International Journal of Accounting Information Systems 13 (2012) 248–262

Contents lists available at SciVerse ScienceDirect

International Journal of Accounting

Information Systems

The antecedents of the use of continuous auditing in the

internal auditing context
George C. Gonzalez a, b,⁎, Pratyush N. Sharma b, 1, Dennis F. Galletta b, 2
a
University of Lethbridge, 615 Macleod Trail SE, Calgary, AB T2G 4T8, Canada
b
Joseph M. Katz Graduate School of Business, University of Pittsburgh, Roberto Clemente Drive, Pittsburgh, PA 15260, United States

a r t i c l e i n f o a b s t r a c t

Article history: The concept of continuous auditing originated over two decades ago.
Received 1 June 2011 Yet despite its much touted benefits, its acceptance and use in practice
Accepted 1 June 2012 has been slow. To gain insight into the state of affairs, we surveyed 210
internal auditors worldwide on the status of their use of continuous
Keywords: auditing. Using the Unified Theory of Acceptance and Use of Technology
Continuous auditing
(UTAUT) we explore the antecedents of internal auditors' intentions to
Continuous assurance
use continuous auditing technology. Employing the Partial Least
Internal auditing
UTAUT
Squares method, we find strong support for the model with an R2 of
Technology use 44.3%. Specifically, we find that internal auditors' perceptions of effort
expectancy and social influence are significant predictors of their
intentions to use continuous auditing. We also find that annual sales
volume of the company and voluntariness of use significantly moderate
the relationship between performance expectancy and social influence
respectively. Additionally, we find regional differences in the signifi-
cance of key UTAUT antecedents. Specifically, we find that the North
American internal auditors are more likely to use continuous auditing
due to soft social coercion pressures of Social Influence through peers
and higher authorities. On the other hand, Middle Eastern auditors are
more likely to use the technology if it is mandated by the higher
authorities.
© 2012 Elsevier Inc. All rights reserved.

⁎ Corresponding author at: University of Lethbridge, 615 Macleod Trail SE, Calgary, Canada AB T2G 4T8. Tel.: +1 403 332 4680.
E-mail address: george.gonzalez@uleth.ca (G.C. Gonzalez).
1
University of Pittsburgh, 229 Mervis Hall, Pittsburgh, PA 15260, United States.
2
University of Pittsburgh, 342 Mervis Hall, Pittsburgh, PA 15260, United States.

1467-0895/$ – see front matter © 2012 Elsevier Inc. All rights reserved.
doi:10.1016/j.accinf.2012.06.009
 G.C. Gonzalez et al. / International Journal of Accounting Information Systems 13 (2012) 248–262 249

1. Introduction

Continuous auditing3 has been touted as offering many important benefits to organizations. Among those
benefits are the minimization of accounting errors, more timely analysis and organizational communication,
and increased audit efficiency and effectiveness. Various research studies have explained the benefits of
continuous auditing (Vasarhelyi et al., 2004; Kuhn and Sutton, 2006), discussed technical aspects of
implementing continuous auditing technology (Kuhn and Sutton, 2010), explored actual implementations in
practice (Hermanson et al., 2006), and examined the psychological effects of continuous auditing on managers
(Hunton et al., 2008, 2010).4 Yet while the concept of continuous auditing, first introduced by Groomer and
Murthy (1989) and Vasarhelyi and Halper (1991), is about two decades old, the actual practice of continuous
auditing has remained the exception rather than the rule (Alles et al., 2008; Chan and Vasarhelyi, 2011). This
puzzling lag in the actual use of continuous auditing is the primary motivation for this study.
Use of continuous auditing technology has so far been almost exclusively limited to the internal audit
function (Chan and Vasarhelyi, 2011). Since 2005 some of the top international accounting firms have surveyed
their clients' Chief Audit Executives (CAEs) and other top internal audit officers to gain an understanding of their
continuous auditing practices (PricewaterhouseCoopers (PwC), 2006,2007; KPMG, 2010; KPMG International
(KPMG), 2010; Grant Thornton, 2011). The results of these surveys vary in terms of how extensively continuous
auditing was being used in practice. For example, one survey showed that of the surveying accounting firm's
clients, 13% had a continuous auditing system that was fully operational and 37% had a system in place but not
yet fully developed (PwC, 2006). Another firm's survey indicated figures of 7% and 13%, respectively (KPMG,
2010). A consistent theme among respondents of these surveys over the years, however, is the uniform
apparent optimism of the respondents, followed each time by minimal usage of these technologies in the
following survey: regardless of the level to which continuous auditing had been used, the survey respondents
expected a considerably higher level of use in two years' time. Even in the face of repeated optimism expressed
by survey respondents, it appears that the implementation of continuous auditing has actually advanced very
slowly. This again raises the question of why there seems to be a lag in the use of continuous auditing.
Given the potential benefits of continuous auditing, some of which were previously mentioned (also see:
Debreceny et al., 2005; Flowerday and von Solms, 2005; Kogan et al., 1999; Rezaee et al., 2002; Vasarhelyi et al.,
2002), this lag is puzzling. We thus set out to gain insight into this state of affairs by conducting an online survey
of industry practitioners' internal audit practices and analyzing their responses through the lens of the Unified
Theory of Acceptance and Use of Technology (UTAUT) from the Management Information Systems (MIS)
discipline (Venkatesh et al., 2003). UTAUT provides a theoretical framework upon which to assess the use of a
particular type of technology. In this study, we find that the UTAUT model explains a substantial amount of
variance in the intentions to use continuous auditing. Most importantly, we find that perceptions of effort
expectancy and social influence are significant predictors of internal auditors' intentions to use, while
performance expectancy and facilitating conditions are not. Annual sales volume of the company and voluntariness
of use significantly moderate the relationship between performance expectancy and social influence respectively.
Our paper is organized as follows. In the next section we briefly present an introduction to the UTAUT
framework. Subsequent sections present our research model and hypotheses, discuss our data collection
method, and outline the data analysis procedure and the results of model testing. Finally we conclude with
a discussion of our findings and implications for practice.

2. Theoretical framework

Many MIS researchers have studied acceptance of new technologies over the past two decades. The Theory of
Reasoned Action (TRA) (Fishbein and Ajzen, 1975) and the Theory of Planned Behavior (TPB) (Ajzen, 1985,
1987, 1991) have greatly informed work in determining behavioral intentions. TRA states that antecedents of
behavioral intentions are attitudes and subjective norms. TPB added perceived behavioral control to the two
antecedents of TRA and also added a direct relationship between perceived behavioral control and actual

3
The focus of a particular continuous auditing technique can range from controls-based (“continuous controls assessment”) to
risk-based (“continuous risk assessment”) (IIA, 2005). Our use of the term “continuous auditing” is intended to encompass any
techniques along this spectrum.
4
For a review of the literature see Brown et al. (2007).
250 G.C. Gonzalez et al. / International Journal of Accounting Information Systems 13 (2012) 248–262

behavior. In her study of external audit support system use, Dowling (2009) develops a model by combining
principles of TPB and Adaptive Structuration Theory (DeSanctis and Poole, 1994) to investigate external auditor
use of firm audit support systems.
Davis (1989) adapted TRA and TPB to the MIS literature to create the widely-cited Technology Acceptance
Model (TAM) (Davis, 1989), which had dramatic influence on the MIS field. TAM, which focuses on
technology acceptance, provides usefulness and ease of use as antecedents to behavioral intentions. Other
versions of TAM have made slight changes to the constructs over the years, but the basic principles have
remained the same. A few technology acceptance studies in the accounting literature have used TAM. One
examined the effects of training on user acceptance of electronic work papers (Bedard et al., 2003). Another
study explored differing attitudes towards technology concluded that social studies students perceived
technology to be less useful and not as easy to use as did business students (Greenfield and Rohde, 2009). A
third study, more closely related to the current one, examined how groups of features of Generalized Audit
Software (GAS) affect internal auditors' technology acceptance behavior, finding that perceived usefulness
has more impact on the usage of basic GAS features than perceived ease of use and, conversely, perceived ease
of use has more impact on the usage of advanced GAS features than perceived usefulness (Kim et al., 2009).
After over a decade of research on TAM, Davis and colleagues proposed the Unified Theory of Acceptance
and Use of Technology (UTAUT) (Venkatesh et al., 2003), which has taken the place of the various TAM
models (see Fig. 1). UTAUT has been used in many contexts to predict behavioral intentions. It served as the
model for examining factors affecting an in-charge external auditor's decision to implement new technology
on an engagement (Curtis and Payne, 2008; Dowling, 2008). The basic notion underlying UTAUT is that three
antecedents will predict behavioral intentions: performance expectancy (formerly perceived usefulness), effort
expectancy (formerly perceived ease-of-use), and social influence (not in the original TAM model). A direct
antecedent of actual behavior is facilitating conditions. Finally, control variables moderate the relationships of
the four antecedents of intentions: gender, age, experience, and voluntariness of use.
TAM and UTAUT have been shown to enrich our understanding of computer-related use behaviors in
many contexts. UTAUT is therefore employed in this study to understand use of continuous auditing tools. If
there is a shortfall in use of such tools, it is important to determine to what extent the various antecedents
play a role. For instance, if the problem is the lack of positive perceptions of effort expectancy among internal
auditors, training programs might be useful to reduce the usability barriers. Alternatively, system designers
may need to develop more user friendly continuous auditing systems that auditors would prefer to use. If, on
the other hand, the problem is the lack of positive perceptions of performance expectancy, demonstrations and
performance statistics could be useful. Therefore, understanding any lack of positive perceptions among
internal auditors concerning the key UTAUT antecedents should allow the firms to take concrete steps in
managing these perceptions and hence encouraging the use of continuous auditing technology.
Since the TAM and UTAUT models were proposed to explain individual level use of technology, it is
reasonable to assume that the intentions of individuals embedded within organizations to use continuous
auditing technologies also depend on most of the same constructs. Executives would need to be cognizant
of perceptions of effort expectancy as well as perceptions of performance expectancy in deciding to
encourage the use of the technology. Facilitating infrastructure needs to be in place as well. Further,

Performance
Expectancy

Effort
Behavioral Use
Expectancy Intention Behavior

Social
Influence

Facilitating
Conditions

Gender Age Experience Voluntariness of use

Fig. 1. UTAUT model (Venkatesh et al., 2003).

 G.C. Gonzalez et al. / International Journal of Accounting Information Systems 13 (2012) 248–262 251

internal auditors even have social pressures from both within and outside of their firms. Internal pressure
points would include those peers and supervisors, as UTAUT specifies. External pressures are likely to
make this variable even more important in the case of continuous auditing; the widespread use of these
technologies by other firms, and the promotion of the technology by influential professional groups such
as the Institute of Internal Auditors (IIA) and the Institute of Management Accountants (IMA), can be
naturally considered to be a benchmark that they should follow so they do not fall behind. Not following
emerging best practices could even create exposure for a firm, such as the failure to detect fraud as a result
of not implementing advanced technologies such as continuous auditing.
While we select the key antecedents in our model based on UTAUT, we believe that the setting of our study
requires a unique set of moderating and control variables. Because we focus on internal auditors embedded in
firms, we need some mechanism to account for differences among the firms that may affect the perceptions of
internal auditors regarding the key UTAUT antecedents and their effect on intentions to use the technology.
For example, firms with higher sales may have access to better infrastructure and/or training mechanisms to
promote the use of continuous auditing among their internal auditors. Therefore we chose annual sales as a
moderator variable in our model to account for the firms' abilities to acquire the technology and promote its
use among their employees. We also use a voluntariness to use as a moderator variable that is identical to the
one in UTAUT, because it accounts directly for situations in which the use of such technologies is dictated by
another party.

3. Research model and hypotheses

As per the discussion above, we expect six constructs to play a significant role in internal auditors'
intentions to use continuous auditing technology: effort expectancy, performance expectancy, facilitating
conditions, social influence, voluntariness of use and annual sales of the company. In the remainder of this
section, we define each of the determinants, specify the role of key moderators (voluntariness and annual
sales), and provide the theoretical justification for the hypotheses. Fig. 2 presents our research model.

3.1. Effort expectancy

Effort expectancy is defined as the degree of ease associated with the use of the system (Venkatesh et al.,
2003). With the advent of continuous auditing systems internal auditors can expect a change in the nature of
their work responsibilities from a traditionally reactive approach to a proactive approach (Chan and Vasarhelyi,
2011). Audit procedures used for transaction and compliance verification are automated in the continuous
auditing environment. The automation of transaction and compliance audit procedures shift the auditor's work
to more complex audit objectives, such as dealing with verifications of estimates, adherence to standards, and
other items that require auditor judgment. Hence, the auditor's main role in the present continuous auditing
environment involves investigating irregularities/exceptions identified by the continuous auditing system and

Effort
Expectancy
H1

Performance
Expectancy H2
Intention
to Use
Facilitating H3
Conditions
H4 H5d H6

Social H5c
Influence H5b
H5a

Sales Voluntary

Fig. 2. Research model.

252 G.C. Gonzalez et al. / International Journal of Accounting Information Systems 13 (2012) 248–262

dealing with audit procedures requiring judgment (Chan and Vasarhelyi, 2011). The more seamlessly and
effortlessly the auditors are able to transition into using the continuous auditing systems in this new role, the
higher will be their intentions to use them. Additionally, continuous auditing systems employ (or will employ)
enabling technologies including statistical methodologies such as belief functions and neural networks, as well
as technologies from computer science such as database and expert systems, intelligent agents, and
technologies for tagging data to facilitate transmission and comparison, most notably XBRL and XBRLGL
(Vasarhelyi et al., 2004). Internal auditors will not only have to overcome the learning curve to become skillful
at using continuous auditing systems, but also to find them easy and efficient to use. Efficient use is especially
relevant because auditors will be interacting frequently with the continuous auditing system. Given the
complexities involved, the use of continuous auditing systems will therefore be facilitated by the positive
perceptions among internal auditors with regard to effort expectancy. Therefore we hypothesize:

H1. Positive perceptions of effort expectancy will increase internal auditors' intentions to use continuous
auditing technology.

3.2. Performance expectancy

Performance expectancy is defined as the degree to which an individual believes that using a system will
help achieve gains in job performance (Venkatesh et al., 2003). Continuous auditing is likely to have many
benefits such as continuous error and fraud detection, and the use of data analytics and data modeling
features (Vasarhelyi et al., 2004), all of which lead to an enhanced internal control system. Given these
benefits, use of continuous auditing will be facilitated by internal auditors' perceptions of usefulness of the
system in their work and the productivity gains they can expect from it. Therefore, we argue that to the extent
that internal auditors perceive continuous auditing as being better than using its precursor, traditional
periodic auditing, they are likely to have positive intentions to use continuous auditing technology.

H2. Positive perceptions of performance expectancy will increase internal auditors' intentions to use
continuous auditing technology.

3.3. Facilitating conditions

Facilitating conditions are defined as the degree to which an individual believes that an organizational
and technical infrastructure exists to support use of the system. These conditions include aspects of the
technological and/or organizational environment that are designed to remove barriers to the use of a
system (Venkatesh et al., 2003). This relates to technical, monetary and training support and the resources
available to the internal auditors in facilitating their use of the continuous auditing system. As mentioned
earlier, the use of continuous auditing systems is likely to involve learning of enabling technologies and
internal auditors who either possess the background knowledge or have access to the resources required
to learn will have more positive perceptions of facilitating conditions. The continuous auditing systems also
have to be compatible with systems auditors are already using.
In the original UTAUT model, facilitating conditions are hypothesized only to impact actual use, not
intentions. In studies that do not examine actual behavior, it is perhaps worthwhile to include facilitating
conditions at least for exploratory purposes. In our study, including facilitating conditions is not purely
exploratory, as the users in this sample are professionals and are more likely to be at least somewhat aware of
those conditions in advance than clerical staff employees. Therefore, in our next hypothesis we propose:

H3. Positive perceptions of facilitating conditions will increase internal auditors' intentions to use
continuous auditing technology.

3.4. Social influence

Social influence is defined as the degree to which an individual perceives that people important to him or
her believe he or she should use the system. Social influence as a direct determinant of behavioral intention is
represented as subjective norm, i.e., the explicit or implicit notion that the individual's behavior is influenced
 G.C. Gonzalez et al. / International Journal of Accounting Information Systems 13 (2012) 248–262 253

by the way in which they believe others will view them as a result of having used the technology (Venkatesh et
al., 2003). Internal auditors' perception of social influence originates from their peers and from their superiors
in the higher management. It is up to the internal auditors' superiors to, among other things, commit to the use
of continuous auditing technology, obtain agreement from all affected influential parties, and secure the
organizational approval of funds needed to implement the technology. We postulate in this hypothesis that:

H4. Positive perceptions of social influence will increase internal auditors' intentions to use continuous
auditing technology.

3.5. Moderating variables

UTAUT calls for several moderators that are dependent on the particular adopter of the technology. Because
this is not personal technology, we adapted the use of the model by replacing Gender, Age and Experience with
Annual Sales. Annual sales is likely to be a much better moderator variable, given that the gender, age and
experience of the potential user embedded in an organization are less likely to impact intentions to use the
technology. Therefore, the annual sales of the organization is considered to be a more appropriate moderating
factor. As mentioned previously, organizations with higher sales may have access to better infrastructure and/or
training mechanisms to promote the use of continuous auditing among their internal auditors For instance, if effort
expectancy (usability) or performance expectancy (usefulness) is considered to be a potential problem, organi-
zations with more resources would be better able to afford training or customization to handle this problem.
The UTAUT model includes those moderators between the antecedents and behavioral intention only
for certain paths. UTAUT predicts that for performance expectancy, effort expectancy, social influence, and
facilitating conditions, gender only moderates the first three, age moderates all, and experience moderates
only the last three on behavioral intentions. Because we replaced gender, age, and experience with annual
sales, we hypothesize that annual sales will moderate all of those relationships.

H5a. Annual sales will moderate the relationship between effort expectancy and internal auditors'
intentions to use the continuous auditing technology.

H5b. Annual sales will moderate the relationship between performance expectancy and internal
auditors' intentions to use the continuous auditing technology.

H5c. Annual sales will moderate the relationship between facilitating conditions and internal auditors'
intentions to use the continuous auditing technology.

H5d. Annual sales will moderate the relationship between social influence and internal auditors'
intentions to use the continuous auditing technology.
The other moderating variable in UTAUT and our model is voluntariness of use. According to the UTAUT
model as depicted in Fig. 1, voluntariness of use only moderates the relationship between social influence
and behavioral intention. We maintain that aspect of UTAUT in our research model. Therefore:

H6. Voluntariness of use will moderate the relationship between social influence and intention to use
the continuous auditing technology.

4. Data and sample

We conducted an online survey by e-mailing respondents a link to our electronic survey site.5 The survey
for this study was e-mailed on our behalf by the IMA to their worldwide members whose membership profile
listed one of the following responsibilities: internal auditing, risk management, information systems or
general accounting. 6 The survey e-mail explained the nature of the survey and asked respondents to complete

5
The Qualtrics tool was used for this survey.
6
The IMA has substantially more members who profile themselves as “accountants” than they do “internal auditors,” “risk
management,” or “information systems.” Hence, while our survey captured practitioners in the latter three groups, we also captured
practitioners in the “accountants” group who either had direct knowledge of their company's internal audit operations or passed on
the email to someone else in the company who did.
254 G.C. Gonzalez et al. / International Journal of Accounting Information Systems 13 (2012) 248–262

the survey provided they were knowledgeable on their company's continuous auditing efforts and, if not, to
forward the e-mail to someone within their company's internal audit function.
The number of surveys e-mailed was 9013 and the number of usable responses was 210 (a response
rate of 2.33%). Of those 210 usable responses, the percentage breakdown by regional geographic location
of company operations was: North America 59.0%; Middle East 28.6%; Asia 5.7%; Europe 4.8%; and others
1.9%. These percentages are comparable to the IMA's worldwide membership breakdown of North
America 72%; Middle East 16%; Asia 8%; Europe 3%; and others 1%. 7
The first question in the survey posed to respondents was “What is the current state of Continuous
Auditing in your company?” The four possible responses and the percentage of respondents selecting each
response were:

• Fully operational in one or more of our company's systems — 21%

• In place but not yet fully developed — 22%
• Not implemented yet but scheduled to be implemented in future — 16%
• Not implemented and no plans for future implementation — 40%.

After the introductory question, respondents were asked to answer a set of approximately 40
questions. The questions were of three main types: (1) company profile questions such as annual sales and
geographic location of operations, (2) questions regarding current use of and future plans for continuous
auditing, and (3) questions structured on the UTAUT technology use framework described in the last
section. The actual survey questions for this latter group are shown in Table 1. 8 Survey responses to these
questions enabled us to analyze the current state of use of continuous auditing technology using the
UTAUT framework. We next describe the method and analysis.

5. Research methods

To estimate the paths between the constructs shown in our research model (Fig. 2), and thereby test
the propositions advanced previously, we used partial least squares (PLS) analysis, which is a powerful
multivariate analysis technique. PLS is useful for analyzing structural equations with latent variables. It is
similar to LISREL, which is probably the best known of the second generation statistical techniques, in the
sense that the measurement and structural (or theoretical) models are analyzed simultaneously. However,
unlike LISREL, PLS relies on ordinary least squares estimation techniques to solve the equations (Compeau
and Higgins, 1995). PLS is most appropriate when sample sizes are small, when assumptions of
multivariate normality and interval scaled data cannot be made, and/or when the researcher is primarily
concerned with prediction of the dependent variable (Birkinshaw et al., 1995). The major benefits of PLS
include robustness for small to medium sample sizes and fewer constraints on the data (e.g., normality
assumptions) compared to covariance-based methods such as LISREL (Wakefield et al., 2008). Simulation
studies have also shown PLS to be robust against inadequacies such as multi-collinearity, skewness and
omission of regressors (Cassel et al., 1999).
In PLS all relationships are modeled simultaneously, sharply reducing concerns about multicollinearity
(Inkpen and Birkinshaw, 1994). The path coefficients obtained from a PLS analysis are standardized
regression coefficients, while the loadings of items on individual constructs are factor loadings. Factor scores
created using these loadings are equivalent to weighted composite indices. Thus, PLS results can be easily
interpreted by considering them in the context of regression and factor analysis (Birkinshaw et al., 1995). The
R2 values are used to assess the proportion of variance in the endogenous constructs which can be accounted
for by the antecedent constructs (Compeau and Higgins, 1995). Generally, PLS results are presented in two
stages. In the first stage, the researcher ensures that the measures used as operationalizations of the
underlying constructs are both reliable and valid. Once convinced of the adequacy of the measurement model,
the researcher proceeds to the second stage and interprets the resulting model coefficients (Birkinshaw et al.,
1995).

7
IMA membership profile information as of April 30, 2011.
8
The actual survey questions for the first two of the three referenced question types were included in the survey for descriptive
data gathering purposes only. Their response data are not used in our theoretical framework and analysis and have therefore not
been included in this paper. They are available upon request.
 G.C. Gonzalez et al. / International Journal of Accounting Information Systems 13 (2012) 248–262 255

Table 1
Constructs and corresponding survey questions based on UTAUT framework.

Construct Indicator Survey questiona

Effort expectancy EE1 Interacting with the continuous auditing system is/would be generally clear and understandable
EE2 It is/would be generally easy to become skillful at using the continuous auditing system
EE3 I (we) find/would find the continuous auditing system easy to use
EE4 Learning to operate the continuous auditing system is/would be easy for me (us)
Performance PE1 A continuous auditing system is/would be useful in my (our) job
expectancy PE2 A continuous auditing system enables/would enable me (us) to accomplish tasks more quickly
PE3 A continuous auditing system increases/would increase my (our) productivity
PE4 A continuous auditing system increases/would increase my (our) chances of improving my
(our) financial position
Facilitating FC1 My (our) company has the resources necessary to use the continuous auditing system
conditions FC2 My (our) company has the knowledge necessary to use the continuous auditing system
FC3 The continuous auditing system is/would be compatible with other systems I (we) use
FC4 A specific person (or group) is/would be available for assistance with the continuous auditing
system difficulties
Social influence SI1 People or parties who influence my (our) behavior think/would think that I (we) should use the
continuous auditing system
SI2 People or parties who are important to me (us) think/would think that I (we) should use the
continuous auditing system
SI3 Senior management has been/would be helpful in the use of the continuous auditing system
SI4 In general, my (our) organization has supported/would support the use of the continuous
auditing system
Intention to use DV1 I (we) intend to use the continuous auditing system in the foreseeable future
DV2 I predict I (we) would use the continuous auditing system in the coming future
DV3 My (our) use of continuous auditing is very likely to occur soon
Annual sales Sales 1 = less than $1 million; 2 = $1 million to $10 million; 3 = $10 million to $100 million;
4 = $100 million to $1 billion; 5 = more than $1 billion
Voluntariness Voluntary The use of continuous auditing system/if in the future your company were to adopt the
continuous auditing system, its use is likely to be: 1 = voluntary; 2 = mandated
a
These questions represent the set of questions we asked of survey respondents, as designated by UTAUT. Prior to asking the
above set of UTAUT questions, we identified whether respondents (A) currently have a continuous auditing system, either fully
operational or in place but not yet fully implemented, or (B) had a scheduled but not yet implemented continuous auditing system,
or did not have any plans for a system. The above UTAUT questions were worded identically for all respondents, except that for “A”
category respondents the questions were worded in the present tense (“is,” etc.), while for “B” category respondents the questions
were worded in the future tense (“would be,” etc.). This minor distinction is reflected above by including both sets of words (e.g., “is/
would be”). For purposes of clarity in the survey, however, actual survey questions included usage of one or the other of the two
tenses, but not both as shown above. For each but the last two questions above, respondents had five choices from which to choose
(1 = strongly disagree; 5 = strongly agree). For the other two questions (annual sales and voluntariness) respondents' choices were
as shown in the survey question column.

6. Data analysis and results

6.1. Measurement model

Convergent validity indicates that measures of constructs that should be theoretically related to each
other are, in fact, observed to be related. A composite reliability value of 0.70 or above and an average
variance extracted value of more than 0.50 are deemed as indicators of acceptable level of convergent
validity of measures (Chin, 1998). As evident in Table 2, all average variance extracted (AVE) values are
above .50 and composite reliability coefficients are above .70 for each construct. This indicates that the
measurements are reliable and the latent constructs account for more than 50% of the variance in the
items. The loadings are also in the acceptable range and all the t-values shown in the table suggest that
they are significant at the .01 level.
Discriminant validity is the extent to which the measure is not a reflection of some other construct. It is
indicated by low correlations between the measure of interest and other measures. If the square root of
the AVE is greater than all of the inter-construct correlations, it is evidence of sufficient discriminant
validity (Chin, 1998). Table 3 suggests that our measurement model demonstrates sufficient discriminant
validity.
256 G.C. Gonzalez et al. / International Journal of Accounting Information Systems 13 (2012) 248–262

Table 2
Loadings of the indicator variable.

Constructa Indicator Mean SD Loading t-value

Effort expectancy (.694) (.900) EE1 3.610 .738 .837 24.208

EE2 3.670 .790 .808 19.847
EE3 3.520 .765 .857 29.844
EE4 3.630 .755 .829 20.705
Performance expectancy (.739) (.918) PE1 3.920 .835 .824 25.568
PE2 3.670 .964 .900 51.707
PE3 3.650 .943 .868 31.637
PE4 3.780 .978 .843 30.594
Facilitating conditions (.713) (.908) FC1 3.380 1.030 .867 36.932
FC2 3.440 .977 .844 29.453
FC3 3.310 .904 .833 31.714
FC4 3.270 .971 .834 28.378
Social influence (.710) (.907) SI1 3.360 .853 .828 26.095
SI2 3.490 .826 .808 18.502
SI3 3.540 .993 .872 47.838
SI4 3.570 .967 .860 34.815
Intention to use (.846) (.942) DV1 3.230 1.114 .925 56.959
DV2 3.330 1.150 .924 64.594
DV3 3.100 1.198 .911 49.480
Annual sales Sales 3.400 1.280 1 n/a
Voluntariness Voluntary 1.220 .415 1 n/a
a
The figures in parentheses shown underneath each construct name are average variance extracted (AVE) and composite
reliability, respectively.

To further assess validity of our measurements, we also constructed a cross loading table (Table 4) as
suggested by Gefen et al. (2000). If each item loading in the table is higher on its assigned construct than
on other constructs, it is evidence of adequate convergent and discriminant validity. As can be seen in
Table 4, all the diagonal elements are high and also greater than off-diagonal elements, suggesting
adequate convergent and discriminant validity of our measures.

6.2. Common method bias

Common method bias may occur if the predictor and the criterion variables share a common method.
In such a scenario, the common method may exert a systematic effect on the observed correlations
between measures. Thus, at least partially, common method biases may pose a rival explanation for the
observed correlations between the measures. Similarly, a common rater bias may occur due to any
artifactual covariance between predictor and the criterion variable produced by the fact that the
respondent providing the measure of these variables is the same (Podsakoff et al., 2003). In order to rule
out any rival explanations due to common method bias, we follow Liang et al. (2007) and include in the

Table 3
Correlations among major constructs.

Intention to use EE FC PE SI Sales Voluntary

Intention to use 0.846

Effort expectancy (EE) 0.416⁎⁎ 0.694
Facilitating conditions (FC) 0.499⁎⁎ 0.489⁎⁎ 0.713
Performance expectancy (PE) 0.441⁎⁎ 0.616⁎⁎ 0.433⁎⁎ 0.739
Social influence (SI) 0.568⁎⁎ 0.501⁎⁎ 0.713⁎⁎ 0.571⁎⁎ 0.710
Annual sales (sales) 0.085 0.060 0.269⁎⁎ 0.078 0.120 1.000
Voluntariness (voluntary) 0.311⁎⁎ 0.106 0.345⁎⁎ 0.183⁎⁎ 0.267⁎⁎ 0.177⁎ 1.000

Diagonals are AVE values.

⁎
p b .05 (2-tailed).
⁎⁎
p b .01 (2-tailed).
 G.C. Gonzalez et al. / International Journal of Accounting Information Systems 13 (2012) 248–262 257

Table 4
Item loadings and cross loadings.

Intention to use EE FC PE SI Sales Voluntary

DV1 0.925 0.385 0.459 0.420 0.523 0.028 0.272

DV2 0.924 0.406 0.440 0.384 0.537 0.073 0.240
DV3 0.911 0.357 0.478 0.412 0.509 0.131 0.343
EE1 0.379 0.837 0.457 0.581 0.533 0.032 0.043
EE2 0.328 0.808 0.337 0.524 0.357 0.064 0.119
EE3 0.347 0.857 0.456 0.473 0.410 0.064 0.134
EE4 0.325 0.829 0.369 0.466 0.353 0.041 0.063
FC1 0.458 0.381 0.867 0.363 0.589 0.261 0.297
FC2 0.384 0.328 0.844 0.335 0.575 0.242 0.282
FC3 0.409 0.473 0.833 0.389 0.644 0.219 0.316
FC4 0.430 0.467 0.834 0.372 0.603 0.187 0.270
PE1 0.350 0.498 0.356 0.824 0.487 0.062 0.231
PE2 0.400 0.546 0.390 0.900 0.505 0.115 0.133
PE3 0.353 0.557 0.350 0.868 0.466 0.058 0.101
PE4 0.406 0.516 0.387 0.843 0.501 0.032 0.166
SI1 0.425 0.411 0.544 0.485 0.828 0.088 0.211
SI2 0.390 0.425 0.468 0.478 0.808 0.031 0.132
SI3 0.540 0.427 0.645 0.531 0.872 0.086 0.235
SI4 0.531 0.430 0.706 0.435 0.860 0.179 0.298
Sales 0.085 0.060 0.269 0.078 0.120 1.000 0.177
Voluntary 0.311 0.106 0.345 0.183 0.267 0.177 1.000

PLS model a common factor whose indicators include all the principal constructs' indicators and calculate
each indicators' variances substantively explained by the principal construct and by the method.
The evidence for common method bias can be obtained by examining the statistical significance of the
factor loadings on the method factor and comparing the variances of each observed indicator explained by
its substantive construct and the method factor (Williams et al., 2003). The squared values of substantive

Table 5
Common method bias analysis.

Construct Indicators Substantive factor loading (R1) R12 Method factor loading (R2) R22

Effort expectancy EE1 0.671*** 0.450 0.198** 0.039

EE2 0.840*** 0.706 −0.045 0.002
EE3 0.893*** 0.797 −0.037 0.001
EE4 0.926*** 0.857 −0.114* 0.013
Performance expectancy PE1 0.825*** 0.681 0.002 0.000
PE2 0.912*** 0.832 −0.012 0.000
PE3 0.916*** 0.839 −0.052 0.003
PE4 0.780*** 0.608 0.066 0.004
Facilitating conditions FC1 0.895*** 0.801 −0.042 0.002
FC2 0.966*** 0.933 −0.14** 0.020
FC3 0.748*** 0.560 0.109* 0.012
FC4 0.767*** 0.588 0.078 0.006
Social influence SI1 0.951*** 0.904 −0.117 0.014
SI2 0.995*** 0.990 −0.185** 0.034
SI3 0.721*** 0.520 0.152* 0.023
SI4 0.715*** 0.511 0.143* 0.020
Voluntariness Voluntary 1.000*** 1.000 0.000 0.000
Annual sales Sales 1.000*** 1.000 0.000 0.000
Intention to use DV1 0.926*** 0.857 0.001 0.000
DV2 0.932*** 0.869 −0.008 0.000
DV3 0.902*** 0.814 0.008 0.000
Average 0.768 0.000 0.009
⁎
p b .05.
⁎⁎
p b .01.
⁎⁎⁎
p b .005.
258 G.C. Gonzalez et al. / International Journal of Accounting Information Systems 13 (2012) 248–262

Effort
Expectancy
.152*

Performance
Expectancy
Intention
to Use
R 2 = .443
Facilitating
Conditions
.331**
Social - .253**
Influence
-.141**

Sales Voluntary

* p < .05 ; ** p < .01

Note: Solid lines in the figure represent significant paths while insignificant paths
are represented by dashed lines.

Fig. 3. UTAUT findings. *p b .05 and **p b .01. Note: solid lines in the figure represent significant paths while insignificant paths are
represented by dashed lines.

and method factor loadings are interpreted as variances explained by the substantive and method
constructs, respectively. If the method factor loadings are insignificant and substantive variances are
substantially greater than the method variances, then we may conclude that common method bias is
unlikely to be of any concern (Liang et al., 2007). As seen in Table 5, the average substantive variance is
.768 while the average method variance is .009. The ratio of substantive variance to method variance is
thus about 85:1. This ratio suggests that the variance explained by the model is 85 times more than the
variance attributable to common method bias. In addition, most method factor loadings are not significant.
Given the small magnitude and insignificance of method variance we argue that common method bias is
unlikely to be of concern for this study. 9

6.3. Hypothesis testing

Fig. 3 presents the estimates obtained from the PLS analysis. The R 2 value of .443 indicates that a
significant amount of variance (44.3%) is explained by the model. The effect of effort expectancy on
intention to use was significant at the .05 level (b = .152), providing support for H1. Our analysis did not
find support for H2 or H3 which represent the effects of performance expectancy and facilitating conditions
on intention to use, respectively. Social influence was found to have a significant impact on internal
auditors' intentions to use continuous auditing technology (b = .331, p b .01) providing support for H4.
As for H5a–H5d, represented by the four arrows extending from the Sales box in Fig. 3, the figure shows
that the links for H5a, H5b and H5d are insignificant thereby failing to confirm the moderating role of
annual sales on effort expectancy, facilitating conditions and social influence, respectively. However, the link
for H5c, performance expectancy, was significant at the .01 level (b = −.141). Finally, the link for H6 was
significant at the .01 level (b = −.253) thereby lending support to the role of voluntariness of use as a
moderator for the relationship between social influence and intention to use.

6.4. Additional analysis

To further analyze our worldwide data set and to explore whether respondents coming from different
parts of the world may be responding differently, we split our data based on the two largest demographic
respondent groups in our sample. With 124 responses, North America was our largest demographic

9
To our knowledge there are no rules of thumb to suggest how large this ratio should be. However, the magnitude of the 85:1
ratio reported in this study compares favorably to the 42:1 ratio reported by Liang et al. (2007).
 G.C. Gonzalez et al. / International Journal of Accounting Information Systems 13 (2012) 248–262 259

Table 6
Region-wise breakdown of results.

North America North America Middle East only Middle East and Other regions
only and other regions other regions

Sample size 124 150 60 86 26

R-squared 41 42.1 26.5 34.3 67
Path t-val Path t-val Path t-val Path t-val Path t-val
EE 0.067 0.628 0.051 0.543 0.155 0.925 0.179 1.45 0.114 1.423
PE −0.053 0.491 0.025 0.281 0.16 1.45 0.202⁎⁎ 2.158 0.21⁎⁎⁎ 3.252
FC 0.166 1.378 0.091 0.852 0.243 1.345 0.13 0.884 −0.046 0.582
SI 0.425⁎⁎⁎ 3.85 0.47⁎⁎⁎ 5.261 0.013 0.089 0.164⁎ 1.65 0.579⁎⁎⁎ 5.565
Sales 0.033 0.482 0.039 0.648 −0.087 0.633 −0.036 0.357 0.086 1.29
Voluntariness 0.124 1.49 0.117⁎ 1.717 0.244⁎⁎ 2.011 0.255⁎⁎⁎ 3.052 0.248⁎⁎⁎ 3.912
⁎
p b .05.
⁎⁎
p b .01.
⁎⁎⁎
p b .005.

respondent group, followed by 60 respondents from the Middle East. The rest of the 26 respondents came
from Europe, Asia, Australasia and South America combined (labeled as “Other Regions”). Using these 3
respondent groups, we analyzed a main-effects only model with no interaction effects because we deemed
the sample sizes of each demographic group to be insufficient for robust moderation analyses. Table 6
presents the region-wise breakdown of the results.
Looking at Table 6 we find that the North America sample is indeed different from the Middle East
sample. Social Influence was found to be highly significant for the North American sample at .01 level (b =
.425). In contrast, we found that Voluntariness of Use construct was highly significant for the Middle East
sample at the .01 level (b = .244). Since we measured voluntariness using a binary scale item, the positive
path coefficient suggests that it is the lack of voluntariness in use (i.e. mandated use) that makes it more
likely for Middle Eastern auditors to use the system.
Quiet surprisingly we also found that, given its small sample size (n= 26) the “Other Regions” model had
a very high R-squared value including highly significant paths for Performance Expectancy, Social Influence and
Voluntariness of Use. Including this sample with either the North American or Middle East samples is likely to
bias results upwards in these two models as is evident in second and the fourth columns of Table 6. On
extrapolating the region wise trends from Table 6, we argue that the critical antecedent for building the
positive intentions to use Continuous Auditing technology among North American users is Social Influence,
while Voluntariness of Use (i.e. mandated use) is the critical factor for Middle Eastern users.
We believe that these interesting results can be explained using Hofstede's “power distance” concept
(Hofstede, 1980). Power distance is the extent to which the less powerful members of organizations and
institutions accept and expect that power is distributed unequally. High power distance cultures such as
the Middle East are more likely to believe that superiors should have greater degree of power over
subordinates. In such cultures, subordinates have a higher tendency to defer to power which we believe
explains the significance of Voluntariness of Use construct in determining the respondents' intentions to
use the system. Therefore we argue that the Middle Eastern respondents are more likely to use the system
if it is mandated by higher authorities. On the other hand, cultures such as North America that believe in
relatively low power distance, hold that authorities should wield lesser degree of power over
subordinates. Respondents in such cultures are more likely to respond to democratic and consultative
pressures such as Social Influence from peers and higher management to use the system. This suggests that
cultural differences may dictate different strategies in convincing respondents to use the Continuous
Auditing system depending on their geographic location.

7. Discussion and conclusions

In spite of the fact that the concept of continuous auditing was first introduced over two decades ago,
and that the concept has garnered a considerable amount of attention in both the academic and
professional literature, to date continuous auditing has been used to a limited extent, and almost
260 G.C. Gonzalez et al. / International Journal of Accounting Information Systems 13 (2012) 248–262

exclusively in the internal audit domain. While past surveys of CAEs and other internal audit executives
have indicated that plans for the implementation of continuous auditing were robust (PwC, 2006; KPMG,
2010; Grant Thornton, 2011), the results of this study's survey indicate that while some progress has been
made, continuous auditing systems are not yet used widely.
This study explores that lack of widespread use of continuous auditing systems, and seeks answers to why
this is the case. We surveyed internal auditors worldwide about the current state of use of continuous
auditing technology in their companies. Through application of the Unified Theory of Acceptance and Use of
Technology (UTAUT) framework we analyzed respondents' answers to find that the significant factors
leading to intentions to use of continuous auditing technology are Effort Expectancy, i.e., the ease and clarity of
use, and Social Influence, i.e., the support and encouragement of key organizational members. These two key
results suggest that two key drivers for increasing the use of continuous auditing systems would be the
demonstration to internal auditors that a continuous auditing system is easy to learn and operate and reduces
the amount of work necessary to carry out audits (Effort Expectancy), and management visibly and actively
promoting the use of continuous auditing as a valuable internal auditing tool (Social Influence).
In addition, the level of annual sales volume seems to moderate another important construct, performance
expectancy. Large firms tend to use continuous systems without overly strict questioning of the systems'
performance expectancy, but use in firms with lower sales volume tends to follow a stricter requirement that
the software be deemed to pass cost/benefit scrutiny. Another moderator, the extent to which the system is
perceived to be voluntary or mandatory, also moderates the effect of social influence on intention to use, as
UTAUT predicts, which is not too surprising as the mandatory use of continuous auditing, possibly by a
governing authority10 or perhaps even by the company's board of directors, would logically carry heavier
pressure than in a voluntary environment.
The UTAUT results regarding Effort Expectancy, as just discussed, seem quite positive and bode well for
future use and expanded utilization of continuous auditing technology. It therefore becomes a matter of
companies' commitment to the use of the technology: an organization that decides to implement continuous
auditing, or expand its use, will likely be able to count on little resistance and, more likely, strong support from
its internal auditors. Indeed, the result regarding Social Influence bears this out — it almost seems as if internal
auditors are ready and waiting for their company's top management to commit to continuous auditing.
Finally, our analysis suggests strong cultural differences among the critical antecedents affecting
internal auditors' intentions to use the system. North American internal auditors are more likely to
respond to soft coercion pressures created by positive Social Influence through peers and higher
authorities. On the other hand, Middle Eastern internal auditors are much more likely to use the system if
it is mandated by the higher authorities. Therefore, we suggest that different strategies depending on the
geographic location should be employed to help create the relevant kinds of pressures on internal auditors
to use the Continuous Auditing system. The prominence of Social Influence and Voluntariness of Use
concepts in our analysis suggests that the lag in the use of Continuous Auditing systems can be readily
explained by the lack of socially conducive or coercive pressures in the auditing field.

Acknowledgments

The authors thank Mary Curtis and Ray Henrickson for their valuable comments and suggestions, as
discussants of an earlier version of this paper presented at the 7th biennial University of Waterloo
symposium on information integrity and information systems assurance (held on October 20–21, 2011).
Editor's Note: As Ray Henrickson's comments and suggestions have been incorporated in the present
version of the paper, a separate set of these comments is not provided in this issue.
We are grateful to the Institute of Management Accountants for their generous support.

10
In our view the mandatory imposition of the use of continuous auditing could eventually arise from the required issuance of
XBRL-based financial statements on frequent basis, perhaps even real-time at some point in the future. Alternatively, the imposition
of continuous auditing itself may not be mandatory but would be deemed all but necessary with mandatory XBRL-based financial
statements.
 G.C. Gonzalez et al. / International Journal of Accounting Information Systems 13 (2012) 248–262 261

References

Ajzen I. From intentions to actions: a theory of planned behavior. In: Kuhl J, Beckman J, editors. Behavior. Action-control: from
cognition to behaviorHeidelberg, Germany: Springer; 1985. p. 11–39.
Ajzen I. Attitudes, traits, and actions: dispositional prediction of behavior in personality and social psychology. In: Berkowitz L,
editor. Advances in experimental social psychology, vol. 20. New York: Academic Press; 1987. p. 1-63.
Ajzen I. The theory of planned behavior. Organ Behav Hum Decis Process 1991;50(2):179–211.
Alles G, Kogan A, Vasarhelyi MA. Putting continuous auditing theory into practice: lessons from two pilot implementations. J Inf Syst
2008;22(2):195–214.
Bedard J, Jackson C, Ettredge M, Johnstone K. The effect of training on auditors' acceptance of an electronic work system. Int J Account
Inf Syst 2003;4:227–50.
Birkinshaw J, Morrison A, Hulland H. Structural and competitive determinants of a global integration strategy. Strateg Manage J
1995;16(8):637–55.
Brown C, Wong J, Baldwin A. A review and analysis of the existing research streams in continuous auditing. J Emerging Technol
Account 2007;4:1-28.
Cassel C, Hackl P, Westlund AH. Robustness of partial least-squares method for estimating latent variable quality structures. J Appl
Stat 1999;26:435–46.
Chan D, Vasarhelyi M. Innovation and practice of continuous auditing. Int J Account Inf Syst 2011;12:152–60.
Chin WW. The partial least squares approach to structural equation modeling. In: Marcoulides GA, editor. Modern methods for
business research. Hillsdale, NJ: Lawrence Erlbaum Associates; 1998. p. 295–336.
Compeau DR, Higgins CA. Application of social cognitive theory to training for computer skills. Inf Syst Res 1995;6(2):118–43.
Curtis M, Payne E. An examination of contextual factors and individual characteristics affecting technology implementation decisions
in auditing. Int J Account Inf Syst 2008;9:104–21.
Davis FD. Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Q Manag Inf Syst
1989;13(3):319–39.
Debreceny RS, Gray GL, Jun-Jin Ng J, Siow-Ping Lee K, Yau W. Embedded audit modules in enterprise resource planning systems:
implementation and functionality. J Inf Syst 2005;19(2):7-27.
DeSanctis G, Poole M. Capturing the complexity in advanced technology use: adaptive structuration theory. Organ Sci 1994;5(2):
121–47.
Dowling C. Discussion of “An examination of contextual factors and individual characteristics affecting technology implementation
decisions in auditing”. Int J Account Inf Syst 2008;9:122–6.
Dowling C. Appropriate audit support system use: the influence of auditor, audit team, and firm factors. Account Rev 2009;84(3):
771–810.
Fishbein M, Ajzen I. Belief, attitude, intention, and behavior. Reading, MA: Addison-Wesley; 1975.
Flowerday S, von Solms R. Continuous auditing: verifying information integrity and providing assurances for financial reports.
Comput Fraud Secur 2005:12–6[July 2005].
Gefen D, Straub D, Boudreau M-C. Structural equation modeling and regression: guidelines for research practice. Commun AIS
2000;4(7).
Grant Thornton. Looking to the future: perspectives and trends from internal audit leaders. Grant Thornton LLP; 2011.
Greenfield G, Rohde F. Technology acceptance: not all organisations or workers may be the same. Int J Account Inf Syst 2009;10:
263–72.
Groomer S, Murthy U. Continuous auditing of database applications: an embedded audit module approach. J Inf Syst 1989;3(2):
53–69.
Hermanson D, Moran B, Rossie C, Wolfe D. Continuous monitoring of transactions to reduce fraud, misuse, and errors. J Forensic
Account 2006;VII:17–30.
Hofstede G. Culture's consequences: international differences in work-related values. 2nd ed. Beverly Hills CA: SAGE Publications; 1980.
Hunton J, Mauldin E, Wheeler P. Potential functional and dysfunctional effects of continuous monitoring. The Accounting Review
2008;83(6):1551–69.
Hunton J, Mauldin E, Wheeler P. Continuous monitoring and the status quo effect. Int J Account Inf Syst 2010;11(3):239–52.
Inkpen A, Birkinshaw J. International joint ventures and performance: an interorganizational perspective. Int Bus Rev 1994;3(3):
201–17.
Kim H, Mannino M, Nieschwietz R. Information technology acceptance in the internal audit profession: impact of technology
features and complexity. Int J Account Inf Syst 2009;10:214–28.
Kogan A, Sudit EF, Vasarhelyi MA. Continuous online auditing: a program of research. J Inf Syst 1999;13(2):87-103.
KPMG. Continuous auditing and monitoring: are promised benefits now being realised? KPMG LLP; 2010.
KPMG International (KPMG). What is driving continuous auditing & continuous monitoring today? KPMG International Cooperative;
2010.
Kuhn J, Sutton S. Learning from WorldCom: implications for fraud detection through continuous assurance. J Emerging Technol
Account 2006;3:61–80.
Kuhn J, Sutton S. Continuous auditing in ERP system environments: the current state and future directions. J Inf Syst 2010;24(1):91-112.
Liang H, Saraf N, Hu Q, Xue Y. Assimilation of enterprise systems: the effect of institutional pressures and the mediating role of top
management. MIS Q 2007;31(1):59–87.
Podsakoff P, MacKenzie S, Lee J, Podsakoff N. Common method biases in behavioral research: a critical review of the literature and
recommended remedies. J Appl Psychol 2003;88(5):879–903.
PricewaterhouseCoopers (PwC). State of the internal audit profession study: continuous auditing gains momentum. Available at:
http://www.pwc.com/en_BE/be/systems-process-assurance/pwc-state-of-internal-audit-2006.pdf2006.
PricewaterhouseCoopers (PwC). State of the internal audit profession study: pressures build for continual focus on risk. Available at:
http://www.pwc.com/en_US/us/internal-audit/assets/state_internal_audit_profession_study_07.pdf2007.
Rezaee Z, Sharbatoghlie A, Elam R, McMickle P. Continuous auditing: building automated auditing capability. Auditing J Pract Theory
2002;21(1):147–63.
262 G.C. Gonzalez et al. / International Journal of Accounting Information Systems 13 (2012) 248–262

The Institute of Internal Auditors (IIA). Global technology audit guide 3: continuous auditing: implications for assurance, monitoring,
and risk assessment. Available at:http://www.theiia.org/guidance/technology/gtag3/2005.
Vasarhelyi M, Halper F. The continuous audit of online systems. Auditing J Pract Theory 1991;10(1):110–28.
Vasarhelyi M, Alles M, Kogan A. Principles of analytic monitoring for continuous assurance. J Emerging Technol Account 2004;1(1):
1-21.
Vasarhelyi M, Kogan A, Alles M. Would continuous auditing have prevented the Enron mess? CPA J July 2002:80.
Venkatesh V, Morris MG, Davis GB, Davis FD. User acceptance of information technology: toward a unified view. MIS Q 2003;27(3):
425–78.
Wakefield RL, Leidner DE, Garrison G. A model of conflict, leadership, and performance in virtual teams. Inf Syst Res 2008;19(4):434–55.
Williams LJ, Edwards JR, Vandenberg RJ. Recent advances in causal modeling methods for organizational and management research. J
Manag 2003;29(6):903–36.