
Panel B of Table 3 presents descriptive data. Again, we perform
parametric and nonparametric tests and ?nd that auditors in the
role-taking group are more likely to make correct decisions than
auditors in the non-role-taking group: t(56) = 2.17, p = 0.017,
one-tailed, and Z = 1.85, p = 0.03, one-tailed, respectively. Our ?nd-
ings suggest that role-taking experience fosters ?nancial-reporting
quality.
We conduct further analyses to assess the link between the
accuracy of auditors’ earnings estimates and ?nancial-reporting
quality. As suggested earlier, auditors who formulate more accu-
rate estimates of actual earnings make sounder decisions, bene?t-
ting ?nancial-reporting quality (see also Peecher et al., 2007). To
formally test this relationship, we regress the number of times that
auditors’ decisions contribute to ?nancial-reporting quality (over
rounds 3–6) on their average absolute estimation error.
We ?nd that the coef?cient of the average estimation error is neg-
ative and statistically signi?cant (b = À0.18, t(56) = À11.25,
p < 0.001). We repeat the analysis substituting the number of times
that auditors make correct accept/reject decisions as the depen-
dent variable and results are similar (b = À0.13, t(56) = À6.77,
p < 0.001).
13
Our ?ndings suggest that the accuracy of
auditors’ earnings estimates are associated with better decisions.
Auditors who have smaller estimation errors, indeed, promote
?nancial-reporting quality.
Tests of H2
Our second hypothesis predicts that perspective taking medi-
ates the relationship between role-taking experience and auditors’
assessment of managers’ reporting choices. As part of the
post-experiment questionnaire, participants respond to a state-
ment that elicits perceived perspective taking, where higher values
suggest better self-assessed perspective-taking ability. We ?nd
that the average response of the role-taking group (M = 5.59,
SD = 1.50) is higher than that of the non-role-taking group
(M = 3.66, SD = 1.86), and the difference is statistically signi?cant
using parametric (t(56) = 4.36, p < 0.001, one-tailed) and nonpara-
metric (Z = 3.89, p < 0.001, one-tailed) tests. We investigate
whether participants’ perspective taking mediates the relationship
between auditors’ role-taking experience and the accuracy of their
earnings estimates.
Fig. 4 presents the results of a path analysis. We observe that
when auditors’ perspective taking is included in the model,
role-taking experience loses its statistical signi?cance. Further,
the path from role-taking experience to auditors’ perspective tak-
ing is statistically signi?cant (p < 0.001), as is the path from per-
spective taking to auditors’ estimation error (p = 0.019). We test
for mediation, as suggested by Preacher and Hayes (2008). Using
the bootstrapping macro, we document a signi?cant mediation
Table 2
The effect of role-taking experience on auditors’ accuracy in estimating earnings:
Experiment one.
a
Statistic Role-taking group Non-role-taking group Total
Mean 7.96 10.43 9.19
Median 7.00 9.75 8.63
Std. Deviation 4.47 4.48 4.61
Min 2.25 1.25 1.25
Max 22.75 20.25 22.75
Mean test: t(56) = À2.10, p = 0.02, one-tailed
Median test: Z = À2.54, p = 0.006, one-tailed
a
In experiment one, auditors in the role-taking group have prior experience in
the manager’s role. Accuracy in estimating earnings is measured by the auditor’s
average estimation error, calculated as
P
|Auditor’s Estimate of Earnings À Actual
Earnings|/4.
Fig. 3. Frequency of auditors’ absolute estimation error: Experiment one. Note: The
?gure shows the frequency of participants’ absolute estimation error, computed as
P
|Auditor’s Estimate of Earnings À Actual Earnings|/4, broken down into categories
based on the magnitude of the error I. The ?gure includes all observations from
rounds 3–6. Using Fisher’s exact test, the proportion of absolute estimation errors
that falls within the materiality bounds (i.e., |E| 6 10) is signi?cantly greater for the
role-taking group than the non-role-taking group at p = 0.003, one-tailed.
13
We re-perform the analyses using logistic regression, controlling for round
effects. The dependent variable is auditors’ decision per round: 1 = contributing to
?nancial-reporting quality (or making a correct decision) and 0 = otherwise. The
independent variable is auditors’ absolute estimation error per round. Inferences are
unaffected: that is, smaller estimation errors are associated with better decisions.
46 B.K. Church et al. / Accounting, Organizations and Society 45 (2015) 40–51
effect (p = 0.031, one-tailed), and we ?nd that the 95% bootstrap
con?dence interval for indirect effects excludes zero. Our results
provide support for H2.
To gain further insight, we consider the effect of perspective
taking on auditors’ performance in each experimental group sepa-
rately. We do so because others (Davis, 1980, 1983; Davis et al.,
1996; Davis, Hull, Young, & Warren, 1987) document individual
differences in perspective-taking ability. The analyses presented
above are indicative of differences between the two experimental
groups (role-taking versus non-role-taking). We probe whether
differences also arise within groups. According to our theoretical
development, effective perspective taking improves individuals’
judgments (e.g., Altiero et al., 2014; Davis et al., 1996; Galinsky &
Moskowitz, 2000; Galinsky & Mussweiler, 2001; Galper, 1976;
Johnson, 1967; Johnson & Johnson, 1982; Regan & Totten, 1975;
Sessa, 1996). We examine whether this relationship holds, irre-
spective of participants’ prior experience.
We regress auditors’ average absolute estimation error on their
perceived perspective taking, performing the analysis separately
for each role-taking group. We ?nd that, in both groups, the coef-
?cient of auditors’ perspective taking is negative and statistically
signi?cant: b = À0.83, t(27) = À1.89, p = 0.035, one-tailed, for the
non-role-taking group, and b = À0.95, t(27) = À1.75, p = 0.046,
one-tailed, for the role-taking group.
14
Hence, participants who per-
ceive that they are better at perspective taking, indeed, are more
accurate in estimating earnings. Our ?ndings suggest that individual
differences in perspective taking in?uence auditor behavior, beyond
their role-taking experience.
Additional analyses
Prior ?ndings suggest that individuals are susceptible to
social-projection bias (e.g., Bowlin et al., 2009; Epley et al., 2004;
Rizzolatti & Craighero, 2004; Van Boven et al., 2000, 2005). To
determine whether social-projection bias is a viable explanation
for our ?ndings, we analyze the association between prior experi-
ence and auditors’ judgments. Social-projection bias predicts that,
in the role-taking group, prior experience (over rounds 1–2)
explains auditors’ estimates of earnings and their decisions to
accept/reject reported earnings (over rounds 3–6). More speci?-
cally, social-projection bias suggests that the degree to which par-
ticipants in?ate reported earnings as a manager affects (1) the
extent to which they discount reported earnings as an auditor
and (2) their propensity to reject reported earnings.
For participants in the role-taking group, we regress auditors’
average downward adjustment of earnings (over rounds 3–6) on
their average reporting bias (over rounds 1–2). Auditors’ down-
ward adjustment is de?ned as the difference between managers’
reported earnings and auditors’ estimated earnings. Reporting bias,
on the other hand, is de?ned as the difference between managers’
reported earnings and actual earnings. We ?nd that the coef?cient
of average reporting bias is not signi?cant (b = À0.06,
t(27) = À0.78, p = 0.44). We repeat the analysis using the number
of times that auditors reject managers’ reported earnings as the
dependent variable and again, the coef?cient of average reporting
bias is not signi?cant (b = 0.01, t(27) = 0.51, p = 0.61). Finally, we
re-perform the analyses substituting a dummy variable for average
reporting bias, de?ned as 1 if the average bias (over rounds 1–2) is
greater than 10 (the materiality threshold) and 0 otherwise, and
inferences are unaffected. Hence, we conclude that speci?c prior
experience does not account for auditors’ behavior, and we rule
out social-projection bias as an alternative explanation for our
results.
Discussion
Our results suggest that role-taking experience facilitates audi-
tors’ assessment of managers’ reporting choices (i.e., determining
whether reported earnings are materially misstated). Auditors
with role-taking experience (as a manager) make more accurate
estimates of earnings than auditors without such experience. In
terms of the underlying cognitive mechanism, we ?nd that
role-taking experience stimulates perspective taking, which
enables auditors to more effectively put themselves in the man-
ager’s shoes. In turn, they are better able to judge managers’
reported earnings. We rule out social-projection bias as an alterna-
tive explanation for our ?ndings, and we offer evidence that links
the accuracy of auditors’ earnings estimates with their decision
to accept/reject manager’s reports, thereby promoting
?nancial-reporting quality. Along these lines, Peecher et al.
(2007) provide a framework that characterizes the audit process
as evidence-driven, belief-based, risk assessment. As part of the
audit process, auditors formulate beliefs about reported amounts
and disclosures, which serve as a basis to evaluate the reasonable-
ness of managers’ reporting choices. Auditors who formulate
richer, more accurate beliefs are expected to provide higher quality
services, bene?tting ?nancial-reporting quality (see also Budescu
et al., 2012). Our ?ndings are consistent with this depiction of
the audit process.
Our results also provide evidence of individual differences in
perspective taking. We ?nd that, even in the non-role-taking
group, auditors’ perspective taking is associated with the accuracy
of their earnings estimates. This result suggests that
perspective-taking disposition can signi?cantly impact individuals’
judgments and decision making in strategic interactions.
Table 3
Auditors’ decisions and role-taking experience: Experiment one.
Round Role-taking group Non-role-taking group
Panel A: Percentage of auditors’ decisions that contribute to ?nancial-reporting
quality
a
3 0.59 0.45
4 0.90 0.69
5 0.76 0.55
6 0.41 0.31
Total M = 0.66 M = 0.50
SD = 0.21 SD = 0.27
Panel B: Percentage of auditors’ decisions that correctly accept/reject managers’
report
b
3 0.66 0.52
4 0.72 0.55
5 0.79 0.66
6 0.45 0.41
Total M = 0.66 M = 0.53
SD = 0.18 SD = 0.24
a
The cell entries in Panel A indicate the percentage of time that auditors’ decision
contribute to ?nancial-reporting quality. This result occurs when (1) the auditor
accepts reported earnings that are within ±10 of actual earnings or (2) the auditor
rejects the manager’s report, the auditor’s earnings estimate is acceptable (within
±10 of actual earnings, and the auditor’s estimate is superior to that of the manager
(closer to actual earnings than the manager’s reported amount). The speci?cs that
underlie high-quality ?nancial reporting are depicted in Fig. 2. The total at the
bottom of the Panel indicates the mean and standard deviation of the percentage of
time that auditor-participants’ decisions contribute to ?nancial-reporting quality.
b
The cell entries in Panel B indicate the percentage of time auditors make the
correct decision to accept or reject reported earnings, irrespective of auditors’
estimates of earnings. The auditor’s decision is correct as long as the auditor accepts
reported earnings that are within ±10 of actual earnings and rejects other amounts.
The total at the bottom of the Panel indicates the mean and standard deviation of
the percentage of time that auditor-participants’ decisions are correct.
14
We also put participants into high and low perspective-taking groups, using a
median split, and repeat the regression analysis for each role-taking group. Inferences
are unaffected.
B.K. Church et al. / Accounting, Organizations and Society 45 (2015) 40–51 47
We note that other research has focused on perspective taking
as a trait or personality dimension (Davis, 1980, 1983; Davis
et al. 1996). More speci?cally, Davis (1980, 1983) has developed
a measure to gauge individuals’ tendency to spontaneously take
the psychological point of view of others, focusing on cognitive
perspective taking.
15
The measure is established and designed to
capture individuals’ capacity for non-egocentric thought, which
allows them to anticipate the behavior and reactions of others. By
comparison, the measure we use in our experiment is
self-assessed. One advantage of our measure is that it is context
speci?c: that is, the measure is elicited based on participants’ consid-
eration of our experimental setting. A potential drawback, however,
is that our measure has not been validated and, as such, it may cap-
ture other dimensions of personality beyond the ability to put one-
self in another’s shoes. The fundamental concern is that our
?ndings might be driven by some other facet of personality, rather
than perspective taking. Indeed, prior research documents that
perspective-taking ability is linked to self-con?dence, self-esteem,
and perceived self-ef?cacy (Corcoran & Mallinckrodt, 2000; Davis,
1983). Further, previous studies document that these personality
dimensions are positively associated with task performance
(Bandura, 1993; Berry, 1987; Berry & West, 1993; Stankov, 2000;
Stankov & Lee, 2008; Wood & Bandura, 1989).
In an effort to appraise the validity of our perspective-taking
measure, we recruit 60 Amazon Mechanical Turk participants
and elicit their responses to the Davis measure and our measure
from experiment one. Participants’ responses indicate that the
two measures are signi?cantly correlated, q = 0.77, p < 0.001.
Notwithstanding, we acknowledge that our measure is a poten-
tially noisy proxy for perspective-taking ability.
To shed further light on the effect of perspective taking on
auditor-participants’ task performance, we design a second exper-
iment using the Davis measure, a validated measure of individual
perspective-taking disposition. Our focus is on individual differ-
ences, absent other situational factors that might stimulate per-
spective taking. The aim is to cleanly examine the effects of
perspective taking on task performance: that is, to determine
whether participants’ perspective-taking disposition affects their
judgments in a predictable way. As discussed previously, effective
perspective taking helps individuals to better judge the behavior of
others in social interactions (Davis et al., 1996; Galinsky &
Moskowitz, 2000; Galinsky & Mussweiler, 2001; Galper, 1976;
Johnson, 1967; Johnson & Johnson, 1982; Regan & Totten, 1975;
Sessa, 1996). Accordingly, we expect participants with high
perspective-taking disposition to have better insight into others’
behavior than those with low perspective-taking disposition. Our
second experiment is designed to provide insight into whether
auditor-participants’ disposition to take another’s perspective
impacts their ability to judge managers’ reporting behavior.
Experiment two
Task and procedures
We recruit 39 students from a public university to participate in
our second experiment. The average age of the participants is
21 years, and 19 are male. Almost all participants are undergradu-
ates (92%), and the majority are business or economics majors
(59%). The experimental task is the same as that used in our ?rst
experiment, except that all participants are assigned to the
non-role-taking condition, meaning that everyone is an auditor
for the entirety of the experiment. As a consequence, we introduce
two practice rounds (replacing rounds 1–2 from our earlier exper-
iment) in which the instructions walk participants through two
examples. The examples explain players’ potential actions and
how such actions map into players’ payoffs. After the two practice
rounds, four experimental rounds are conducted. We use the same
managers’ earnings reports (rounds 3–6) as in our ?rst experiment.
After ?nishing the experimental task, participants complete a
post-experiment questionnaire, which includes the Davis (1980)
scale. The scale measures participants’ spontaneous
perspective-taking disposition and is comprised of seven items
that elicit how well various statements describe respondents.
Each statement involves respondents’ cognitive ability to
empathize with others.
16
Participants’ mean score on the Davis
scale is 4.42 (SD = 0.85). Following Davis et al. (1987), we split them
at the median and put them into high and low perspective-taking
groups. The average response is 5.00 (SD = 0.52) for participants in
the high perspective-taking group and 3.69 (SD = 0.61) for those in
the low perspective-taking group.
Role-Taking
Experience
Perspective Taking
Accuracy in
Estimating Earnings
?
1
= 2.47
p = 0.016
?
1
= 1.93
p < 0.001
?
2
= 0.88
p = 0.019
?
1
= 0.78
p = 0.623
Fig. 4. Path analysis results with perspective taking as a mediator: Experiment one. Note: The ?gure shows the path analysis results linking role-taking experience, perceived
perspective taking, and auditors’ accuracy in estimating earnings. The coef?cient for the direct path from role-taking experience to auditor’s accuracy in estimating earnings is
represented by c
1
when the mediator is not included in the path model and d
1
when the mediator is included in the model. Using Preacher and Hayes’ (2008) macro, we
document a signi?cant mediation effect at p = 0.031, one-tailed and we also ?nd that the 95% bootstrap con?dence interval for indirect effects excludes zero.
15
Davis (1980, 1983) develops the Interpersonal Reactivity Index, designed to
measure dispositional empathy. The instrument includes four separate subscales that
delve into separate facets of empathy, including perspective taking, empathic
concern, personal distress, and fantasy. We are interested in the perspective-taking
subscale that deals with cognition and understanding others’ beliefs, behavior, and
actions.
16
The scale has seven items: (1) I sometimes ?nd it dif?cult to see things from the
‘‘other person’s’’ point of view (reverse-coded); (2) I try to look at everybody’s side of
a disagreement before I make a decision; (3) I sometimes try to understand my
friends better by imagining how things look from their perspective; (4) If I’m sure I’m
right about something, I don’t waste much time listening to other people’s arguments
(reverse-coded); (5) I believe that there are two sides to every question and try to look
at them both; (6) When I’m upset at someone, I usually try to ‘‘put myself in his
shoes’’ for a while; and (7) Before criticizing somebody, I try to imagine how I would
feel if I were in their place. For each item, participants respond on a seven-point Likert
scale, anchored by 1 = does not describe me well to 7 = describes me well.
Participants’ average response across the seven items is a measure of their
perspective-taking disposition. Based on participants’ responses, we obtain a
Cronbach’s alpha of 0.88.
48 B.K. Church et al. / Accounting, Organizations and Society 45 (2015) 40–51
Results
We examine whether perspective-taking disposition is associ-
ated with auditors’ evaluation of managers’ earnings reports. We
compute auditors’ average absolute estimation error over the four
rounds. Descriptive statistics are presented in Table 4, partitioned
by perspective-taking group. We ?nd that the mean (median) esti-
mation error of the high perspective-taking group is less than that
of the low perspective-taking group: 7.11 (7.00) versus 9.49 (9.00),
respectively. A parametric t-test and a nonparametric Mann–
Whitney test indicate that the difference between the two groups
is statistically signi?cant at conventional levels: t(37) = À1.69,
p = 0.05, one-tailed and Z = À1.67, p = 0.049, one-tailed, respec-
tively. Our ?nding suggests that participants’ natural
perspective-taking disposition improves task performance.
To further judge the accuracy of auditors’ earnings estimates,
we examine the frequency of participants’ estimation errors over
the four rounds, partitioned into categories based on whether the
magnitude of the error is within ±10 of actual earnings. Fisher’s
exact test indicates that the proportion of absolute estimation
errors that falls within the materiality bounds (i.e., |E| 6 10) is sig-
ni?cantly greater for the high perspective-taking group than the
low perspective-taking group at p = 0.043, one-tailed. This ?nding
provides additional evidence of the effect of perspective-taking
disposition on participants’ performance.
Next, we investigate the effect of perspective-taking disposition
on auditors’ decisions. Panel A of Table 5 presents descriptive data,
summarizing the percentage of time that auditors’ decisions con-
tribute to ?nancial-reporting quality (refer to Fig. 2), partitioned
by perspective-taking group. Examination of the data indicates
that the percentage is larger for the high perspective-taking group
in three of four rounds. However, collapsing the data across the
four rounds, parametric and nonparametric tests indicate that
the difference between the two groups is not statistically signi?-
cant at conventional levels: t(37) = 0.85, p = 0.20, one-tailed, and
Z = 0.73, p = 0.23, one-tailed, respectively. Panel B of Tables 5 pre-
sents descriptive data, summarizing the percentage of time that
auditors’ make the correct accept/reject decision, irrespective of
their earnings estimates. Again, collapsing the data across the four
rounds, we ?nd that difference between the two groups is not sta-
tistically signi?cant: t(37) = 0.03, p = 0.49, one-tailed, and Z = 0.00,
p = 0.50, one-tailed. We suggest that perspective-taking disposition
may have a more subtle effect on auditors’ decisions as compared
to other variables (e.g., role-taking experience) that are designed to
stimulate perspective taking. Our conjecture is consistent with
?ndings in psychology that behavior is in?uenced by both contex-
tual factors and personality, but contextual factors are likely a
stronger predictor of behavior in a speci?c situation, whereas
personality is a more consistent predictor of behavior over time
(Fleeson, 2001; Fleeson & Noftle, 2008; Ross & Nisbett, 1991).
Subsequently, we scrutinize whether participants with more
accurate estimates of earnings promote high-quality ?nancial
reporting. As before, we regress the number of times that auditors’
decisions contribute to ?nancial-reporting quality on their average
absolute estimation error. We ?nd that the coef?cient of the aver-
age estimation error is negative and statistically signi?cant
(b = À0.18, t(37) = À5.17, p < 0.001), which suggests that estima-
tion accuracy is associated with superior decision making. We
repeat the analysis using the number of times that auditors make
the correct accept/reject decision and inferences are unaffected
(b = À0.11, t(37) = À3.41, p = 0.002).
Finally, we perform a path analysis to assess the links between
perspective-taking group, estimation accuracy, and promoting
high-quality ?nancial reporting. We document a signi?cant path
from perspective taking to estimation accuracy (path coef?cient
of À2.37, p = 0.042) and from estimation accuracy to promoting
high-quality ?nancial reporting (path coef?cient of À0.17,
p < 0.001). Results are similar examining the links between per-
spective taking, estimation accuracy, and making correct accep-
t/reject decisions. All in all, the results suggest that dispositional
perspective taking impacts auditor-participants’ assessment of
managers’ reporting choices, aside from the effects of other factors
that might be introduced to stimulate perspective taking.
Conclusion
We report the results of two experiments designed to examine
the effect of perspective taking on auditors’ ability to determine
whether managers’ reported earnings are materially misstated
and, in turn, to contribute to high-quality ?nancial reporting. In
our ?rst experiment, we manipulate auditors’ prior experience in
the manager’s role (experience versus no experience). We predict
Table 4
The effect of perspective-taking disposition on auditors’ accuracy in estimating
earnings: Experiment two.
a
Statistic Perspective-taking group Total
High Low
Mean 7.11 9.49 8.15
Median 7.00 9.00 7.50
Std. Deviation 3.79 5.01 4.46
Min 1.75 2.50 1.75
Max 15.75 21.25 21.25
Mean test: t(37) = À1.69, p = 0.05, one-tailed
Median test: Z = À1.67, p = 0.049, one-tailed
a
In experiment two, participants respond to the Davis scale, which allows us to
measure their perspective-taking disposition. We use a median split to partition
participants into a high and low perspective-taking group. Accuracy in estimating
earnings is measured by the auditor’s average estimation error, calculated as
P
|Auditor’s Estimate of Earnings À Actual Earnings|/4.
Table 5
Auditors’ decisions and perspective-taking disposition: Experiment two.
Round Perspective-taking group
High Low
Panel A: percentage of auditors’ decisions that contribute to ?nancial-reporting
quality
a
3 0.45 0.41
4 0.86 0.53
5 0.77 0.65
6 0.41 0.59
Total M = 0.63 M = 0.54
SD = 0.29 SD = 0.31
Panel B: Percentage of auditors’ decisions that correctly accept/reject managers’
report
b
3 0.45 0.41
4 0.64 0.47
5 0.86 0.82
6 0.41 0.65
Total M = 0.59 M = 0.59
SD = 0.25 SD = 0.25.
a
The cell entries in Panel A indicate the percentage of time that auditors’ decision
contribute to ?nancial-reporting quality. This result occurs when (1) the auditor
accepts reported earnings that are within ±10 of actual earnings or (2) the auditor
rejects the manager’s report, the auditor’s earnings estimate is acceptable (within
±10 of actual earnings, and the auditor’s estimate is superior to that of the manager
(closer to actual earnings than the manager’s reported amount). The speci?cs that
underlie high-quality ?nancial reporting are depicted in Fig. 2. The total at the
bottom of the Panel indicates the mean and standard deviation of the percentage of
time that auditor-participants’ decisions contribute to ?nancial-reporting quality.
b
The cell entries in Panel B indicate the percentage of time auditors make the
correct decision to accept or reject reported earnings, irrespective of auditors’
estimates of earnings. The auditor’s decision is correct as long as the auditor accepts
reported earnings that are within ±10 of actual earnings and rejects other amounts.
The total at the bottom of the Panel indicates the mean and standard deviation of
the percentage of time that auditor-participants’ decisions are correct.
B.K. Church et al. / Accounting, Organizations and Society 45 (2015) 40–51 49
and ?nd that role-taking experience stimulates perspective taking,
which allows auditors to more accurately estimate managers’
reported earnings, ultimately bene?tting ?nancial-reporting qual-
ity. In our second experiment, we appraise auditor-participants’
ability to spontaneously take another’s psychological viewpoint,
beyond the effects of role-taking experience. As expected, we ?nd
that auditors with high perspective-taking disposition are better
able to judge managers’ reported earnings than auditors with
low perspective-taking disposition. On the whole, our ?ndings
highlight the importance of perspective taking as a means to pro-
mote audit quality.
We recognize that our study is subject to various limitations.
Our experimental setting abstracts away from a speci?c audit con-
text. However, the use of an abstract setting allows us to carefully
focus on the underlying constructs and, importantly, to directly
test our theoretical arguments. Indeed, our experimental ?ndings
support the underlying theory. In designing the experimental task,
we sought to incorporate the key features of auditor–manager
interactions, speci?cally both parties’ situational incentives.
Importantly, we are not aware of any audit content factor that
would interact with role-taking experience or perspective-taking
ability in such a way as to alter our ?ndings. We leave this issue
for future study.
In addition, our study focuses on the cognitive aspects of per-
spective taking – mentally understanding another’s point of view.
But, other aspects of perspective taking can come into play in
strategic interactions. For example, affective empathy may be
important in gauging others’ behavior (e.g., Batson, 2009; Davis,
1983; Davis et al., 1987). Auditors may bene?t from sensitivity to
their counterpart’s feelings, including situational stress and dis-
comfort that are triggered by the reporting environment, which
impact managers’ actions. In our experimental setting, the affective
aspects of perspective taking are suppressed (to avoid potential
confounds). Our experimental task is calculative and game-like,
and it precludes social interaction (e.g., auditors observe managers’
reports and nothing else). Accordingly, our study does not shed
light on the associations between affective empathy, perspective
taking, and auditor performance. We readily acknowledge that
the emotional aspects of perspective taking are important, and
we encourage others to study this area in the future.
Despite the limitations, our results directly point to the bene?ts
of enhanced perspective taking in promoting audit quality. Our
?ndings suggest that audit ?rms can bene?t from hiring individu-
als who have been on the corporate side of the ?nancial-reporting
process, including boomerangs. Audit ?rms also may be able to
capture the bene?ts of role-taking experience by implementing
training programs that include role taking. In addition, our ?ndings
suggest that audit ?rms may want to consider dispositional per-
spective taking in staf?ng assignments. The importance of includ-
ing some team members with high perspective taking likely is
magni?ed when the auditor and client are involved in resolving
signi?cant disagreements over accounting matters. Finally, we
encourage audit ?rms to consider other ways to facilitate perspec-
tive taking as a means to improve auditor performance.
References
Altiero, E., Kang, Y. J., & Peecher, M. E. (2014). The investor perspective and its
in?uence on auditor materiality judgments. Working paper. University of
Illinois.
Anderson, C. M., & Camerer, C. F. (2000). Experience-weighted attraction learning in
sender–receiver signaling games. Economic Theory, 16(3), 689–718.
Badal, J. (2006). On the job: More ’boomerangs’ return to old jobs. The Wall Street
Journal 15 October.
Bandura, A. (1993). Perceived self-ef?cacy in cognitive development and
functioning. Educational Psychologist, 28(2), 117–148.
Batson, C. D. (2009). These things called empathy: Eight related but distinct
phenomena. In J. Decety & W. Ickes (Eds.), The social neuroscience of empathy
(pp. 3–15). Cambridge, MA: MIT Press.
Bazinger, C., & Kühberger, A. (2012). Is social projection based on simulation or
theory? Why new methods are needed for differentiating. New Ideas in
Psychology, 30(3), 328–335.
Berry, J. M. (1987). A self-ef?cacy model of memory performance. In Paper presented
at the meeting of the American Psychological Association, New York.
Berry, J. M., & West, R. L. (1993). Cognitive self-ef?cacy in relation to personal
mastery and goal setting across the life span. International Journal of Behavioral
Development, 16(2), 351–379.
Biddle, B. J. (1986). Recent development in role theory. Annual Review of Sociology,
12, 67–92.
Bloom?eld, R. J. (1995). Strategic dependence and inherent risk assessments. The
Accounting Review, 70(1), 71–90.
Bonner, S., & Lewis, B. L. (1990). Determinants of auditor expertise. Journal of
Accounting Research, 28(3), 1–20.
Bowlin, K. (2011). Risk-based auditing, strategic prompts, and auditor
sensitivity to the strategic risk of fraud. The Accounting Review, 86(4),
1231–1253.
Bowlin, K. O., Hales, J., & Kachelmeier, S. J. (2009). Experimental evidence of how
prior experience as an auditor in?uences managers’ strategic decisions. Review
of Accounting Studies, 14(1), 63–87.
Brandts, J., & Holt, C. H. (1992). An experimental test of equilibrium dominance in
signaling games. American Economic Review, 82(5), 1350–1365.
Brandts, J., & Holt, C. H. (1993). Adjustment patterns and equilibrium selection in
experimental signaling games. International Journal of Game Theory, 22(3),
279–302.
Budescu, D. V., Peecher, M. E., & Solomon, I. (2012). The joint in?uence of the extent
and nature of audit evidence, materiality thresholds, and misstatement type on
achieved audit risk. Auditing: A Journal of Practice and Theory, 31(2), 19–41.
Burns, S. M., & Brainerd, C. J. (1979). Effects of constructive and dramatic play on
perspective taking in very young children. Developmental Psychology, 15(5),
512–521.
Chalmers, J. B., & Townsend, M. A. (1990). The effects of training in social
perspective taking on socially maladjusted girls. Child Development, 61(1),
178–190.
Chandler, M. J. (1973). Egocentrism and antisocial behavior: The assessment and
training of social perspective-taking skills. Developmental Psychology, 9(3),
326–332.
Colman, A. M. (2003). Depth of strategic reasoning in games. Trends in Cognitive
Sciences, 7(1), 2–4.
Cooper, D. J., Garvin, S., & Kagel, J. H. (1997a). Adaptive learning vs. equilibrium
re?nements in an entry limit pricing game. The Economic Journal, 107(442),
553–575.
Cooper, D. J., Garvin, S., & Kagel, J. H. (1997b). Signaling and adaptive learning in an
entry limit pricing game. RAND Journal of Economics, 28(4), 662–683.
Corcoran, K. O., & Mallinckrodt, B. (2000). Adult attachment, self-ef?cacy,
perspective taking, and con?ict resolution. Journal of Counseling and
Development, 78(4), 473–483.
Daneman, M., & Carpenter, P. A. (1980). Individual differences in working
memory and reading. Journal of Verbal Learning and Verbal Behavior, 19(4),
450–466.
Davis, M. H. (1980). A multidimensional approach to individual differences in
empathy. JSAS Catalog of Selected Documents in Psychology, 10, 85–104.
Davis, M. H. (1983). Measuring individual differences in empathy: Evidence for a
multidimensional approach. Journal of Personality and Social Psychology, 44(1),
113–126.
Davis, M. H., Conklin, L., Smith, A., & Luce, C. (1996). Effect of perspective taking on
the cognitive representation of persons: A merging of self and other. Journal of
Personality and Social Psychology, 70(4), 713–726.
Davis, M. H., Hull, J. G., Young, R. D., & Warren, G. G. (1987). Emotional reactions to
dramatic ?lm stimuli: The in?uence of cognitive and emotional empathy.
Journal of Personality and Social Psychology, 52(1), 126–133.
Deloitte (2011). Good riddance or good business – How should companies manage
alumni relationships? Deloitte debates. .
Dulas, M. R., & Duarte, A. (2013). The in?uence of directed attention at encoding on
source memory retrieval in the young and old: An ERP study. Brain Research,
1500, 55–71.
Epley, N., Caruso, E. M., & Bazerman, M. H. (2006). When perspective taking
increases taking: Reactive egoism in social interaction. Journal of Personality and
Social Psychology, 91(5), 872–889.
Epley, N., & Caruso, E. M. (2009). Perspective taking: Misstepping into others’ shoes.
In K. D. Markman, W. M. P. Klein, & J. A. Suhr (Eds.), Handbook of imagination and
mental simulation. New York, NY: Psychology Press.
Epley, N., Keysar, B., Boven, L. F., & Gilovich, T. (2004). Perspective taking as
egocentric anchoring and adjustment. Journal of Personality and Social
Psychology, 87(3), 327–339.
Feffer, M. H. (1959). The cognitive implications of role taking behavior. Journal of
Personality, 27(2), 152–168.
Fiske, S. T., Taylor, S. E., Etcoff, N. L., & Laufer, J. K. (1979). Imaging, empathy, and
causal attribution. Journal of Experimental Social Psychology, 15(4), 356–377.
Fleeson, W. (2001). Towards a structure- and process-integrated view of
personality: Traits as density distributions of states. Journal of Personality and
Social Psychology, 80(6), 1011–1027.
Fleeson, W., & Noftle, E. E. (2008). Where does personality have its in?uence? A
super-matrix of consistency concepts. Journal of Personality, 76(6), 1355–1386.
50 B.K. Church et al. / Accounting, Organizations and Society 45 (2015) 40–51
Galinsky, A. D., & Moskowitz, G. B. (2000). Perspective-taking: Decreasing
stereotype expression, stereotype accessibility, and in-group favoritism.
Journal of Personality and Social Psychology, 78(4), 708–724.
Galinsky, A. D., & Mussweiler, T. (2001). First offers as anchors: The role of
perspective-taking and negotiator focus. Journal of Personality and Social
Psychology, 81(4), 657–669.
Galper, R. E. (1976). Turning observers into actors: Differential causal attributions
as a function of ‘‘empathy’’. Journal of Research in Personality, 10(3), 328–
335.
Hedden, T., & Zhang, J. (2002). What do you think I think you think?: Strategic
reasoning in matrix games. Cognition, 85(1), 1–36.
Heider, F. (1958). The psychology of interpersonal relations. New York, NY: Wiley.
Hidi, S. (1995). A reexamination of the role of attention in learning from text.
Educational Psychology Review, 7(4), 323–350.
Hoffman, V. B., & Zimbelman, M. F. (2009). Do strategic reasoning and
brainstorming help auditors change their standard audit procedures in
response to fraud risk? The Accounting Review, 84(3), 811–837.
Holmes, D. S. (1968). Dimensions of projection. Psychological Bulletin, 69(4),
248–268.
Hurtt, R. K. (2010). Development of a scale to measure professional skepticism.
Auditing: A Journal of Practice & Theory, 29(1), 149–171.
Hyland, T. (2006). ‘Alumni’ programs aim to bring back former employees.
Philadelphia Business Journal (April 3, 2006).
Iannotti, R. J. (1978). Effect of role-taking experiences on role taking, empathy,
altruism, and aggression. Developmental Psychology, 14(2), 119–124.
James, W. (1890). The principles of psychology. New York: Holt.
Johnson, D. W. (1967). The use of role reversal in intergroup competition. Journal of
Personality and Social Psychology, 7(2), 135–141.
Johnson, D. W., & Johnson, R. T. (1982). Joining together: Group theory and group
skills. Englewood Cliffs, NJ: Prentice Hall.
Kahneman, D. (1973). Attention and effort. Englewood Cliffs, NJ: Prentice Hall.
Kahneman, D., Knetsch, J. L., & Thaler, R. H. (1991). Anomalies: The endowment
effect, loss aversion, and status quo bias. The Journal of Economic Perspectives,
5(1), 193–206.
King, R. R. (2002). An experimental investigation of self-serving biases in an
auditing trust game: The effect of group af?liation. The Accounting Review, 77(2),
265–284.
Krueger, J., & Clement, R. W. (1994). The truly false consensus effect: An
ineradicable and egocentric bias in social perception. Journal of Personality
and Social Psychology, 67(4), 596–610.
Kübler, D., Müller, W., & Normann, H.-T. (2008). Job-market signaling and
screening: An experimental comparison. Games and Economic Behavior, 64(1),
219–236.
Magilke, M., Mayhew, B., & Pike, J. (2009). Are independent audit committee
members objective? The Accounting Review, 84(6), 1959–1981.
Majors, T. M., Shefchik, L. B., & Vitalis, A. M. (2014). The interactive effect of ego
depletion and trait skepticism on auditor performance. Working paper.
University of Illinois at Urbana-Champaign, Indiana University, and Georgia
Tech.
Mead, G. H. (1934). Mind, self, and society. Chicago: University of Chicago Press.
Moreno, J. L. (1934). Who shall survive?: A new approach to the problem of human
interrelations. Nervous and mental disease monograph series (Vol. 58, pp. 22–66).
Washington, DC: Nervous and Mental Disease Publishing Co.
Moreno, J. L. (1946). Role theory and role practice. Psychodrama (Vol. 1). Beacon, NY:
Beacon House.
Musso, M., Kyndt, E., Cascallar, E., & Dochy, F. (2012). Predicting mathematical
performance: The effect of cognitive processes and self-regulation factors.
Education Research International, 1–13.
Niedenthal, P. M., Barsalou, L. W., Winkielman, P., Krauth-Gruber, S., & Ric, F. (2005).
Embodiment in attitudes, social perception, and emotion. Personality and Social
Psychology Review, 9(3), 184–211.
Oberauer, K. (2002). Access to information in working memory: Exploring the focus
of attention. Journal of Experimental Psychology: Learning, Memory, and Cognition,
28(3), 411–421.
Oberauer, K. (2009). Design for a working memory. Psychology of Learning and
Motivation, 51, 45–100.
Peecher, M. E., Schwartz, R., & Solomon, I. (2007). It’s all about audit quality:
Perspectives on strategic-systems auditing. Accounting, Organizations and
Society, 32(4–5), 463–485.
Perner, J., & Wimmer, H. (1985). John thinks that Mary thinks that. . . Attribution of
second-order beliefs by 5- to 10-year-old children. Journal of Experimental Child
Psychology, 39(3), 437–471.
Piaget, J. (1932). The moral judgment of the child. London: Kegan, Paul, Trench,
Trubner.
Piaget, J. (1950). The psychology of intelligence. London: Routledge and Kegan Paul.
Posner, M. I., & Petersen, S. E. (1990). The attention system of the human brain.
Annual Review of Neuroscience, 13, 25–42.
Potters, J., & van Winden, F. (1996). Comparative statics of a signaling game:
An experimental study. International Journal of Game Theory, 25(1), 329–
353.
Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling strategies for
assessing and comparing indirect effects in multiple mediator models. Behavior
Research Methods, 40(3), 879–891.
Regan, D. T., & Totten, J. (1975). Empathy and attribution: Turning observers into
actors. Journal of Personality and Social Psychology, 32(5), 850–856.
Rerko, L., & Oberauer, K. (2013). Focused, unfocused, and defocused information in
working memory. Journal of Experimental Psychology: Learning, Memory, and
Cognition, 39(4), 1075–1096.
Rizzolatti, G., & Craighero, L. (2004). The mirror-neuron system. Annual Review of
Neuroscience, 27, 169–192.
Rose, J. M. (2007). Attention to evidence of aggressive ?nancial reporting and
intentional misstatement judgments: Effects of experience and trust. Behavioral
Research in Accounting, 19, 215–229.
Ross, L., & Nisbett, R. E. (1991). The person and the situation: Perspectives of social
psychology. New York, NY: McGraw-Hill.
Sco?eld, S. B., Phillips, T. J., Jr., & Bailey, C. D. (2004). An empirical reanalysis of the
selection-socialization hypothesis: A research note. Accounting, Organizations
and Society, 29(5–6), 543–563.
Sessa, V. I. (1996). Using perspective taking to manage con?ict and affect in teams.
The Journal of Applied Behavioral Science, 321(1), 101–115.
Smith, P. L., & Sewell, D. K. (2013). A competitive interaction theory of attentional
selection and decision making in brief, multielement displays. Psychological
Review, 120(3), 589–627.
Sogunro, O. A. (2004). Ef?cacy of role-playing pedagogy in training leaders: Some
re?ections. Journal of Management Development, 23(4), 355–371.
Stankov, L. (2000). Complexity, metacognition, and ?uid intelligence. Intelligence,
28(2), 121–143.
Stankov, L., & Lee, J. (2008). Con?dence and cognitive test performance. Journal of
Educational Psychology, 100(4), 961–976.
Staub, E. (1971). The use of role playing and induction in children’s learning of
helping and sharing behavior. Child Development, 42(3), 805–816.
Trotman, K. T., Wright, A., & Wright, S. (2005). Auditor negotiations: An
examination of the ef?cacy of intervention methods. The Accounting Review,
80(1), 349–367.
Turner, R. (2006). Role theory. In J. H. Turner (Ed.), Handbook of Sociological Theory.
New York, NY: Springer.
Van Boven, L. F., Dunning, D., & Loewenstein, G. (2000). Egocentric empathy gaps
between owners and buyers: Misperceptions of the endowment effect. Journal
of Personality and Social Psychology, 7(1), 66–76.
Van Boven, L. F., Loewenstein, G., & Dunning, D. (2005). The illusion of courage in
social predictions: Underestimating the impact of fear of embarrassment in
other people. Organizational Behavior and Human Decision Processes, 96(2),
130–141.
West, P. M. (1996). Predicting preferences: An examination of agent learning.
Journal of Consumer Research, 23(1), 68–80.
Wilson, B. J., Will, M. N., Schoen?eld-McNeill, J., & Montague, R. (2013). The
attention skills of boys with and without developmental delays: Associations
with academic performance. Early Child Development & Care, 183(5), 651–666.
Wood, R., & Bandura, A. (1989). Impact of conceptions of ability on self-regulatory
mechanisms and complex decision making. Journal of Personality and Social
Psychology, 56(3), 407–415.
Wyer, R. S., Srull, T. K., Gordon, S. E., & Hartwick, J. (1982). Effects of processing
objectives on the recall of prose material. Journal of Personality and Social
Psychology, 43(4), 674–688.
Zimbelman, M. F., & Waller, W. S. (1999). An experimental investigation of auditor–
auditee interaction under ambiguity. Journal of Accounting Research, 37(3),
135–156.
B.K. Church et al. / Accounting, Organizations and Society 45 (2015) 40–51 51
doc_453482365.pdf