`
Does Forensic Accounting Matter? Diagnosing Fraud Using the Internal
Control System and Big Data on Audit Institutions in Indonesia
ENNY SUSILOWATI MARDJONO1, ENTOT SUHARTONO1, GURUH TAUFAN HARIYADI2
1Department of Accounting,
Universitas Dian Nuswantoro,
INDONESIA
2Department of Management,
Universitas Dian Nuswantoro,
INDONESIA
Abstract: This study aims to determine the relationship between awareness of the use of forensic accounting and
the role of the internal control system (COSO) mediated by Big Data Analysis (BDA) towards interest in using
forensic accounting in detecting fraud. The research design is a case study with a quantitative approach. The
sample for this study was 331 auditor respondents spread across KAP, BPK and BPKP in Indonesia. The data
used is primary data with research methods through interviews and surveys at the Indonesian auditor
institutions. The weakness of companies is that on average they still use data systems that are not integrated, so
there are risks in terms of data security. The results of this study prove that Big Data Analysis mediates the
relationship between Awareness of Forensic Accounting on Intentions of Forensic Accounting. Also, Big Data
Analysis mediates the relationship between COSO on Intentions of Forensic Accounting. The results show that
the seventh hypothesis proposed is statistically proven. This study proves that the implementation of a good
internal control system will be an effective tool to control fraud risk. Internal controls can be fully effective if
the organization is able to understand the most vulnerable risks and how to respond to fraud. BDA with data
mining techniques that contribute to decision-making and fraud detection. Auditors can find and extract hidden
patterns in large amounts of data by using Big Data to detect fraud.
Key-Words: - big data analysis, forensic accounting, internal control system, fraud detection, Audit Institutions,
Indonesia.
Received: March 19, 2023. Revised: December 21, 2023. Accepted: January 16, 2024. Published: February 9, 2024.
1 Introduction
One of the primary purposes of financial statements
is to furnish prospective investors, creditors, and
other users with information that facilitates logical
investment, credit, and analogous decision-making.
Financial statements ought to satisfy the qualitative
requirements and be presented in adherence to the
stipulations outlined in the Statement of Financial
Accounting Standards (PSAK No. 1), [1]. However,
the results of audits conducted by public accountants
do not fully guarantee that financial statements are
free from fraud or fraud, [2]. According to a study
by Price Waterhouse Coopers, thirty percent of the
businesses surveyed had fallen victim to fraud, [3].
Fraud cases that occur in Indonesia occur at PT.
Garuda Indonesia Tbk for fiscal year 2018 and PT.
Waskita Karya tbk for the 2019 financial year
concerning financial statement engineering. 80% of
fraud committed by companies in Indonesia is
financial statement fraud, [4].
This study is also reinforced by the results of the
annual report from ACFE (Association of Certified
Fraud Examiners), where 81% of the organizations
surveyed have been victims of fraud with losses of
USD $ 100,000 per case and 42% of the main
perpetrators are from internal management itself, [2].
Due to the company's failure to implement internal
controls, fraud occurs, and auditors have the
knowledge and abilities to identify fraud in the
company's books, [5]. The implementation of the
management control system is in the scope of
improving information systems and increasing
auditor competence, especially in the provision of
non-audit services (Tax, investigative audit,
consulting services) improving the performance of
public accounting firms, [6].
The existence of fraud cases in America, namely
Enron, Worldcom, Adelphia and others, where the
case is very detrimental to stakeholders and users of
financial statements (creditors, investors,
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2024.21.53
Enny Susilowati Mardjono,
Entot Suhartono, Guruh Taufan Hariyadi
E-ISSN: 2224-2899
638
Volume 21, 2024
`
shareholders, the public) then this triggers the
publication of Sarbanes-Oxley (SOX). The
Sarbanes-Oxley Act (SOA) was created by the
American Association of Certified Fraud Examiners
(ACFE). Research on Taiwanese audit firms shows
that the effect of SOX is further enhanced control
systems using MASs (Management Advisory
Service) strategies and alliance collaboration among
audit firms, [7].
The speed at which financial information or
statements generate data and the volume of data they
contain make it difficult for auditors to identify and
extract hidden patterns in such data, rendering
analysis and drawing accurate and pertinent
conclusions only with big data analytics (BDA), [8].
BDA can find unexpected patterns that cannot be
detected when small samples are used in regular
audits, [8]. But at the time auditors used BDA there
was also a gap in the role of BDA towards
practitioners' intention to adopt forensic accounting,
[8]. Limited auditor skills in analyzing large
volumes of data have an impact on auditors'
difficulties in accessing and identifying relevant
data. Auditors must ensure the accuracy and
reliability of audit results on large enough data;
therefore it becomes a problem if there are no tools
to be able to find potential fraud, [9].
It is critical, both inside and outside the auditing
profession, that BDA techniques be recognized as
legitimate in order to guarantee their application
during audit engagements. In the execution of
digitally advanced audits for client financial
statements, the incorporation of Big Data Analytics
(BDA) as an additional source of audit evidence
reinforces the practical legitimacy of BDA within
the auditing profession, [9]. There are numerous
prospects for applying big data methodologies in the
field of auditing, specifically when stringent
analytical protocols are integrated with conventional
auditing approaches and expert opinion. Recent
advancements in financial hardship models and big
data financial misconduct could benefit audits.
Additional research is warranted on sentiment
analysis and natural language processing as they are
potentially valuable audit tools. Emerging research
avenues in auditing, including collaborative
platforms, peer-to-peer marketplaces, and real-time
information management, are ideally adapted to big
data methodologies, [9]. The purpose and objectives
of this research is to examine the relationship
between forensic accounting, internal control
systems (COSO) and the use of big data to detect
fraud in Audit Institutions in Indonesia.
Auditors are cognizant of the importance of
utilizing forensic accounting to detect fraud, and big
data analysis (BDA) reinforces and facilitates these
activities. This assertion is grounded in the findings
of studies examining internal and external auditors
in Indonesia, [10], [11] and India, [12]. Additional
studies demonstrate the efficacy of the Committee of
Sponsoring Organizations (COSO) model in fraud
detection and prevention. According to the study, all
control activities, risk assessment and response,
internal control, event identification, and risk
assessment and response significantly impact
detecting and preventing financial statement fraud at
commercial banks operating in Jordan, [13].
The main contribution in this research is that this
study integrates the elements of forensic accounting
implementation, big data analysis, and the internal
control framework (COSO) as assessed by internal
and external auditors to identify fraudulent activities,
drawing upon findings from previous research. The
innovation of this research is that the researcher
investigates the potential influence of auditors
incorporating big data analytics and organizations
implementing the internal control framework
(COSO) on the auditors' capacity to identify
fraudulent activities at KAP, BPK, BPKP in
Indonesia. This combination of 3 research elements
such as big data, COSO, forensic accounting is still
rarely carried out by previous research.
2 Empirical Literature Review and
Hypotheses Development
2.1 Forensic Accounting
The fusion of investigative and auditing
methodologies within the accounting field has led to
the cultivation of a specialized skill set known as
"forensic accounting." This skill set is specifically
geared towards identifying and preventing
accounting fraud, [12]. Forensic accounting
encompasses the processes of ascertaining,
documenting, assessing, categorizing, reporting, and
validating historical financial data or other
accounting procedures to address legal issues. This
historical data is also used for the evaluation of
financial data in the resolution of future legal
disputes, [12].
2.2 Big Data
The term "big data" pertains to extensive and
intricate data collections that can be examined to
uncover patterns, trends, and connections. It comes
from various sources, including social media,
machine-generated data, and transactional data. Big
data analytics can create value for organizations by
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2024.21.53
Enny Susilowati Mardjono,
Entot Suhartono, Guruh Taufan Hariyadi
E-ISSN: 2224-2899
639
Volume 21, 2024
`
providing insights into operational and strategic
approaches, [12].
2.3 COSO (Internal Control System)
Internal control is measured by the original COSO
2013 framework standards adapted from the
research of, [13]. They use COSO standards
implemented in local governments to prevent fraud.
COSO comprises control activities, information and
communication, risk assessment, monitoring
activities, and the control environment. Respondents
were asked to answer questions and were assessed
on a Likert scale of 1 to 5 regarding the five
components of integrated internal control.
2.4 Fraud
Engaging in any activity that manipulates others for
personal advantage constitutes "fraud." When
"knowingly misrepresenting the truth or concealing
a material fact to induce another to act to their
detriment," fraud is considered a criminal offence,
[13]. Criminalist Dr. Donald Cressey devised the
Fraud Triangle idea in the 1970s. The hypothesis
states that a person is prone to commit fraud if all
three components are present.
The three components are financial pressure,
perceived opportunity, and rationalization. Financial
pressure refers to the need for money or other
resources, while perceived opportunity refers to the
belief that one can commit fraud without getting
caught. Rationalization refers to the justification or
excuses that a person uses to convince themselves
that committing fraud is acceptable. Anti-fraud
professionals frequently employ the Fraud Triangle
to delineate circumstances that might incentivize
organizations or individuals to partake in fraudulent
activities. Additionally, the model can be employed
to underscore industry-wide or economic
circumstances that may contribute to an increased
overall risk.
2.5 Previous Studies
Auditors face several problems in implementing data
analysis to detect fraud. In addition, the increasing
complexity of internal control and the non-
integration of one data with another data is also an
obstacle for auditors in controlling and detecting
fraud. The influence of social media information
literacy technology that has not been used for
example, is expected to detect fraud accurately, [13].
Therefore, to overcome the above problems, this
study aims to develop an effective fraud detection
and prevention model by using the IT internal
control system (COSO) in forensic accounting and
using BDA (data analysis of financial statements,
annual reports, social media). This method can assist
auditors in overcoming the problem of limited skills
in analyzing large amounts of data, as well as in
finding, identifying, and extracting hidden patterns
in large amounts of data so that audit results are
accurate and reliable, [13].
This study examines how Indonesian auditors use
BDA, internal control, and forensic accounting to
uncover fraud. BDA helps derive structural equation
models for fraud detection from research models and
theories on forensic accounting awareness. Do
results indicate that forensic accounting awareness
affects fraud detection practitioners' intentions?
BDA extracts hidden patterns in massive data sets to
detect financial accounting fraud. We develop a
systematic internal control management system
rooted in internal control principles and artificial
intelligence theory. This paper aims to establish a
data analysis model incorporating intelligent
identification and early warning mechanisms.
Furthermore, the internal control reverse order
method facilitates the autonomous evaluation of
financial indicators.
The novelty of this study is the creation of a new
model that explains in depth the factors that
influence auditors using forensic accounting with
BDA and IT control systems (COSO) to detect
fraud. The model obtained is strengthened using the
content analysis method where the new model is
obtained from processed and strengthened variables
or robustness test, [14], with the use of the Atlas.ti
application with the depth interview method, [14] ,
which can generate new indicators to update the first
model by finding more complex factors that
influence auditors to detect fraud. With this method,
more fraud is detected.
The research roadmap on big data analytics, internal
control, and forensic accounting for fraud detection
can be seen in Figure 1 (Appendix).
In Figure 2, a Conceptual Framework is
depicted, showcasing the correlation between
awareness and the utilization of forensic accounting
and COSO's internal control system. Big Data is
integrated into the framework as a determining
factor that shapes the interest in employing forensic
accounting to detect fraud.
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2024.21.53
Enny Susilowati Mardjono,
Entot Suhartono, Guruh Taufan Hariyadi
E-ISSN: 2224-2899
640
Volume 21, 2024
`
Fig. 2: Empirical Model
Source: Elaborated by the Author
2.6 Hypotheses Development
2.6.1 Awareness of Forensic Accounting and Big
Data
Auditors in carrying out their work face several
problems in implementing data analysis to detect
fraud. Awareness of the use of forensic accounting
in detecting fraud is very important for auditors in
analyzing financial statements. Where with the use
of Forensic Accounting Auditors or examiners can
identify misuse of assets and can make financial data
or information more reliable and accurate.
The growing recognition of the value of forensic
accounting will necessitate an integrated data
structure and diminish the significance of utilizing
Big Data to detect more fraud and operate more
rapidly.
Big data techniques and data analysis are vital for
forensic accounting due to the complexity of
business transactions and the limits of traditional
ways of identifying fraudulent activities. Forensic
accountants can apply a variety of advanced anti-
fraud techniques such as data visualization,
predictive analytics, behavioral analysis, content
analysis, social network analysis, and geo-spatial
analysis to identify abnormal operations, high-risk
areas, and potentially fraudulent activity.
Therefore, it is important to incorporate theoretical
knowledge and practical skills into forensic
accounting training programs to increase
understanding, usefulness, and interest in big data
analytics and data analytics. Accountants must have
an educational or training background in the idea
and implementation of big data and big data
analysis, [14]. Forensic accounting awareness may
increase the desire to use big data technology to
uncover fraud, [14].
H1: Awareness of Forensic Accounting has a
significant impact on Big Data Analysis
2.6.2 Awareness of Forensic Accounting and
Intentions of Forensic Accounting
The intention of using forensic accounting for fraud
detection is influenced by forensic accounting
awareness. This hypothesis is based on the
assumption that practitioners who have a higher
level of forensic accounting awareness will be more
willing to apply forensic accounting for fraud
detection.
The study used data analysis to test this hypothesis
and found that this hypothesis was supported by
data, which suggests that practitioners who have a
higher level of forensic accounting awareness will
be more willing to apply forensic accounting to
fraud detection, [15]. The need for forensic
accounting arises because an organization's
conventional internal audit system fails to detect
accounting fraud effectively.
Forensic accounting is a relatively new discipline for
practitioners. However, it has become notorious for
the rapid increase in scams over the decades.
Forensic accounting plays an important role in
uncovering hard-to-find fraud through mere internal
audits using accounting, auditing, and investigative
skills, [15].
H2: Awareness of Forensic Accounting has a
significant impact on Intentions of Forensic
Accounting
2.6.3 Big Data using and Intentions of Forensic
Accounting
Big data techniques can improve the quality and
effectiveness of financial statement audits and
forensic accounting techniques can help in detecting
potential fraud in financial data. By examining
transactions and financial patterns, forensic
accountants can assist auditors in identifying areas
of risk that require further investigation. This data
can subsequently be employed in constructing
extensive data models specifically designed to
identify financial fraud accurately. The significance
of collaboration between forensic accountants and
data analysts is underscored in the paper,
emphasizing the efficient development and
utilization of big data models.
In summary, forensic accounting studies suggest
an enhancement in big data audits, enabling auditors
to create models that enhance fraud detection
accuracy through the integration of forensic
accounting methodologies. Effective collaboration
between forensic accountants and data analysts
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2024.21.53
Enny Susilowati Mardjono,
Entot Suhartono, Guruh Taufan Hariyadi
E-ISSN: 2224-2899
641
Volume 21, 2024
`
remains imperative for proficiently developing and
utilizing big data models, [15].
H3: Big Data using has a significant impact on
Intentions of Forensic Accounting
2.6.4 COSO and Big Data Using
The COSO framework acknowledges the
significance of big data in the realm of risk
management by characterizing "risk" as involving
"customer relationship management (CRM) or
other databases and big data analysis." The
application of big data analytics allows for
predictive capabilities, aiding organizations in
making pivotal decisions. By utilizing diverse
sources of big data, organizations can garner
valuable insights into organizational risks,
facilitating evaluating and mitigating potential
hazards.
Employing data science technology with
predictive algorithms to analyze extensive datasets
alongside risk assessment enables financial
institutions to gain immediate, real-time insights
into their risks, informing and guiding their risk
management strategy. Therefore, big data analytics
can be used to improve risk management, and the
COSO framework recognizes its importance in
managing risk, [15].
H4: COSO has a significant impact on Big Data
Analysis
2.6.5 COSO and Intentions of Forensic
Accounting
According to, [16], investigate the efficacy of
COSO internal control framework in averting
occupational fraud within local government
organizations. Their findings indicate that COSO
internal controls successfully mitigate asset
misappropriation and financial statement fraud. The
internal control components, encompassing the
control environment, risk assessment, control
activities, information and communication, and
monitoring activities, have proven effective in fraud
prevention. Consequently, COSO is anticipated to
heighten auditors' inclination to employ forensic
accounting to prevent and detect fraud.
In, [16], contend that the internal control system,
proper compensation, and competence of the village
apparatus serve as deterrents to village fund fraud.
The findings suggest that strengthening the internal
control system (COSO) effectively prevents fraud.
Furthermore, moral sensitivity reinforces the
relationship between the effectiveness of village
officials in averting village fund fraud, the internal
control system, and appropriate compensation.
According to, [16], show that forensic accounting
and management control may combat banking
cybercrime. The report recommends fraud risk
management controls for banks to maintain
reputation and regulatory compliance. This proves
that the auditor implementing a good control system
will prevent the risk of fraud. A good COSO will
also increase interest in using forensic accounting to
detect fraud.
H5: COSO has a significant impact on Intentions of
Forensic Accounting
2.6.6 Mediating effect of Big Data in the
relationship between Awareness of
Forensic Accounting and Intentions of
Forensic Accounting
Awareness of Forensic Accounting has a positive
relationship with the intention to use forensic
accounting (Intentions of Forensic Accounting) in
fraud prevention and detection, [16], [17].
The higher the individual's awareness of forensic
accounting, the more positive the correlation with
the intention to use the practice. This awareness
raises an understanding of forensic accounting's role
in detecting financial fraud and irregularities and can
drive people to use it to improve professional skills
and minimize risk. Ethical considerations in
financial reporting also influence an individual's
intention to adopt forensic accounting practices,
[15], [17].
Big Data Analytics (BDA) serves as a tool for
recognizing patterns indicative of fraud, [12], [15].
The integration of BDA into audit processes ensures
the enhancement of audit quality and the efficacy of
fraud detection, [15], [17]. Forensic accountants are
required to possess proficiency in the extraction,
analysis, and visualization of data, given the
substantial volume, speed, and variety inherent in
such datasets, [17].
Big Data Analytics (BDA) functions as a mediator
in facilitating the correlation between Advanced
Forensic Analytics (AFA) and Investigative Forensic
Analytics (IFA) for the purpose of fraud detection,
[15], [17]. The application of BDA stands to
enhance the efficiency of inspections and
investigations within the domain of forensic
accounting, [17]. Forensic accountants are equipped
to leverage advanced anti-fraud methodologies,
including data visualization, predictive analytics,
behavioral analysis, content analysis, social network
analysis, and geospatial analysis, enabling the
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2024.21.53
Enny Susilowati Mardjono,
Entot Suhartono, Guruh Taufan Hariyadi
E-ISSN: 2224-2899
642
Volume 21, 2024
`
detection of anomalous operations, high-risk areas,
and potentially fraudulent activities, [17].
Using big data analysis, forensic accountants can
collect, integrate, and analyze data from multiple
sources to identify suspicious patterns and trends,
[15], [17]. Assist forensic accounting in developing
predictive models to detect fraud and fraud, by
developing predictive models that can identify
suspicious patterns and behaviors and provide early
warnings of potential fraud and fraud, [17].
Forensic accountants can utilize predictive modeling
and advanced analytics to identify suspicious
transactions, high-risk occurrences, and deceptive
behavior by analyzing activity data or past
transactions. Additionally, forensic accountants can
apply entity resolution algorithms to uncover
concealed relationships, addresses, and aliases,
enabling the examination of conflicts of interest,
fraudulent identities, and individuals and businesses
subject to sanctions in database data mining, [18].
H6: Big Data Analysis Mediates the Relationship
between Awareness of Forensic Accounting on
Intentions of Forensic Accounting
2.6.7 Mediating effect of Big Data in the
Relationship between COSO and
Intentions of Forensic Accounting
Control systems help detect and prevent fraud
through policy, surveillance, and data analysis. This
can complement forensic accounting efforts in
identifying and addressing violations. In addition to
early detection, control systems also provide critical
data and documentation for forensic investigations,
[18].
Forensic accounting, however, continues to be of
paramount importance as the last step towards
identifying and tackling fraud which has not been
identified through control systems. Both make an
important contribution to the preservation of
organizational integrity and financial soundness, by
cooperating in parallel with each other, [18]. The
basis for the establishment of effective internal
controls is the COSO framework. This framework's
structure revolves around monitoring the
environment, risk assessment, control activities,
information and communication, and monitoring,
[13], [18].
The fusion of forensic accounting and corporate
governance can enhance internal control. Applying
forensic accounting practices to the COSO
framework allows organizations to fortify their
internal controls and enhance financial reporting
processes, [13], [18].
In order to allow auditors to identify patterns and
anomalies that might indicate fraudulent activities,
BDA could help improve the efficiency of internal
controls through real time monitoring and analysis
of financial transactions and by analyzing large
amounts of data, [15], [18]. BDA can mediate
correlations between COSO's internal control
framework and forensic accounting objectives by
providing real-time monitoring and analysis of
financial transactions.
Forensic accounting and internal auditing can use
big data analytics to identify and investigate
financial fraud, while corporate governance can be
strengthened by integrating forensic accounting
practices, [19].
H7: Big Data Analysis Mediates the Relationship
between COSO on Intentions of Forensic
Accounting
3 Research Method
3.1 Data
This research was conducted in 2 large provinces in
Indonesia, with the number of respondents in
Central Java (Semarang, Solo) being 84% of the
total respondents and DIY (Yogyakarta) being 16%,
as seen in Figure 3. The reason for selecting the
three city locations was because it was based on
OJK data in 2022; the highest concentration of fraud
checks will be concentrated in these three provinces,
[4].
Fig. 1: Composition of Respondents by Region
Source: Elaborated by the Author
This research was undertaken within national
auditing institutions, encompassing audit firms
(KAP), the Indonesian Supreme Audit Institution
(abbreviated nationally as Badan Pemeriksa
Keuangan - BPK), and the State Development Audit
Agency (abbreviated nationally as Badan
Pengawasan Keuangan dan Pembangunan - BPKP).
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2024.21.53
Enny Susilowati Mardjono,
Entot Suhartono, Guruh Taufan Hariyadi
E-ISSN: 2224-2899
643
Volume 21, 2024
`
The first group of respondents to be studied is the
condition of the KAP financial statement auditor in
55%, factors that influence auditors using the COSO
framework, and the ability of BDA by auditors to
detect financial statement fraud. Auditors will be
highlighted from the behavioral aspects of
awareness using forensic accounting, knowledge,
affection, and cognition as well as the attitudes of
auditors who have conducted examinations using
forensic accounting with the help of BDA. The
second group of respondents to be studied in 45%
are BPK auditors (Figure 5) who use forensic
accounting and BDA in carrying out fraud detection
activities and implementing COSO control
Fig. 5: Origin of Respondent Institution
Source: Elaborated by the Author
The research model proposed in this research
can be seen in Figure 4 (Appendix).
3.2 Measures
This study uses primary data as the main input to
improve and develop auditor awareness literacy
using forensic accounting and BDA in detecting
fraud. Primary data were obtained by distributing
questionnaires by snowball sampling method, [20],
in-depth interviews, and group discussions with
auditors in BPK and KAP. Questionnaires are used
to obtain data related to factors that affect auditor
awareness using forensic accounting for decision-
making in preventing and detecting fraud.
Interviews with BPK were conducted to explore
regulatory policies and enforcement of rules
regarding financial statement fraud, while
interviews with some KAP respondents were
conducted to explore factors that influence decision
making to prevent or control and detect fraud. The
technical analysis used is descriptive analysis
(quantitative and qualitative) with a presentation in
the form of tables, narratives, graphs, and charts,
[14], [21].
Descriptive analysis is used to obtain a
comprehensive picture of financial literacy using
SMART PLS (Partial Least Square) analysis, [14],
[21]. Research is also combined with aspects of
cognition, affection, and conation. Map supporting
and inhibiting factors, both internal and external,
[20]. This research will also be complemented by
qualitative analysis related to emerging aspirations
to formulate a model of increasing literacy in the use
of forensic accounting in the detection of financial
statement fraud and financial fraud literacy in
aspects of marketing, social media, and
communication.
The data were collected using a questionnaire
featuring a five-point Likert scale, where 1 denoted
strong disagreement and 5 represented strong
agreement. A pilot test was done on Audit
Institutions in Indonesia (KAP, BPKP, BPK) before
the questionnaires were distributed, to check the
clarity of the questions. The findings from the pilot
test indicated that every item comprising the
research constructs exhibited validity and reliability,
as evidenced by Cronbach's alpha coefficients
ranging from 0.75 to 0.85 for each construct.
4 Results and Discussion
The questionnaires that came in both from the form
and electronic form were 355, after going through
the eligibility selection there were 331 respondents'
answers that were worthy of being processed in the
study. Respondents in this study are Accountants or
Auditors who have been assigned to conduct
examinations or investigations at a government
institutions, SOEs, and private institutions.
Fig. 2: Respondent Profile Graph by Functional
Position
The profile of respondents in this study has
various auditor functional positions, consisting of
Junior Accountants and Senior Accountants for
auditors within the Public Accounting Firm (KAP).
Research respondents within BPKP and BPK auditor
functional positions consisted of First Auditor,
Junior Auditor, and Associate Auditor. Based on
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2024.21.53
Enny Susilowati Mardjono,
Entot Suhartono, Guruh Taufan Hariyadi
E-ISSN: 2224-2899
644
Volume 21, 2024
`
Figure 6, the most respondent profiles based on audit
functional positions within KAP are Junior
Accountants (110) compared to Senior Accountants
who number 42.
This is because in the implementation of audit
operations, many are handled by Junior
Accountants, while Senior Accountants act as
managers and supervisors of audit implementation.
In contrast to the profile of respondents in the BPKP
/ BPK environment, most respondents were those
with the functional position of Young Auditor as
many as 64 respondents followed by respondents
with the functional positions of First Audit and
Associate Audit each as many as 53 respondents.
This is because the characteristics of BPKP / BPK
respondents are government institutions, where
assignments to conduct examinations or audits are
mostly carried out by auditors who have middle to
upper-functional positions, such as Young Auditors
and Associate Auditors.
Figure 7 shows a graph of respondent profiles based
on experience in audits, it is known that the most
respondent profiles in this study over 10 years are 95
people or 29% of all respondents. This shows that
the most respondents have very high flight hours or
experience in conducting audit examination
activities with forensic accounting in detecting
fraud, especially added to the profile of respondents
who have experience between 7-10 years as many as
52 people or 16%, and respondents with experience
between 4-6 years as many as 62 people or 19% of
the total respondents.
Fig. 3: Graph of respondent profiles by Experience
in Audit
Fig. 4: Respondent Profile Graph Based on Audit
Engagement Frequency
Overall, respondents to this study answered that
they had received assignments or carried out
examination or audit activities with forensic
accounting and big data, but how often or frequency
they received these tasks can be seen in Figure 8.
The most respondents answered that they often get
audit tasks as many as 187 respondents or 56% both
within the KAP and BPKP / BPK audit units, while
respondents who answered very often as many as 19
respondents or 6% who only exist within the BPKP /
BPK only. This shows that research respondents in
general often conduct examination or audit activities
using forensic accounting and big data analytics to
detect fraud. However, some respondents rarely get
assignments or carry out examinations or audits,
namely as many as 122 people or 37% both within
BPKP / BPK and KAP.
1. Distribution of Respondents' Answers Regarding
the Awareness of Forensic Accounting (AFA)
Table 1. Respondents' answers related to Awareness
Forensic Accounting
Very Disagree
Disagree
Very Agree
AFA1
2
6.0%
0
0.0%
235
71.0%
94
28.4%
AFA2
0
0.0%
0
0.0%
231
69.8%
100
30.2%
AFA3
0
0.0%
2
0.6%
225
68.0%
104
31.4%
AFA4
0
0.0%
2
0.6%
245
74.0%
84
25.4%
AFA5
0
0.0%
2
0.6%
227
68.6%
102
30.8%
AFA6
0
0.0%
0
0.0%
253
76.4%
78
23.6%
AFA7
0
0.0%
0
0.0%
241
72.8%
90
27.2%
71.5%
28.1%
Based on Table 1, respondents generally
answered Agree on average as much as 71.5% and
Strongly Agree on average as much as 28.1%, the
majority of participants indicated a comprehensive
comprehension and acknowledgement of the
significance of forensic accounting in the
identification of fraudulent activities.
2. Distribution of Respondents' Answers Related to
Big Data Analytics Perception
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2024.21.53
Enny Susilowati Mardjono,
Entot Suhartono, Guruh Taufan Hariyadi
E-ISSN: 2224-2899
645
Volume 21, 2024
`
Table 1. Distribution of BDA respondents' answers
Indicators
Strongly
disagree
Disagree
Agree
Strongly
Agree
BDA1
0
0.0%
6
1.8%
268
81.0%
57
17.2%
BDA2
0
0.0%
6
1.8%
265
80.1%
60
18.1%
BDA3
0
0.0%
5
1.5%
274
82.8%
52
15.7%
BDA4
0
0.0%
4
1.2%
257
77.0%
70
21.1%
BDA5
0
0.0%
6
1.8%
277
83.7%
48
14.5%
BDA6
0
0.0%
3
0.9%
295
89.1%
33
10.0%
BDA7
0
0.0%
14
4.2%
278
84.0%
39
11.8%
BDA8
0
0.0%
6
1.8%
276
83.4%
49
14.8%
1.9%
82.6%
15.4%
Based on Table 2, the data distribution of big
data analytics among the respondents indicates a
comprehensive understanding of the function of big
data analytics in facilitating the application of
forensic accounting for fraud detection. This can be
seen from the answers of respondents who expressed
agreement as much as 82.6% and strongly agreed
15.4%, although there are a few respondents who are
still not familiar with the role of big data analytics
but only 1.9% answered disagree.
3. Distribution of Respondents' Answers to the
COSO framework
Table 2. Distribution of respondents' answers about
COSO
Very
Disagree
Disagree
Agree
Very Agree
COSO1
1
0.3%
13
3.9%
167
80.7%
50
15.1%
COSO2
1
0.3%
2
0.6%
251
75.8%
77
23.3%
COSO3
1
0.3%
4
1.2%
265
80.1%
61
18.4%
COSO4
1
0.3%
2
0.6%
283
85.5%
45
13.6%
COSO5
0
0.0%
4
1.2%
248
74.9%
79
23.9%
COSO6
1
0.3%
4
1.2%
243
73.4%
83
25.1%
COSO7
0
0.0%
3
0.9%
266
80.4%
62
18.7%
COSO8
1
0.3%
1
0.3%
244
73.7%
85
25.7%
COSO9
0
0.0%
3
0.9%
261
78.9%
67
20.2%
COSO10
1
0.3%
2
0.6%
265
80.1%
63
19.0%
0.2%
1.1%
78.4%
20.3%
In general, respondents already understand and
apply the COSO framework as an internal control
framework when implementing forensic accounting
to detect fraud. This can be seen from the answers of
respondents (Table 3) who agreed as much as 78.4%
and strongly agreed 20.3%, although there were a
few respondents who still did not understand the role
of the COSO framework but only 1.1% answered
disagree and strongly disagree as much as 0.2%.
4. Distribution of Respondents' Answers Related to
Intention of Forensic Accounting
Table 3. Distribution of respondents' answers
intention of forensic accounting
Very
Disagree
Disagree
Agree
Very Agree
IFA1
0
0.0%
2
0.6%
287
86.7%
42
12.7%
IFA2
0
0.0%
2
0.6%
273
82.5%
56
16.9%
IFA3
0
0.0%
2
0.6%
258
77.9%
71
21.5%
IFA4
1
0.3%
9
2.7%
272
82.2%
49
14.8%
0.1%
1.1%
82.3%
16.5%
With strong intentions and a general willingness
to use forensic accounting techniques to prevent and
detect fraud, the majority of respondents indicated.
This can be seen from the answers of respondents
(Table 4) who expressed agreement with as much as
82.3% and strongly agreed 16.5%, although there
were a few respondents who were still not willing
and intended to use forensic accounting techniques,
namely only 1.2% who answered disagree and
strongly disagree as much as 0.1%.
4.1 Measurement Model Evaluation Results
4.1.1 Internal Consistency and Reliability
Internal consistency and reliability are two important
indicators that are evaluated in research studies.
These indicators help to determine whether the
variables in the scale converge to a single latent
structure. Cronbach's Alpha and Composite
Reliability assess internal consistency and reliability.
Cronbach's Alpha over 0.7 indicates scale
dependability. In Table 5, all components meet the
0.7 criteria, suggesting discriminant validity for all
constructs. Referring to, [22], we also checked the
cross-factor loading values and the Heterotrait-
monotrait ratio (HTMT) criterion and the results
were also similar. Thus, the discriminant validity of
the constructs is assured.
Table 4. Reliability and convergence value
evaluation results
Variable
AVE
CR
Cronbach`s
Alpha
AFA
0.677
0.936
0.920
COSO
0.544
0.892
0.858
BDA
0.660
0.951
0.942
IFA
0.574
0.842
0.751
4.2 Convergent Validity
Convergent validity requires an average variance
extracted (AVE) of 0.5 for each latent variable,
according to, [29]. The latent variable can explain
over half of its indicators' variance. Latent variables
with AVEs below 0.5 should be removed from the
study model. The data processing results in Table 5
reveal that the AVE values for all scales are higher
than the minimum criterion of 0.5. Therefore, all
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2024.21.53
Enny Susilowati Mardjono,
Entot Suhartono, Guruh Taufan Hariyadi
E-ISSN: 2224-2899
646
Volume 21, 2024
`
nine research variables have strong convergent
validity.
In addition, the convergence value is calculated
using the observed variables' external load
coefficients and average extracted variance. The
observed variables' external stress coefficient must
be 0.708 or higher to be statistically significant. It
means the outcome variables will measure the
observed variable's dependability.
The results in Table 6 demonstrate that the external
load coefficient of the observed variables must be at
least 0.708 and statistically significant. Therefore,
the variables in the model satisfy the convergence
value criteria.
4.3 Evaluation of the Structural Model PLS-
SEM
According to, [23], study, the criteria considered to
evaluate the reference structure model. The PLS-
SEM method is used to evaluate different factors and
their importance. The evaluation of the measurement
model includes assessing the fit of the model,
measuring the R2 coefficient, and testing the
research hypothesis by the path coefficient and T-
value. The PLS-SEM bootstrapping sample size of
N = 5000 is used to test the consistent thesis. The
proposed hypotheses' p-values < 1%, 5%, and 10%
are considered statistically significant at 99%, 95%,
and 90% confidence. From the estimation results,
the evaluation results of the PLS-SEM structural
model are as follows:
4.3.1 Evaluate Collinearity in the Structural
Model
Table 3 shows all predictors' VIF (Variance Inflation
Factor) values (Outer VIF Values) below threshold
5. Therefore, collinearity between predictors is
acceptable in the structural model and can be
continued.
4.3.2 Model Fit Assessment
The Standard root mean square residual (SRMR) is a
measure of how well a model fits the data. It is
calculated by comparing the model's predictions
with the actual data. A value of 0 means a perfect fit,
and a value less than 0.05 is considered good.
However, for PLS path models, a value less than
0.08 is more appropriate. The results of Table 8
show that the SRMR values are almost less than or
equal to 0.1, which means that the model is suitable
for the dataset.
Table 5. External load coefficients of observed
variables
Variables
Indicator
Outer
Loading
Infor
matio
n
Awareness
Forensic
Accounting
can be used to analyze
financial statements (AF1)
0.781
Valid
very important to use in fraud
detection (AF2)
0.855
Valid
can be used to identify misuse
of assets (AF3)
0.796
Valid
can make financial data or
information more reliable and
accurate (AF4)
0.875
Valid
ensure compliance with
applicable laws and
regulations (AF5)
0.814
Valid
be able to evaluate the
corporate governance system
(AF6)
0.870
Valid
be able to evaluate internal
control (AF7)
0.760
Valid
Internal
Control
(COSO)
always use Forensic
Accounting techniques or
methods to detect fraud
(COS1)
0.678
Valid
senior management sets a
good example in supporting
and implementing internal
control (COS2)
0.898
Valid
identification and evaluation
of fraud risks that may occur
within the organization being
audited (COS3)
0.816
Valid
using a risk-based approach
in audit planning and
execution (COS4)
0.871
Valid
systematically evaluate the
design, implementation, and
effectiveness of internal
controls (COS5)
0.786
Valid
inspect and test information
systems (COS6)
0.798
Valid
ensure that information
regarding internal controls
and fraud-related policies is
communicated to company
management (COS7)
0.777
Valid
ensure that reports of fraud or
indications of fraud are
communicated appropriately
and in a timely manner
(COS8)
0.864
Valid
monitor and supervise
regularly the implementation
of internal control to detect
fraud (COS9)
0.842
Valid
routine internal checks to
evaluate the effectiveness of
internal controls related to
fraud detection (COS10)
0.898
Valid
Big Data
Analytics
can analyze data to detect
fraud in real-time (BG1)
0.846
Valid
provides advanced social
network analysis, data
visualization, and analysis
algorithms (BG2)
0.750
Valid
provides powerful and high-
quality and cost-effective
processing in Forensic
accounting to detect fraud
(BG3)
0.757
Valid
Provides high-speed data
processing (BG4)
0.766
Valid
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2024.21.53
Enny Susilowati Mardjono,
Entot Suhartono, Guruh Taufan Hariyadi
E-ISSN: 2224-2899
647
Volume 21, 2024
`
Variables
Indicator
Outer
Loading
Infor
matio
n
is very familiar used in
Forensic Accounting to detect
fraud (BG5)
0.661
Valid
make a significant
contribution to the use of
Forensic Accounting to detect
fraud (BG6)
0.656
Valid
confidence can improve the
reliability of forensic
accounting results for fraud
detection (BG8)
0.708
Valid
Intention to
Use
Forensic
Accounting
I am interested in using
Forensic Accounting to detect
fraud (NAF1)
0.783
Valid
I prioritize compliance needs
when using Forensic
Accounting to detect fraud
(NAF2)
0.629
Valid
I prioritize risk identification,
analysis, and control more
when using Forensic
Accounting to detect fraud
(NAF3)
0.858
Valid
I will always use Forensic
Accounting techniques or
methods to detect fraud
(NAF4)
0.743
Valid
4.3.3 Test the Predictive Level of the Structural
Model (R²)
The results in Table 7 show that the explanatory
level of the adjusted R² is quite strong, [22]. The
explanatory level of factors affecting business
performance (BP) is 63.5 (corresponding to an
adjusted R² of 62.5). Thus, to sum up, the
independent variable that explain 62.5% of the
research dependent variable is the business
performance of enterprises. This result is considered
quite well.
Table 6. Results of the VIF index of the model's
predictor
AFA
BDA
COSO
IFA
AFA
1.436
1.634
BDA
1.861
COSO
1.436
1.803
IFA
Table 7. Evaluation of model fit
Variable
Value
R Square
Big Data Analytics
0,463
Intentions Forensic Accounting
0,591
SRMR
0,095
NFI
0,613
According to the data presented in Table 8, the
R Square values for both Big Data Analytics
variables and Intentions in Forensic Accounting fall
above 0.25 and below 0.65. It suggests that the
structural model of this study aligns with medium
criteria, indicating a moderate level of influence of
the independent variable on the dependent variable.
The estimated outer model SRMR value is 0.095,
which means that the research model is included in
the Goodness of Fit criteria because it is in the range
between 0.08 to 0.10.
To see the impact of the independent variables on
the dependent variable in the model, we consider the
path coefficient and the coefficient f2. As described
in chapter 3, the coefficient f2 indicates the degree
of influence of the variable when removed from the
model. If f2 > 0, the independent variable has a
potentially suitable prediction for the dependent
variable. Corresponding to a coefficient f equal to
0.02 is the low probability, 0.15 is the medium
probability and 0.35 is the high probability. The
influence coefficient f2 is shown in Table 9
indicating that the explanatory level of the
independent variables for the dependent variable is
relatively low.
Table 8. Results of the coefficient f square
AFA
BDA
COSO
IFA
AFA
0.138
0.121
BDA
0.145
COSO
0.255
0.112
IFA
4.4 Results of Testing the Research
Hypothesis
This study's seven hypothesized causal relationships
were statistically verified to exhibit a positive
impact. The inner model testing results,
encompassing R-Square, path coefficient, and T-
statistics, were utilized for hypothesis testing. The
significance values between constructs are evident in
the T-statistics and P-Values obtained via
bootstrapping testing with the SmartPLS 4.0
package. A significance level of 5% (P = 0.05) and a
positive beta coefficient (T = 1.96) are the criteria
utilized in this study.
Table 10 shows that H1, H2, H3, H4, and H5, are
accepted (value of p < 0.05), i.e., variables
AW_FORAC, BDA, COSO all have a positive impact
on the Intentions of Forensic Accounting
(INT_FORAC). Table 11 shows that the research
hypotheses from H6 and H7 are accepted (p-value <
0.05), i.e., variables AW_FORAC through BDA have
a positive impact on Intentions of Forensic
Accounting (INT_FORAC). Also, variables COSO
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2024.21.53
Enny Susilowati Mardjono,
Entot Suhartono, Guruh Taufan Hariyadi
E-ISSN: 2224-2899
648
Volume 21, 2024
`
through BDA have a positive impact on Intentions of
Forensic Accounting (INT_FORAC).
Table 9. Results of model hypothesis testing
Estimate
Stdev
t-
Statistics
P
Value
R
Square
H1
AW_FORAC->
BDA
0.326
0.075
4.321
0.000
0.463
H2
AW_FORAC->
INT_ FORAC
0.284
0.070
4.074
0.000
H3
BDA ->
INT_FORAC
0.332
0.082
4.045
0.000
0.591
H4
COSO -> BDA
0.444
0.050
8.799
0.000
H5
COSO ->
INT_FORAC
0.287
0.068
4.258
0.000
Table 10. Specific indirect effect
Specific Indirect
Effects
P-Values
H6
AW_FORAC-> BDA->
INT_FORAC
0.108
0.000
H7
COSO -> BDA->
INT_FORAC
0.148
0.000
Based on the indirect effects table in Table 11, it can
be concluded that:
1. The indirect influence of AW_FORAC on
INT_FORAC through BDA is 0.108 which
means that if AW_FORAC increases by one unit
then INT_FORAC can increase indirectly
through by BDA by 10.8%. This influence is
positive.
2. The indirect influence of COSO on INT_FORAC
through BDA is 0.148 which means that if
COSO increases by one unit then INT_FORAC
can increase indirectly through BDA by 14.8%.
This influence is positive.
Table 11. Total effect
AW_FOR
AC
BIG DATA_
ANALY
COSO
INTENT_
FORACC
AW_FORACC
0.326
0.393
BIGDA_ANALY
0.332
COSO
0.444
0.435
INTENT_FORAC
Based on Table 12, it can be concluded as follows:
1. The effect of AW_FORAC on INT_FORAC
through BDA is 0.393 which means that if one-
unit increases, INTENT_FORAC can increase
directly and indirectly through BDA by 39.3%.
This influence is positive.
2. The effect of COSO on INT_FORAC through
BDA is 0.435 which means that if COSO
increases by one unit then INT_FORAC can
increase directly and indirectly through BDA by
43.5%. This influence is positive.
5 Discussion
Hypothesis 1 testing proves that Awareness of
Forensic Accounting influences Big Data. Increasing
awareness of the use of forensic accounting will
cause auditors to need complex and integrated big
data.
In the validity test table, the AF4 indicator
exhibits the highest outer loading or correlation
value of 0.875, indicating that awareness-based
forensic accounting has the potential to enhance the
dependability and precision of financial data or
information. Auditors and examiners feel the
importance of using Big Data to work quickly and
find more fraud and can make financial information
more reliable and accurate. Auditors use big data
techniques when dealing with the complexity of
commercial transactions to uncover fraudulent
transactions. Forensic accountants can apply various
advanced anti-fraud techniques by increasing the use
of big data such as data visualization, predictive
analysis, behavioral analysis, geo-spatial analysis to
identify fraudulent activities. The results of this
research support, [14], and, [18], which states that
forensic accounting awareness can increase the use
of big data technology to detect fraud.
Testing hypothesis 2 proves that Awareness
Forensic Accounting (AW_FORAC) influences the
Intention to use Forensic Accounting
(INT_FORAC). to detect fraud. Practitioners who
have a higher level of forensic accounting awareness
will be more willing to apply forensic accounting for
fraud detection.
In the validity test table, it can be seen that the
AF4 indicator has the highest outer loading or
correlation value, namely 0.875, which shows that
Awareness Forensic Accounting can make financial
data or information more reliable and accurate
because examiners or auditors prioritize risk
identification and analysis to detect fraud. The
results of this research support, [17], and, [18],
which states that awareness of forensic accounting
plays an important role in increasing the intention to
use forensic accounting in uncovering complicated
fraud through audits, accounting and investigative
skills.
Testing hypothesis 3 establishes the influence of
Big Data utilization on the Intentions of Forensic
Accounting. The application of big data techniques
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2024.21.53
Enny Susilowati Mardjono,
Entot Suhartono, Guruh Taufan Hariyadi
E-ISSN: 2224-2899
649
Volume 21, 2024
`
has the potential to enhance the quality and
effectiveness of financial report audits and
contribute to the advancement of forensic
accounting methods for detecting potential fraud
within financial data.
The validity test table indicates that the BG1
indicator exhibits the highest outer loading or
correlation value of 0.846, signifying the capability
of Big Data to analyze data and identify fraud in real
time. Therefore, using big data for real-time fraud
detection can further amplify the intention to employ
forensic accounting. By using big data models that
can identify financial fraud more precisely, auditors
have a higher intention to use forensic accounting to
identify risk areas that require further investigation.
The results of this research support, [19], which
states that big data models can be used efficiently if
auditors increase their interest in collaborating on
the use of forensic accounting with data analysts.
Testing hypothesis 4 proves that COSO influences
the use of Big Data. The COSO framework for risk
detection requires big data to predict and help
businesses make important decisions. In the validity
test table, it can be seen that the COS2 indicator has
the highest outer loading or correlation value of
0.898, which shows that senior management
provides a good example in the implementation of
internal control so that with good internal control
support it allows auditors to use specific big data to
detect and prevent the risk of fraud.
In order to implement risk management
strategies, financial institutions acquire real-time
insight into potential hazards by utilizing big data
technology that integrates risk assessment with
predictive algorithms for big data analysis. The
results of this research support, [20] which states
that risk management can be more effective by using
the COSO framework and big data.
Testing hypothesis 5 proves that COSO
influences the Intentions of Forensic Accounting.
Internal control components (COSO), including risk
assessment, control activities, information and
communication, as well as monitoring activities,
have been proven to be effective in preventing fraud.
So, the use of COSO will increase the interest of
auditors or examiners in using forensic accounting to
prevent and detect fraud, [24].
In the validity test table, it can be seen that the
COS2 indicator has the highest outer loading or
correlation value of 0.898, which shows that senior
management at KAP, BPKP, BPK provides an
example of implementing effective internal control
thereby increasing the interest of auditors or
examiners in using forensic accounting in finding
more fraud. The results of this research support,
[20], [18], and, [25], which prove the effectiveness
of the internal control framework in preventing work
fraud in local government organizations.
Hypothesis H6 testing proves access to
Awareness of Forensic Accounting (AW_FORAC)
through Big Data Analysis (BDA) has the most
significant influence on intentions of Forensic
Accounting (INT_FORAC). This result is similar to
the studies of, [18], [26].
BDA can be used to increase the effectiveness of
inspections and investigations in forensic
accounting. Big data with its volume, velocity, and
variety of data offers many sources of evidence.
Forensic Accountants with Big Data can extract,
visualize, and analyze data originating from various
sources. Apart from that, auditors can also identify
suspicious patterns and trends using Big Data.
Auditors utilize Entity resolution algorithms and
data mining databases to detect hidden relationships,
forged addresses, conflicts of interest, and false
identities.
However, most KAPs still haven't maximized the
use of Big Data due to the limited availability of the
data being examined. Meanwhile, BPK and BPKP
have used Big Data a lot because the data is
centralized so data availability and data access can
be obtained easily. In addition, there are government
regulations and The Financial Services Authority
(OJK) that require related companies and audited
clients to disclose and provide their data if requested
by the BPK and BPKP. The Financial Services
Authority (OJK) is a trustworthy monitoring
institution that oversees the financial services
industry in Indonesia.
Hypothesis H7 testing proves that the indirect
influence of COSO through Big Data Analysis
(BDA) is the most critical factor affecting the
intentions of Forensic Accounting (INT_FORAC).
This result is similar to the studies of, [19], [27],
[28], [29].
By incorporating forensic accounting practices into
the COSO framework, organizations can strengthen
their internal controls and improve their financial
reporting processes. BDA can enhance the efficacy
of internal control by monitoring and analyzing
financial transactions in real-time. Auditors can use
Big Data to evaluate and analyze large amounts of
data so that the auditor can identify data patterns and
anomalies that can be identified as fraudulent
activity. The role of BPK and BPKP in internal
control with COSO to use forensic accounting is
very large.
The Constitution of the Republic of Indonesia
mandates the roles of the BPK and BPKP. The
BPK's primary role is to audit and supervise the
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2024.21.53
Enny Susilowati Mardjono,
Entot Suhartono, Guruh Taufan Hariyadi
E-ISSN: 2224-2899
650
Volume 21, 2024
`
local and central government finance, acting as
national external auditors. The BPKP's role is to
perform audits, evaluations, reviews, and monitoring
of asset utilization and financial planning belonging
to local and central governments, concerned with
financial and operational terms, and focused on
performance measurement. The BPKP and internal
control systems are intricately intertwined in
strategy formulation and implementation. Further
provisions regarding the BPK are regulated by law.
The BPK is not responsible for financial
management, but it plays a role as an external
auditor to improve public financial management.
The BPK's roles and functions have strengthened
since the third amendment of the 1945 Constitution
in 2001. The Indonesian Constitution of 1945 has
been reinstated in 1959 and revised in 2002. The
BPK is one of the main actors in the Indonesian
health system, along with the Badan Pemeriksa
Keuangan (Supreme Audit Agency) or BPK
6 Managerial Policy Implications
From the research results, some policy implications
are proposed for state agencies. This research found
interesting evidence. The empirical evidence
presented in our study assists scholars and
consumers in assessing the level of complexity
demonstrated by internal control frameworks
designed to prevent fraud. Internal controls are not
universally effective in preventing all forms of fraud
within firms, particularly in instances involving
corruption.
The organization takes the existence of a
conspiracy orchestrated by a powerful group
extremely seriously; consequently, internal control
will be rendered less effective, [29] [30].
Meanwhile, the organization's internal control makes
it very easy to prevent fraud related to technical
tasks. So, it is necessary to design additional
measures from existing COSO guidelines to
appropriately address the risk of corruption fraud,
[31] .
This has theoretical implications for managerial
organizations. There are several alternative
mechanisms for implementing controls related to the
use of Big Data such as "regular staff rotation" and
"fraud awareness training" as well as increasing the
use of big data to get evidence findings quickly and
in real-time. By implementing this measure,
businesses can establish a highly ethical
environment and gain control over financial and
human resource deception. Some elements facilitate
the execution of internal control, such as
establishing a mechanism for reporting misconduct.
Auditors can highlight the mechanisms for
implementing controls and Big Data using on their
audited clients.
7 Conclusion
The study identifies key attributes influencing the
intention to detect fraud in forensic accounting. Big
Data and awareness of forensic accounting play
pivotal roles. Seven hypotheses were posited,
establishing the significant impact of Forensic
Accounting Awareness on Big Data Analysis, the
substantial influence of Big Data Analysis on
Intentions in Forensic Accounting, the impactful
relationship between COSO and Big Data Analysis,
the consequential effect of COSO on Intentions in
Forensic Accounting, and the mediating role of Big
Data Analysis in the relationship between COSO
and forensic accounting intentions. Statistical
support is found for the seventh hypothesis.
This study provides empirical evidence that a
resilient internal control system reduces the
likelihood of fraudulent activities. Effective internal
controls require an organization to comprehensively
understand the most susceptible risks and the
appropriate course of action to address fraudulent
activities. It is imperative for organizations to
methodically integrate components for identifying,
assessing, and responding to fraud risk.
This approach facilitates the incorporation of
anti-fraud measures into risk management initiatives,
establishing a systematic approach to document
acquired knowledge for guiding future endeavors in
detection or mitigation. Big Data is an indispensable
tool for the continuous and timely accumulation of
evidence and data to analyze trends over time to
identify misconduct instances.
8 Limitation and Future Research
The limitation of this research is that the scope or
object of research is auditors in Indonesia, so it may
not necessarily be applicable in other countries with
different cultures and regulations. In addition, this
research has not been in-depth in identifying barriers
and opportunities for using forensic accounting, big
data, and the COSO framework to detect fraud. It is
necessary to conduct in-depth follow-up research
with qualitative methods from the results of auditor
interviews.
Based on the results of this study, the researcher
provides suggestions for future research, such as (1)
Conduct further research to delve into auditors'
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2024.21.53
Enny Susilowati Mardjono,
Entot Suhartono, Guruh Taufan Hariyadi
E-ISSN: 2224-2899
651
Volume 21, 2024
`
understanding of the use of forensic accounting in
fraud detection. Focus on how this awareness can be
enhanced and applied more effectively in audit
practices. (2) Deepen the role of big data analysis
(BDA) in supporting forensic accounting activities.
Research can explore specific BDA techniques that
are most effective in detecting fraud indications and
identify barriers and opportunities that may arise in
its implementation. (3) Explore the impact of
implementing the Committee of Sponsoring
Organizations (COSO) model in preventing and
detecting fraud in other industry sectors. Focus on
different sectors to gain further insights into the
effectiveness of the COSO model outside the
banking sector, such as manufacturing or services
industries. (4) Build an integrated framework that
combines aspects of forensic accounting, big data
analysis, and the COSO model. Research can
evaluate how this integration can be applied by
internal and external auditors to enhance efficiency
in fraud detection.
Acknowledgements:
This research is funded by Indonesian Ministry of
Education, Culture, Research and Technology,
Indonesian Directorate General of Higher
Education, Research and Technology.
References
[1] N. Christian, “Behavioral strategy analysis
using the fraud diamond theory approach to
detecting corporate fraud in Indonesia,”
International Journal of Business and
Management Invention (IJBMI), vol. 9, no.
4, pp. 66–74, 2020.
[2] C. C. Yu and H. W. Huang, “Audit Office’s
unused capacity and audit quality,” Asia
Pacific Management Review, vol. 28, no. 2,
pp. 146–162, Jun.2023,
doi:10.1016/j.apmrv.2022.07.005.
[3] ACFE, “Association of Certified Fraud
Examiners (ACFE). ACFE Report,” 2022.
[4] ACFE, “Association of Certified Fraud
Examiners (ACFE) Indonesia. Survei Fraud
Indonesia,” 2019.
[5] S. Alayli, “The Impact of Internal Control
Practices on Fraud Prevention: The Case of
Lebanese Small-Medium Enterprises,”
European Journal of Business and
Management Research, vol. 7, no. 5, pp.
141–147, Oct. 2022, doi:
10.24018/ejbmr.2022.7.5.1671.
[6] A. Gepp, M. K. Linnenluecke, T. J. O’Neill,
and T. Smith, “Big data techniques in
auditing research and practice: Current trends
and future opportunities,” Journal of
Accounting Literature, vol. 40, pp. 102–115,
Jun. 2018, doi: 10.1016/j.acclit.2017.05.003.
[7] F. De Santis and G. D’Onza, “Big data and
data analytics in auditing: in search of
legitimacy,” Meditari Accountancy
Research, vol. 29, no. 5, pp. 1088–1112,
2020, doi: 10.1108/MEDAR-03-2020-0838.
[8] Z. Rezaee and J. Wang, “Relevance of big
data to forensic accounting practice and
education,” Managerial Auditing Journal,
vol. 34, no. 3, pp. 268–288, May 2019, doi:
10.1108/MAJ-08-2017-1633.
[9] Y. A. A. Hezam, L. Anthonysamy, and S. D.
K. Suppiah, “Big Data Analytics and
Auditing: A Review and Synthesis of
Literature,” Emerging Science Journal, vol.
7, no. 2. Ital Publication, pp. 629–642, Apr.
01, 2023. doi: 10.28991/ESJ-2023-07-02-
023.
[10] L. H. Bambang, Ameliya R., Nada A., and
Adeliya Y.B, “The Impact of Big Data
Analytics and Forensic Audit in Fraud
Detection,” in the 12th International
Workshop on Computer Science and
Engineering (WCSE 2022), 2022, pp. 67–71.
[11] A. Alfiar and Jaeni, "Pengaruh Audit
Forensik, Audit Investigasi, Kompetensi
Auditor, Profesionalisme, Dan Kecerdasan
Spiritual Terhadap Pencegahan Fraud (Studi
Pada BPKP Perwakilan Jawa Tengah),"
Jurnal Ilmiah Komputerisasi Akuntansi, vol.
15, no. 1, pp. 159-157, Jul. 2022,
https://doi.org/10.51903/kompak.v15i1.628.
[12] P. Mittal, A. Kaur, and P. K. Gupta, “The
Mediating Role of Big Data to influence
Practitioner to use forensic Accounting for
Fraud Detection,” European Journal of
Business Science and Technology, vol. 7, no.
1, pp. 47–57, 2021.
[13] Abdul Aziz A, Abdul Rahman, and Othman
Hel Ajmi Al-Dhaimesh, “The effect of
applying COSO-ERM model on reducing
fraudulent financial reporting of commercial
banks in Jordan,” Banks and Bank Systems,
vol. 13, no. 2, pp. 107–115, 2018.
[14] Y. S. Chen, E. S. Mardjono, and Y. F. Yang,
“MASs, alliance, and performance: an
evidence of SOX effects,” Managerial
Auditing Journal, vol. 37, no. 5, pp. 521–
541, Apr. 2022, doi: 10.1108/MAJ-05-2021-
3164.
[15] B. İ. Kılıç, “The Effects Of Big Data On
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2024.21.53
Enny Susilowati Mardjono,
Entot Suhartono, Guruh Taufan Hariyadi
E-ISSN: 2224-2899
652
Volume 21, 2024
`
Forensic Accounting Practices And
Education,” in Contemporary Studies in
Economic and Financial Analysis, vol. 102,
Emerald Group Holdings Ltd., 2020, pp. 1–
10. doi: 10.1108/S1569-
375920200000102005.
[16] A. Maulidi and J. Ansell, “Corruption as
distinct crime: the need to reconceptualize
internal control on controlling bureaucratic
occupational fraud,” J Financ Crime, vol. 29,
no. 2, pp. 680–700, Mar. 2022, doi:
10.1108/JFC-04-2021-0100.
[17] O. Esther Akinbowale, H. Eckart
Klingelhöfer, and M. Fekadu Zerihun, “The
Integration of Forensic Accounting and the
Management Control System as Tools for
Combating Cyberfraud,” Academy of
Accounting and Financial Studies, vol. 25,
no. 2, Mar. 2021, [Online].
https://www.researchgate.net/publication/350
496079.
[18] S. Saluja, A. Aggarwal, and A. Mittal,
“Understanding the fraud theories and
advancing with integrity model,” J Financ
Crime, 2021, doi: 10.1108/JFC-07-2021-
0163.
[19] D. H. Akbulut and I. Kaya, “Big data
analytics in financial reporting and
accounting,” Pressacademia, vol. 7, no. 1,
pp. 256–259, Sep. 2018, doi:
10.17261/pressacademia.2018.892.
[20] A. Nawawi and A. S. A. P. Salin, “Internal
control and employees’ occupational fraud
on expenditure claims,” J Financ Crime, vol.
25, no. 3, pp. 891–906, Jul. 2018, doi:
10.1108/JFC-07-2017-0067.
[21] Y. S. Chen, E. S. Mardjono, and Y. F. Yang,
“Competition and sustainability: Evidence
from a professional service organization,”
Sustainability (Switzerland), vol. 12, no. 18,
pp. 1–15, Sep. 2020, doi:
10.3390/su12187266.
[22] J. F. Hair, W. C. Black, B. J. Babin, and R.
E. Anderson, “Multivariate Data Analysis
Eighth Edition,” 2019.
[Online].Available:www.cengage.com/highe
red
[23] J. F. Hair Jr., R. E. Anderson, R. L. Tatham,
and W. C. Black, Multivariate Data
Analysis, 5th Editio. Prentice Hall, 1998.
[24] J. Lanz, “Enterprise Technology Risk in a
New COSO ERM World - The CPA
Journal.”, [Online].
https://www.cpajournal.com/2018/06/19/ente
rprise-technology-risk-in-a-new-coso-erm-
world/ (Accessed Date: September07, 2023).
[25] N. Shonhadji and A. Maulidi, “Is it suitable
for your local governments? A contingency
theory-based analysis on the use of internal
control in thwarting white-collar crime,” J
Financ Crime, vol. 29, no. 2, pp. 770–786,
Mar. 2022, doi: 10.1108/JFC-10-2019-
0128/FULL/XML.
[26] F. J. M. Arboleda, J. A. Guzman-Luna, and I.
D. Torres, “Fraud detection-oriented
operators in a data warehouse based on
forensic accounting techniques,” Computer
Fraud & Security, vol. 2018, no. 10, pp. 13–
19, Oct. 2018, doi: 10.1016/S1361-
3723(18)30098-8.
[27] W. A. Günther, M. H. Rezazade Mehrizi, M.
Huysman, and F. Feldberg, “Debating big
data: A literature review on realizing value
from big data,” The Journal of Strategic
Information Systems, vol. 26, no. 3, pp. 191–
209, Sep. 2017, doi:
10.1016/J.JSIS.2017.07.003.
[28] Y. Yan, “Management Accounting in The
Era of Big Data,” in Advances in Economics.
Business and Management Research,
Proceedings of the 2022 7th International
Conference on Financial Innovation and
Economic Development (ICFIED 2022),
2022, pp. 793–798.
[29] A. Mwange and M. Chansa, “Emerging
Issues in Accounting: A Theoretical
Review,” journal of accounting finance and
auditing studies (JAFAS), Oct. 2022, doi:
10.32602/jafas.2022.032.
[30] D. L. Crumbley, L. E. Heitger, and G. S.
Smith, Forensic and Investigative
Accounting (7th Edition), 7th ed. CCH Inc,
2015.
[31] I. B. A. Yasa, I Ketut Sukayasa, and Ni
Made Mega Abdi Utami, “Organizational
culture moderates the effect of bystander
effect and internal control on accounting
fraud trends in village credit institutions in
Jembrana Regency,” International Journal of
Research in Business and Social Science
(2147- 4478), vol. 11, no. 7, pp.210–
217,Nov.2022,doi:
10.20525/ijrbs.v11i7.2097.
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2024.21.53
Enny Susilowati Mardjono,
Entot Suhartono, Guruh Taufan Hariyadi
E-ISSN: 2224-2899
653
Volume 21, 2024
`
APPENDIX
Fig. 1: Roadmap for Research Variables: Big Data, Internal control, and Interest in using Forensic
Accounting for Fraud Detection
Fig. 4: Full model
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2024.21.53
Enny Susilowati Mardjono,
Entot Suhartono, Guruh Taufan Hariyadi
E-ISSN: 2224-2899
654
Volume 21, 2024
`
Contribution of Individual Authors to the
Creation of a Scientific Article (Ghostwriting
Policy)
- Enny Susilowati Mardjono: Conceptualization,
Investigation, Supervision, Writing – original
draft, Formal analysis, Review & editing.
- Entot Suhartono: Data curation, Methodology,
Funding acquisition, Writing – review & editing.
- Guruh Taufan: Software, Writing – review &
editing.
Sources of Funding for Research Presented in a
Scientific Article or Scientific Article Itself
This research is funded by Indonesian Ministry of
Education, Culture, Research and Technology,
Indonesian Directorate General of Higher
Education, Research and Technology.
Conflict of Interest
The authors have no conflicts of interest to declare.
Creative Commons Attribution License 4.0
(Attribution 4.0 International, CC BY 4.0)
This article is published under the terms of the
Creative Commons Attribution License 4.0
https://creativecommons.org/licenses/by/4.0/deed.en
_US
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2024.21.53
Enny Susilowati Mardjono,
Entot Suhartono, Guruh Taufan Hariyadi
E-ISSN: 2224-2899
655
Volume 21, 2024