Effect of Big Data

Analytics on Audit
An exploratory qualitative study of
data analytics on auditors’ skills and
competence, perception of
professional judgment, audit
efficiency and audit quality

Hamadou Kandeh, Mohamad Alsahli

Department of Business Administration

Master's Program in Accounting
Master's Thesis in Business Administration III, 30 Credits, Spring 2020
Supervisor: Tobias Svanström
[This page was intentionally left blank]

I
 Abstract

Purpose: The primary goal of this thesis is to provide a deeper understanding of how big
data affect professional judgment, audit efficiency, and perceived audit quality. It also aims
to explore the effect of Big Data Analytics (BDA) on the skills and competence required
by auditors to perform an audit in a big data environment.

Theoretical perspectives: Theoretical concepts base on previous research and publications

by practitioners and regulators on BDA, professional judgment, audit efficiency and audit
quality. Literature was used to derive the research gap and research questions.

Methodology: A qualitative method base exploratory approach. A literature review was

conducted to uncover areas of interest that require more research. The effect of data
analytics on the audit was identified as a potential area for research; a focus on audit
quality was chosen, including key factors that contribute to overall audit quality. The
research is based on semi-structured interviews with auditors from big four audit firms in
Sweden.

Empirical foundation: Empirical evidence was generated through an interview with

seven auditors at different levels of the professional hierarchy. Empirical data was
analyzed using a thematic data analysis approach.

Conclusions: The findings of this research show that using BDA in the audit methodology
affect the required skills and competence by auditors to carry out audit engagement
activities. More IT related skills and knowledge gaining prominent in the audit field.
Implementing data analytics will not be efficient in the early stage but will save time as
auditors become more familiar with the tools. Data analytics improve audit quality.
Auditors use analytics to gain more insight into the client’s business and communicate
such insights to clients. It was found that data analytics generate fact-based audit evidence.
The visualization ability enables auditors to visualize and analyze audit evidence to guide
their professional judgment and decision making.

Key words: Big data, Data analytics, Auditors skills and competence, Audit process, Audit
efficiency, Audit quality and Professional judgment.

II
[This page was intentionally left black]

III
Acknowledgment

We would like to express our gratitude to Umea School of Business, Economics, and
Statistics for providing us with such a wonderful enabling environment to flourish to our
supervisor Tobias Svanström for supporting us throughout the whole process, especially by
sharing critical feedback with us.
To our participants, we greatly value the time and effort you devoted to participating in
this study in spite of your busy schedules. Your inputs were critically valuable to our
study.
We would like to express sincere thanks to our families and classmates for the
unconditional support and encouragement during our study.
Finally, a special thanks to the Swedish Institute for the support provided to Hamadou’s
education.

UMEÅ 25-05-2020
……………………… ………………………
Hamadou Kandeh Mohamad Alsahli

IV
[This page was intentionally left blank]

V
1. Table of Contents
1. Table of Contents.......................................................................................................................VI
1. INTRODUCTION........................................................................................................................... 1
1.1. Background ......................................................................................................................... 1
1.2. Problematization and research gap .................................................................................... 4
1.3. Research Questions ............................................................................................................ 5
1.4. Research purpose ............................................................................................................... 5
1.5. Delimitation ........................................................................................................................ 6
1.6. Structure of the paper ........................................................................................................ 6
2. THEORETICAL FRAMEWORK ....................................................................................................... 7
2.1. Big Data and Big Data Analytics .......................................................................................... 7
2.1.1 Big Data .............................................................................................................................. 7
2.1.2 Big Data as Audit Evidence ................................................................................................ 9
2.1.3 Big Data Analytics ............................................................................................................ 10
2.1.4 Overview of previous literature ....................................................................................... 12
2.2. Legitimacy Theory ............................................................................................................. 13
2.3. Legitimacy in Audit ........................................................................................................... 14
2.4. Audit processes ................................................................................................................. 15
2.4.1 Pre-audit engagements ................................................................................................... 16
2.4.2 Audit plan......................................................................................................................... 16
2.4.3 Assess internal control over financial reporting .............................................................. 17
2.4.4 Perform a substantive test on financial transactions ...................................................... 17
2.5.5 Report the findings .......................................................................................................... 18
2.5. Competence and skills of auditors.................................................................................... 18
2.5.1 Competence ..................................................................................................................... 18
2.5.2 Skills ................................................................................................................................. 19
2.5.3 Effect of Advanced Technology on Skills and Competencies .......................................... 19
2.6. Professional Judgment...................................................................................................... 20
2.7. Audit efficiency ................................................................................................................. 22
2.8. Audit Quality ..................................................................................................................... 23
3. METHODOLOGY ........................................................................................................................ 27

VI
 3.1. Theoretical Methodology ................................................................................................. 27
3.1.2. Research philosophy ....................................................................................................... 28
3.1.3. Research Approach ......................................................................................................... 30
3.1.4. Research Method............................................................................................................ 31
3.1.5. Research Design .............................................................................................................. 32
3.2. Practical Methodology ...................................................................................................... 33
3.2.1. Data Collection................................................................................................................ 33
3.2.2. Sampling ......................................................................................................................... 35
3.2.3 Participant Selection ........................................................................................................ 36
3.2.4. Interview Guide and Interview Process .......................................................................... 37
3.2.5 Data Analysis .................................................................................................................... 40
3.2.6 Literature Search ............................................................................................................. 42
4. EMPIRICAL RESULTS AND FINDINGS ......................................................................................... 44
4.1. General bachground ......................................................................................................... 44
4.2. Skills and competences ..................................................................................................... 46
4.3. Professional Judgment...................................................................................................... 49
4.4. Audit Efficiency ................................................................................................................. 53
4.5. Audit Quality ..................................................................................................................... 55
5. ANALYSIS AND DISCUSSION ...................................................................................................... 61
5.1. Skills and competence ...................................................................................................... 61
5.1.1 Skill sets required to perform an audit in a BDA environment........................................ 61
5.1.2 The need for a data analytics specialist ........................................................................... 62
5.1.3 Potential barriers to Future Auditors .............................................................................. 63
5.1.4 The need to redesign the academic curriculum for Accountants ................................... 64
5.2. Professional judgment ...................................................................................................... 65
5.2.1 Understanding the client Business environment and industry ....................................... 65
5.2.2 Quality of evidence .......................................................................................................... 65
5.2.3 Professional experience ................................................................................................... 66
5.3. Audit Efficiency ................................................................................................................. 67
5.4. Audit Quality ..................................................................................................................... 68
5.4.1 Input ................................................................................................................................. 68
5.4.2 Process ............................................................................................................................. 68

VII
 5.4.3 Output .............................................................................................................................. 70
5.4.4 Interaction ....................................................................................................................... 70
6. CONCLUSION............................................................................................................................. 72
6.1. General Conclusion ........................................................................................................... 72
6.2. Contribution & Implication of research ............................................................................ 73
6.2.1 Theoretical contribution .................................................................................................. 73
6.2.2 Practical implication......................................................................................................... 73
6.2.3 Societal implication.......................................................................................................... 73
6.3. Limitations ........................................................................................................................ 74
6.4. Avenues for future research ............................................................................................. 74
6.5. Quality Criteria .................................................................................................................. 75
6.6. Ethical consideration ........................................................................................................ 77
7. Reference: ................................................................................................................................. 79
8. Appendix ................................................................................................................................... 86
8.1. Appendix 1: Email to prospective participants ................................................................. 86
8.2. Appendix 2: Interview guide ............................................................................................. 87
8.3. Appendix 3: Information sheet ......................................................................................... 89
8.4. Appendix 4: List of contacted big four Forms for interview ............................................. 90

Table of figures
Figure 1: Connolly’s’ (2012) definition of Big Data ............................................................................. 8
Figure 2: Paths to expand data analytics in financial statement audits ........................................... 11
Figure 3: Phases of the audit process ............................................................................................... 16
Figure 4: IAASBs’ Framework for Audit Quality ................................................................................ 24
Figure 5: Data analytics – Impact on audit quality ........................................................................... 69

Table of tables
Table 1: Overview of previous literature .......................................................................................... 12
Table 2: Summary of philosophical assumptions ............................................................................. 33
Table 3: Participants interview details.............................................................................................. 40
Table 4: Phases of the Thematic Analysis (Braun & Clarke, 2006, p. 87) ......................................... 42
Table 5: Participants background ..................................................................................................... 45

VIII
1. INTRODUCTION
In the first chapter of this manuscript, the authors introduce the choice of the research
topic and research purpose as well as the overall objective of the thesis. The first chapter
provides an overview of the use of technology in audit and the demand for audit quality.
Secondly, the authors dive into the theoretical background to motivate the relevance of the
topic as a current trend in business administration. Thirdly, the authors identified the
research gap and explained how the research would contribute to filling the gap with the
research questions. Finally, they outline the research purpose, delimitation of the
research, and structure of this paper.

1.1. Background

The auditing profession has gained much attention in the past decades, particularly after
Enron, WorldCom, and other auditing scandals (Alles, 2015, p. 440). An audit involves
examining both financial and non-financial records of companies to establish patterns of
events. Regulations are being reviewed to adequately guide audit processes and procedures
in gathering sufficient evidence. This process is subjective upon the professional judgment
of the auditors (Adrian & Viorica, 2015, p. 141). Nowadays, audit engagement involves
several stages, such as planning the audit, conducting the client risk assessment,
performing internal control tests, collecting evidence, and sharing information to
appropriate stakeholders at the various stages of the audit process (Appelbaum et al., 2017,
pp. 3- 4). The advancement in technology played and continues to play a crucial role in the
evolution of the audit profession. The advent of technology such as digitalization, artificial
intelligence, and data analytics drastically changes the methods employed in audit
processes geared towards improving engagement output (ICAEW, 2016, p. 6). The
continuous development and incorporation of technology into audit methods call for
auditors to widen the scope of their knowledge related to using advanced information
technology in different stages of the audit process (Salijeni et al., 2019, pp. 98-100; Kokina
& Davenport, 2017, p. 115).
For long, auditors have relied on the application of data analytics in the execution of audit
engagements. For the past two to three decades, auditors are making the best use of the
available technology to gain a better understanding of audit clients and to conduct a risk
assessment. More specifically, they gain an overview of industrial risk, and substantive
procedures to gather sufficient evidence to issue an informed opinion about management
assertion of financial statements (Salijeni et al., 2019, p. 98). Data analytics simply
determine the processing of raw data to generate valuable information for decision-making
purposes. For long, auditors have used data analytics mainly to improve audit quality and
efficiency (Cao, et al., 2015, p. 426).
In recent years, the world is witnessing a growing impact of Big Data (hereafter BD) in
business. According to Mckinsey Institute (2011), BD has a significant potential to change
the world and not just the business (Alles, 2015, p. 439). Many organizations have invested
considerable resources in BD to generate value and increase profit. For instance, by
predicting sales numbers and employing new marketing strategies (Alles, 2015, pp. 439-

1
440). BD is expected to generate valuable economic benefits for audit firms and audit
clients (Salijrni, 2019, p. 11). According to Salijeni (2019, p. 11) it is anticipated that using
of BD will generate around £322 billion in UK by 2020 and €425 billion cost reduction in
the manufacturing industry in Europe by 2020. Richins et al. (2017) argued that auditing
firms should adopt BD proactively in auditing procedures to exploit its potential benefits.
In this regard, The Institute of Chartered Accountants of England and Wales (ICAE&W)
further states “it is critical for the audit profession to keep pace with these changes and be
proactive in understanding how new technology trends can transform the audit
approaches” (ICAE&W cited by Joshi & Marthandan, 2018, p. 2).
BD consists of various forms of traditional financial and non- financial data such as email
correspondence, records of telephone conversations, text messages from private and
business-related, social media platforms, news outlets, and other publications (Appelbaum
et al., 2017, p. 4). Furthermore, BD can be described by 4Vs: volume, variety, velocity,
and veracity. Volume refers to a large data set. Variety refers to different sources of
collecting data. Velocity is how rapidly the data is changing. Veracity is the relevance and
integrity of data that changes at a very high level (Vasarhelyi et al., 2015, p. 382; Zhang et
al., 2015, p. 470; Yoon et al., 2015, p. 432). Many organizations are increasingly investing
in the collection and storage of mass volumes of data in a variety of forms, and the growth
is increasing rapidly (Cao et al., 2015, p. 426).
The growing trend in technology development makes it possible to record, store, and
measure all kind of data (Cao et al., 2015, p. 423). However, the challenge auditors face is
filtering the vast amount of data to obtain relevant and manageable information needed in
audit procedures (Brown-liburd & Vasarhelyi, 2015, p. 3). Therefore, big four audit firms
have responded to new technological developments by introducing Big Data Analytics
(hereafter BDA). BDA can be defined as: ’’the science and art of discovering and
analyzing patterns, identifying anomalies and extracting other useful information in data
underlying or related to the subject matter of an audit through analysis, modeling, and
visualization for the purpose of planning or performing the audit’’ (AICPA 2014, cited by
Salijeni et al. 2019, p. 98). Audit firms utilize BDA in audit methodology to explore the
potential benefits and to serve their clients effectively.
BDA require an ample amount of investment in hardware, software, and skills
development to enable the appropriate data mining from the clients and the other third-
party organizations (Cao et al., 2015, p. 427). Recently, the big four audit firms have
invested a considerable amount in acquiring or developing new BDA tools (Salijeni et al.,
2019; p. 98. kokina & Davenport, 2015, p. 116). For instance, Deloitte, in cooperation with
Kira System, has started to develop a text mining program to extract significant
information from unstructured data (Richins et al., 2017, p. 73). KPMG in cooperation
with Formula One Racing and other technology firms, has assigned a $100 million to
develop analytics systems that may generate additional value to its clients. Similarly, EY
has invested $ 400 million in developing audit methodologies and data analytics
capabilities. PWC has developed the Halo platform internally to add data analytics
capabilities to its applications (Salijeni et al., 2019, p. 98; Richins et al., 2017, p. 74;
Kokina & davenport, 2015, p. 116) and to be ‘a next-generation software application that
analyses and assures data using a suite of algorithms’ (PwC, 2014b, p.15; cited by

2
Salijeni, 2019, p. 14). Moreover, medium, and small tier auditing firm has used available
software like Spotlight, Lavastorm, and Alteryx in order to develop their data analytics
abilities (Salijeni et al., 2019, p. 99).
Earley (2015, P. 493) argued that BDA is a game-changer for the audit profession. Other
researchers predicted that BDA would make a major change in audit methodology (Gepp
et al., 2018; Salijeni, 2019). Cao et al. (2015) argued that BDA has huge potential to
improve financial statements audit. There is an increasing interest and recognition of BDA
in auditing by academia and audit firms (Appelbaum, 2017, p. 2). However, the auditing
profession is different from other businesses. The profession is guided by sets of standards
that constrain the audit firms from adopting new technologies. Therefore, auditors should
consider the constraints before adopting BDA (Alles, 2015, p. 440) to legitimate or
maintain the legitimacy of incorporating BDA in the audit profession (Salijeni et al., 2019,
p. 98). Although, this development has resulted to recognition and approval of BDA by
international professional auditing bodies such as IAASB and ACCA (Salijeni et al., 2019,
p. 96). Regulatory agencies and professional bodies recognize the ability of BDA to
minimize the risk of losing physical records. BDA eliminate the use of paper records to
digitalized records and cloud storage (Salijeni et al., 2019, p. 96). The regulatory bodies
are making efforts to assess the effect of BDA in audit methodology and how it could be
regulated (Salijeni, 2019, p. 12).
Auditors have embraced the use of BDA to increase the effectiveness and credibility of the
audit reports. Additionally, BDA may have potential effects in reducing the cost of
auditing and enhancing the bottom line of audit firms as a result (Alles, 2015, p. 440). In
contrast to traditional audit methodology, auditors can perform analysis on an entire
population. Through the use of BDA, auditors can improve risk assessment processes,
substantive procedures, and the test of internal control (Earely, 2015, p. 495). The
development of data analytics tools geared toward improving audit quality and audit
efficiency and effectiveness. However, it is imperative to note employing data analytics
alone cannot enhance audit quality. It is the combination of data analytics and professional
judgment facilitated by data analytics that can improve audit quality (Salijeni et al., 2019,
p. 105; ICAEW, 2016, p. 6)
Audit quality has been under continuous scrutiny as a result of failure from major audit
engagements such as Enron, WorldCom, Lehman Brothers, Freddie Mac, etc. These
incidents lead the audit engagements to be questioned by regulators, governments, and
investors. The reliability of audit opinion is dependent upon the validity and quality of
audit evidence collected during the auditing process. BDA enables auditors to extract data
using simple data structures in place of the traditional record file (ICAEW, 2016, p. 5).
Auditors can also make the use of external data sources such as suppliers, customers,
banks, and tax agencies. Notably, auditors can perform extraction and analysis of such data
independently and this is crucial in gathering sufficient evidence (ICAEW, 2016, p. 5).
Most importantly, it is possible to perform the analysis in more detail to gain precise
insights into performed activities and business operations. This supports the auditor’s
judgment in forming an appropriate audit opinion (ICAEW, 2016, pp. 6- 9)

3
1.2. Problematization and research gap

In auditing, BDA have offered several opportunities for auditors’ activities. This includes
bankruptcy assessment, management fraud, risk assessment associated with audit
engagement, and identification of material misstatements in the financial statements (Cao,
et al., 2015, p. 424). Furthermore, instead of the traditional focus on financial transactional
data, BDA provides further opportunities for auditors to focus on structured data, semi-
structured data, and unstructured data (Alles & Gray, 2015, p. 8). BDA provides a
possibility to test the whole population instead of the current traditional sampling approach
(Cao et al., 2015, p. 427; Richins, 2017, p. 72). This can lead to enhance audit quality by
increasing the sufficiency of audit evidence. Nevertheless, other than the role of accessing
massive data, BDA is considered as a complement to internal audit evidence (Yoon et al.,
2015, pp. 432- 433). Similarly, BDA provides support to auditors, especially, in the case of
traditional audit insufficiency of evidence, and it increases the reliability and relevance of
audit evidence (Yoon et al., 2015, pp. 432- 433).
Testing 100% of financial transactions and nonfinancial information is considered a
paradigm shift in audit methodology from traditional. Auditors can focus on higher risk
transactions to identify patterns and anomalies in which this may lead to higher audit
quality. Moreover, BDA in auditing provides opportunities to use non-financial data to
make a better professional judgment. In cases like going concerned and valuation, auditors
can use BDA to evaluate and establish new models to predict future events (Earley, 2015,
pp. 496- 497). Salijeni et al. (2019, p. 106) argued that the rise of BDA in auditing
provides possibilities to outsource routine and repetitive tasks. Consequently, this will
allow auditors to spend more time on complex issues that need professional judgments.
Additionally, this may lead to reducing audit costs and avail audit firms a competitive
advantage.
In contrast, previous researchers have found that access to BDA may raise new issues like
the ambiguity of information, information overload, and false positives. These may lead to
a decreased reliability of information and the benefit of using BDA (Salijeni et al., 2019, p.
99). Furthermore, BDA may cause additional work for auditors, which could result in an
increased audit cost (Salijeni et al., 2019, p. 109). Other concerns are related to the
knowledge and expertise of auditors. In BDA environment, the auditor should process a
large amount of data, and most of them are non-financial data and evaluate anomalies in
patterns. The skills and knowledge required to perform such tasks are not yet fully
integrated in the academic syllabus. This concern raised an essential question among
scholars on what knowledge and expertise auditors should have to work within BDA
(Earley, 2015, p. 497). Cao et al. (2015), proposed that audit firms can introduce third-
party providers of BDA service which, however, may raise privacy and confidentiality
issues.
Previous research identified several pros and cons of using BDA in the audit process.
However, as suggested by many researchers there remain several unanswered questions
related to the impact of BDA on the audit process (Salijeni et al., 2019, p. 99). Yet,
existing literatures in auditing have not identified specific skills required for extracting data
from the clients and evaluating anomalies in patterns (Earley, 2015, p. 497). The

4
composition of the audit team is another critical factor that would be affected by the advent
of BDA, and the issue overlooked in available research (Appelbaum et al., 2017, p. 15).
Researchers suggested one of the advantages of using BDA in the methodology will make
it possible to outsource some of the processes to others like IT professionals (Salijeni et al.,
2019, p. 106). This would raise several concerns in assurance engagements, including
confidentiality and privacy, in accessing client’s data (Salijeni et al., 2019, p. 106; Yoon et
al., 2016, p. 436; Cao et al., 2015, p. 428). Furthermore, most of the researchers argue that
incorporating big data in the audit methodology will contribute to audit quality. However,
the concept of audit quality is a subjective term, and achieving high audit quality is
dependent upon different factors (Salijeni et al., 2019, p. 111). Achieving higher audit
quality will depend on the competence and experience of audit team and methodology
used. Finally, there is not enough scientific research on the topic of big data analytics, and
available research focuses on the impact of BDA on auditing from a conceptual point of
view (Brown-Liburd et al., 2015; Cao et al., 2015; Salijeni et al., 2019). The issue of BDA
lacks sufficient empirical evidence from a practical point of view (Salijeni et al., 2019, p.
96). Therefore, more research is needed to exploit this new phenomenon. The above
mention factors motivate the authors to conduct this study.
Despite the lack of empirical evidence on the use of BDA on audit methodology, the
insufficiently available literature offers a positive view of using BDA in audit
methodology (Salijeni et al., 2019; Alles & Gray, 2018). Moreover, prior research calls for
more investigation to provide references on the topic from a practical perspective. Many
researchers identified key areas worth investigating, and researchers proposed research
questions that need to be addressed in future studies (Salijeni et al., 2019; Alles & Gray,
2016; Richins et al., 2017). For instance, what and how the BDA affects auditor’s opinion,
audit report and audit evidence? Does BDA increase the audit quality? How BDA affect
audit efficiency and professional judgment? What role regulations could play in BDA?
(Alles & Gray, 2016, p. 58; Salijeni et al., 2019, p. 112, Appelbaum et al., 2017). This
suggest that there is a need to provide empirical support for BDA fit within traditional
audit methodology.
1.3. Research Questions

What is the minimum level of skills and competence that are required for auditors to work
with BDA tools?

How BDA may impact perception of professional judgment, audit efficiency, and audit
quality?

1.4. Research purpose

The purpose of this study is to explore and gain a deeper understanding of how the use of
BDA in the audit methodology may affect perception of professional judgment, audit
efficiency and audit quality from practitioners’ perspectives. This thesis also aims to
increase the understanding of skills and knowledge of auditors that are required to perform
an audit in the BDA environment. BDA is developing rapidly, and the debate centered
around the phenomenon concerning the audit process and methodology continues to

5
intensify. Furthermore, this thesis may provide a valuable contribution to the discussion on
the subject.
1.5. Delimitation

Some constraints limited the application of the findings of this study. First, the study
focuses only on big four Swedish audit firms. Therefore, the findings are relevant to the
Swedish context; if this context is altered, the conclusions may change. The study uses
qualitative methods, where individual perceptions, knowledge, and experience play a key
role. The findings may not be wholly true if perception, knowledge, and experience of the
participants differ from the time this research was conducted.
Moreover, this thesis is concerned with a relatively new topic. Another possible constrain
is related to access to relevant and sufficient academic and scientific literature. Finally, the
authors choose to interview auditors as they are being involved and affected by the new
technology. Therefore, it considered that the empirical data would be sufficient to answer
the research question in this context.

1.6. Structure of the paper

The structure of this paper is outlined in the following order. The first chapter contains the
research background, problem statement, research questions, and research purpose. The
second chapter determines the theoretical point of departure. This chapter contains a
previously conducted literature review on the topic that is relevant in answering the
research questions. Afterwards, the third chapter follows with the research methodology.
The chapter is divided into two main sections. The first section is the philosophical
assumptions, and the second section contains the practical steps the authors took to collect
the necessary data to answer the research questions. The fourth chapter comprises the
empirical data obtained from the seven interviews the authors conducted through semi-
structured interviews. The Fifth chapter involves the analysis and discussion of the
empirical results. Finally, in the sixth chapter, the authors conclude with a summary of the
main findings, implications of the findings, limitations of the research work, and identify
areas for future research.

6
2. THEORETICAL FRAMEWORK
This chapter presents the theoretical point of departure. The chapter provides an overview
of the theories related to the topic. First, the chapter discusses the general concept of Big
data before moving to the specific use of data analytics in audit methodologies. After
deliberating on the use of data analytics in audit, the discussion proceeds with theories of
legitimizing the use of big data in an audit. A summary of the audit process in connection
to big data is presented. Theories relating to the research questions are then discussed
chronologically, starting with auditors’ skills and competence, professional judgment,
audit efficiency, and finally, audit quality. Audit quality is presented in accordance with
the IAASB framework for audit quality.
2.1. Big Data and Big Data Analytics

2.1.1 Big Data

Big Data (hereafter BD) is a new notion that has been used heavily in different sectors of
our life like marketing, health care, and policies. BD becomes the currency of current
business, and it is considered as a potential business asset (Appelbaum, 2016, p. 17). BD
has brought a wide range of opportunities to businesses. However, firms should effectively
exploit these opportunities to generate value from BD. Therefore, companies must have
individuals who are capable of associating BD and their analytics to firm strategy and
business fundamentals. As stated by the CEO of AICPA, “Big Data has increased the
demand for information management specialists, while dramatically increasing the
potential for visionary professional growth and positioning. CPAs are perfectly suited to
take a leadership role in deciphering and using Big Data to achieve strategic business
goals” (Rishins et al., 2017, p. 64). Professional bodies and audit firms started to recognize
the significance of BD since 2013 and refer to it in their presentations and publications.
The discussion focused on the value that can be extracted in the financial reporting
(Salijeni, 2019, p. 41).
External auditors are motivated to use BD in their audit engagements for different reasons.
Firstly, the audit’s client has used BD heavily in the decision-making process and
accounting judgment, which could have a material impact on financial statements
(Appelbaum, 2016, p. 17). Secondly, there exists the aspiration of auditors to use it in risk
analysis, assessing client and industry, and confirmation (Appelbaum, 2016, p. 17). Due to
reason that the audit profession is highly regulated, BD has been incorporated lately into
the audit environment. BD has no unified definition, and the researchers suppose that
readers will understand its meaning on an intuitive basis (Vasarhelyi et al., 2015, p. 381).
Various definitions of BD can be found on different websites. However, the Wikipedia
definition is considered one of the most comprehensive definitions which represent
unanimous perspectives. “Big data is the term for a collection of data sets so large and
complex that it becomes difficult to process using on-hand database management tools or
traditional data processing applications. The challenges include capture, curation,
storage, search, sharing, transfer, analysis, and visualization. The trend to larger data sets
is due to the additional information derivable from analysis of a single large set of related
data, as compared to separate smaller sets with the same total amount of data, allowing

7
correlations to be found to "spot business trends, determine quality of research, prevent
diseases, link legal citations, combat crime, and determine real-time roadway traffic
conditions’’(Wikipedia; cited by Alles & Gray, 2016, p. 48).
The researchers define BD from different perspectives, on the one hand, is from the
content of BD and on the other hand from dimension and characteristics of big data (Alles
& Gray, 2015, p. 8). In terms of characteristics, BD is mainly described in 4Vs, namely:
volume, variety, velocity, and veracity. Volume refers to a massive amount of data in the
database. This volume varies among different companies and different industries, and what
may be considered as the BD in one company or industry may not be regarded as such by
another. For example, what a small audit firm regards as BD may not be a BD for the big
audit firms (Vasarhelyi et al., 2015, p. 382). Variety implies the source of data as internal
and external data and different types of data as structured data and unstructured data.
Velocity refers to any extent data are being changed and updated continuously. Veracity is
the most concerning issue for auditors, and it relates to any extent BD provides truthful
information (Yoon et al., 2015, p. 432) and integrity of the data (Alles & Gray, 2015, p. 8).
Further, the audit profession defines BD in terms of contents. BD refers to the combination
of different types of data as internal and external data, structured and unstructured data,
traditional financial data and non-financial data, emails, telephone calls, sensor data,
logistics data, blogs, RFID data and social media data (Alles & Gray, 2015, p. 8). Connoly
(2012, cited by Alles & Gray, 2015 p. 9) identified BD in terms of mathematical equation
“Big Data= Transactions + Interactions + Observations’’ (see figure 1).

Figure 1: Connolly’s’ (2012) definition of Big Data

Source: Alles & Gray (2015 p. 9)
8
Transactions refer to traditional structured financial data that is stored within SQL
databases like ERP data systems. Interactions are the interactions between people, things,
and firms like social media comments and users’ clickstreams. Observations refer to
unstructured data that come from the internet of things (IoT) such as RFID, GPS within
mobile devices, and ATMs.
2.1.2 Big Data as Audit Evidence

The main purpose of the auditors is to collect evidence regarding audit engagement to
evaluate them to express their opinion regarding financial statements. Audit evidence is a
different kind of information that is used by auditors to express their opinions (Appelbaum,
2016, p. 19). According to standards, audit evidence should be sufficient and appropriate.
Further, auditors must test physical evidence in the risk assessment process according to
audit standards (PCAOB 2010, AS 15; AICPA 2012, SAS 122; IAASB 2009, ISA 500)
(Appelbaum, 2016, p. 19).
With the continuous advancement in technology, traditional audit evidence becomes no
longer sufficient due to the change of the nature and competence of audit evidence
(Appelbaum, 2016, p. 19). In their study (2015), Yoon et al. argued that external BD is a
complement to traditional client’s data in case of insufficiency of evidence. Therefore, both
internal and external auditors should have access to data to ensure that it is secure and
trustworthy. External auditor’s access to BD can contribute to the following audit phases.
Acquiring additional knowledge about the audit client and industry in the engagement
phase. Providing support to the auditor in the risk assessment phase in audit planning.
Using non-financial information like social media data in the substantive test phase to
conduct a fair value assessment of intellectual property, for example (Appelbaum, 2016, p.
18). Viewing the audit results in deep insight and is comparable with the client's industry
in the review phase. Providing auditors more knowledge about their clients in the
continuous activities phase (Appelbaum, 2016, p. 18).
Moreover, external data provides support to the auditor to detect fraud because traditional
data might hide important information and include motivation and rationalization about
individual lifestyle, conduct, and morality. Therefore, evaluating external data like emails
can be a helpful tool for auditors to detect fraud (Yoon et al., 2015, p. 433). BD evidence
is considered sufficient due to the volume and variety of the data available in real-time.
Appropriateness refers to the reliability and relevant evidence (Yoon et al., 2015, p. 432).
In the context of using BD as evidence, reliability is a questionable issue. On the one hand,
BD evidence is highly reliable as most of the evidence is generated from external sources
and retrieved by auditors directly. Further, due to its vast volume it is very difficult to
manipulate. On the other hand, the BD in nature is the noise that can lead to false positives
and data overload, hence, decrease the reliability of the data (Yoon et al., 2015, p. 433).
BD is relevant because of the availability of data in real-time (Yoon et al., 2015, p. 432).
Additionally, the relevance of data depends heavily on professional judgment similar to
traditional data (Brow-liburd & Vasarhelyi, 2015, p. 6). besides, using BD as audit
evidence raises the concern about the verification of the accuracy of data. As most of the
data are from external sources and especially if the BD has a significant effect on financial
statements (Appelbaum, 2017, pp. 6-7).
9
2.1.3 Big Data Analytics

Alles and Gray (2016, p. 13) stated that “Big Data—whether considered as an evolutionary
or revolutionary development in technology—remains a means towards an end and not, as
the hype sometimes expresses it, as an end in itself. If auditors are to find value in it, Big
Data and related analytics must lead to better audit outcomes”. BD itself has no value;
therefore, BD should be analyzed by advanced analytics techniques to extract potential
value to help decision-makers to make sound and informed decisions (Gandomi & Haider,
2014, p. 140). The analytics techniques mentioned above are called Big Data Analytics
(BDA). BDA has plenty of definitions however in auditing it is defined as “the science and
art of discovering and analyzing patterns, identifying anomalies, and extracting other
useful information in data underlying or related to the subject matter of an audit through
analysis, modeling, and visualization for the purpose of planning or performing the audit”
(AICPA, 2014, cited in Salijeni, 2019, p. 42). The big four audit firms have responded to
the later claims to introduce BDA in financial statement audits (Salijeni, 2019, p. 41). They
assigned considerable resources to develop new advanced technology internally or in
collaboration with specialized firms within this area, such as Microsoft, SAP, and Oracle
(Richins et al., 2017, p. 73).
The audit profession is mainly dependent on standards and regulations. Therefore, audit
engagement must be conducted in alignment with regulation regardless of the advanced
level of the audit's client in terms of IT and accounting complexity (Appelbaum et al.,
2017, p. 4). Richins et al. (2017, p. 75) argued that the evolution of auditing standards to
consider BDA will determine the role of the auditor within the BDA environment. Further,
auditing standards remain unchanged. Therefore, using BD techniques in gathering audit
evidence instead of traditional audit methods is still unclear (Richins et al., 2017, p. 75).
Some practitioners see the silence of standard setters and regulators as an obstacle to limit
their engagement with BDA. These practitioners will extend their use when the standard
setters provide answers to their uncertainty about BDA use (Salijeni et al., 2019, p. 110).
In this regard, Cao et al. (2015, p. 427) argue that auditing standards should be modified
before the auditor adopts BDA. On the other hand, other practitioners see this silence as an
opportunity to widely use BDA without considering any possible regulatory restrictions
(Salijeni et al., 2019, p. 110). In contrast, standards setting communities argue that the
current audit standards allow the auditor to exploit BDA in the audit context due to the
flexibility of these standards (Salijeni et al., 2019, p. 110).
Incorporating BDA in audit engagement provides the audit profession an opportunity to
use more advanced predictive and prescriptive-oriented analytics (Gyun No et al., 2019, p.
128). Appelbaum et al. (2017, p. 5) argued that close attention should be taken when it
comes to discussing analytical procedures and business analytics (BA). Since both terms
might not be exchangeable. The analytic procedure is a significant part of the audit
process, which includes analyzing data to find out any plausible relationship between
financial and non-financial data. At the same time, BA is “the use of data, information
technology, statistical analysis, quantitative methods, and mathematical or computer-
based models to help managers gain improved insight about their operations, and make
better, fact-based decisions” (Appelbaum et al., 2017, p. 5).

10
BA has three dimensions - domain, orientation, and techniques - that are very useful to
provide a deeper understanding of the scope of BA (Appelbaum et al., 2017, p. 6). The
domain is the environment where audit teams apply analytics like client’ enterprise and
management. The orientation refers to the vision of analytics descriptive, predictive, or
prescriptive. The technique is the analytical approach or method (Appelbaum et al., 2018,
p. 93). The variety of data, techniques, and client’s enterprise system capabilities are
behind the dimensions of BA. Descriptive analytics is backward-looking and provides
answers to what happened. It is used extensively in business, and it uses visualization,
graphs, Key Performance Indicators (KPIs), dashboards, and descriptive statistics to
transform the analysis into valuable information (Appelbaum et al., 2017, p. 6). Predictive
analytics consists of different types of techniques that use historical and current data to
predict future outcomes (Gandomi & Haider, 2015, p. 143). The majority of businesses use
descriptive analytics as the first step of constructing predictive analytics. Prescriptive
analytics is future-oriented by taking predictive analytics further. It is considered an
optimization approach to analyze and determine the best possible alternatives (Appelbaum
et al., 2018, p. 93).

Figure 2: Paths to expand data analytics in financial statement audits

Source: Alles & Gray (2016, p. 45)
Alles & Gray (2016, p. 45) argued that BD and BDA can be interrelated even if both are
independent concepts. They determine the association between the two concepts by means
of above figure (see figure2). For so long, audit firms have been operating in cell “A,” in
which they use traditional tools such as Excel, ACL, and Idea to analyze accounting data.
With a small part of non-financial data in cell “B.” Recently, audit firms have started to use
visualization tools to analyze 100% of traditional financial data in cell B instead of
sampling by following traditional audit procedures (Salijeni, 2019, p. 51). Alles & Gray
(2016, p. 45) added that audit firms should move to cell D to fully exploit the opportunities
of BD and advanced analytics techniques in an audit. In this regard, Appelbaum et al.

11
 (2018) established a framework for future research to provide a deeper understanding of
the increased use of BDA in audit engagements. Another study by No et al.
(2019) proposed a framework to guide auditors through the data selection process and
introduce the obstacles in dealing with a high number of outliers in testing the entire data.
2.1.4 Overview of previous literature

To summarize the finding of previous literatures in the areas of BD and BDA in terms of
their effects on audit methodology, the table below provides some insights about their role
in audit.

Table 1: Overview of previous literature

Author Research Approach Findings

Richins et al., 2017 Salijeni et al., 2019

the study looks at BDA effect on relationship between

Qualitative :22 interviews with
auditor and their clients as a result changing in analyzing,
individuals who have experience in
storage and using financial and non-financial data.
implementing, evolving, or
Looking also on BDA effect on changing audit
evaluating the effect of BDA in audit.
methodology. Overview about the challenges of
Looking also on published paper on
implementing BDA on audit process. the findings are used
BDA within the audit.
to establish agenda for future research.
Accounting professions have knowledge and experience
to work with structured data which help them to make a
Conceptual: looking on pervious shift to deal with unstructured data. Therefore, BDA
literature and publication of big four complement the skills and knowledge of accountant
firms. instead of replacing them. However, the curriculum of
professional bodies and education must be adjusted to
cope with potential challenges of BDA.
Brown-Liburd

Conceptual: looking on pervious implementing BD in audit process does not come without
et al., 2015

literature from auditing and pitfall. As a massive amount of data from different
psychology to find out the behavioral sources to be processed could affect the cognitive ability
implications of BD on professional of auditor to form professional judgement and decision-
judgment. making process.
Data analytics is a game changer in the way auditor
Earley 2015

Conceptual: looking on pervious conduct audit. Addressing the problem of data analytics
paper and popular press that issued in like skills and knowledge of auditors, integrity and
recent years. relevance of available data and stakeholder expectation of
data analytics for further research.
Gray, 2016 al., 2015

Conceptual: looking on pervious BDA does not use heavily in audit. However, its use could
Cao et

literature, publication of big four increase the efficiency and effectiveness of financial
firms and ISA, etc. statement audit.
Conceptual: looking on pervious
Alles &

provide discussion about advantage and disadvantage of

literature, publication of big four
using BD in audit. The authors also suggested areas for
firms, white paper, and blogs of
further research to fill the gap in literature.
practitioner

12
2.2. Legitimacy Theory

According to Salijeni (2019, p. 56) adopting new technology in the audit field like BDA in
this case, audit firms should prove to their stakeholder that BDA is adequate and socially
desirable to meet audit requirements. In other words, audit firms should gain legitimacy
from their stakeholders like clients, regulators, and standard setters. Legitimacy defined by
Suchman (1995, p. 574) “Legitimacy is a generalized perception or assumption that the
actions of an entity are desirable, proper, or appropriate within some socially constructed
system of norms, values, beliefs, and definitions.” Legitimacy within the legitimacy theory
is about the idea of the social contract between organizations and societies they operate
within. The legitimacy will be affected negatively in case the organization does not comply
with the terms of the social contract, which can lead to a legitimacy gap (Deegan, 2018, p.
2315).
The literature in legitimacy theory is divided into two schools: strategic and institutional.
The strategic approach discusses legitimacy from a managerial perspective by focusing on
the organization's effort to gain societal support. In contrast, the institutional approach
deals with legitimacy from an industry perspective as a broader point of view (Suchman,
1995, p. 572). In his study of 1995, Suchman stated that legitimacy is visible in form three
types, namely: pragmatic legitimacy, moral legitimacy, and cognitive legitimacy. Every
kind of legitimacy requires different behavior. Pragmatic legitimacy, which involves
actions that support the best interest of an organization's immediate audience (O’Dwyer et
al., 2011, p. 36). It also comprises the organization's activities and effects on broader
political, economic, and social audience’s well-being (Suchman, 1995, p. 572). Pragmatic
legitimacy is also divided into three types: exchange, influence, and dispositional
legitimacy (O’Dwyer et al., 2011, p. 36). In relation to influence legitimacy, auditors seek
to gain legitimacy from other groups like regulators, in addition to legitimacy from
immediate audiences such as clients (Salijeni, 2019, p. 63).
Moral legitimacy is the second type of legitimacy. Suchman (1995, p. 579) identified
moral legitimacy as “reflects a positive normative evaluation of the organization and its
activities’’. The main point in moral legitimacy is the assessment of constituents to
determine if the practice is the right thing to do (O’Dwyer et al., 2011, p. 36). Moral
legitimacy has four types, namely, consequential, procedural, structural, and personal
legitimacy (Suchman, 1995, p. 579). These types are determined based on the evaluation of
outputs and consequences, techniques and procedures, categories and structures, and
leadership and representative (O’Dwyer et al., 2011, P. 36).
Cognitive legitimacy includes practices that pursue objectives and activities which are
accepted by constituents as taken for granted, ultimate, and desirable (O’Dwyer et al.,
2011, p. 36). According to Suchman (1995, p. 582), taken for granted “is distinct from the
evaluation: one may subject a pattern to positive, negative, or no evaluation, and in each
case (differently) take it for granted.” Therefore, this type is very difficult to attain
because, it depends on cognition instead of interest or evaluation. In relation to this point,
sustainability assurance is taken for granted by constituents due to the necessity and
relevance of this kind of practice in sustainability reporting (O’Dwyer et al., 2011, p. 36).

13
The types mentioned above of legitimacy co-exist in most real-world settings, and only in
some cases, there is a distinction among each other (O’Dwyer et al., 2011, p. 36). For
instance, audit firms’ efforts to legitimizing non-audit services (pragmatic legitimacy) may
compromised auditors' independence. Which, in turn, can affect the whole legitimacy of
the audit profession (Salijeni, 2019, p. 65). According to Suchman (1995), the distinction
appears in two salient points. Firstly, self-interest is the main purpose of audiences in the
pragmatic legitimacy assessment. Whereas, the cultural rules are the base in assessing
moral and cognitive legitimacy. Secondly, pragmatic, and moral legitimacy depends on
cost-benefit assessments and ethical judgments of audiences (Suchman, 1995, p. 585)
whereas, cognitive legitimacy is based on cumulative experiences of audiences (O’Dwyer
et al., 2011, p. 36). Suchman (1995, p. 585) argued based on these observations that are
moving from the pragmatic to the moral and to cognitive legitimacy; legitimacy becomes
more difficult to attain and sometimes needs to reinforce one type to another. In relation to
implementing BDA in audit, audit firms seek to gain pragmatic and moral legitimacy. As
incorporating BDA in the audit process is at the early stages; additionally, legitimacy
process starts from pragmatic and moral legitimacy (Salijeni, 2019, p. 65)
2.3. Legitimacy in Audit

The legitimacy in audit has been discussed from two perspectives. The first point of view
considers auditing as a tool or mechanism to generate legitimacy, while the other point of
view, auditing seeks to be legitimized (Salijeni, 2019, p. 65). Power (2003, p. 380)
suggests that the legitimacy of both perspectives (producing or seeking) is co-produced.
The audit firms and professional bodies cultivate a lot of efforts to standardize the audit
process. However, there are differences in the style and application of audit routines. These
differences are between two approaches - the structured formal style approach, and the
other approach that gives a room for professional judgment. According to Power (2003, p.
381) the difference relies in the nature of auditing as a social entity dependent on “deeply
embedded perspectives’’ instead of a series of technical steps.
Further, the conflict between structure and judgment in the audit process is related to one
mechanism and the other organism. The mechanism seeks an integrative formal audit
approach, which depends on algorithmic knowledge, while organisms focus on whole
knowledge instead of specificity as the whole is greater than individual parts. In their
study, Dirsmith and Haskins (1991 cited by Power, 2003, p. 381) examine the conceptual
conflict within the assessment of inherent risk, and they found that audit firms use broader
factors in assessing the inherent risk when they follow an organic approach. They also
found that the structure-judgment could not be realized as a perfect approach, even in the
predominance of the structure audit approach. This predominance increases the need for
legitimacy and transparency to standardize processes for management control purposes.
Therefore, legitimacy and control are very significant in the structure audit approach, even
if it is not associated with enhancing audit efficiency (Power, 2003, p. 381).
Some researchers argue that incorporating technologies into the audit is used to legitimize
program and audit ideas (Robson et al., 2007; Andon & Free, 2014; cited by Saljini, 2019,
p. 68). The use of technologies either aims to help stakeholders to achieve their objectives
or as a tool to finalize work. Further, these technologies appear to main stakeholders like

14
clients, professional bodies, educators, regulators, and standard setters as appropriate and
desirable because they are associated with shared norms and cultural value of auditing
(Salijeni, 2019, p. 68). In their study Robson et al. (2007, p. 433) found that audit firms
introduce a new audit methodology known as Business Risk Audit (hereafter BRA). To
respond to their failure in detecting management fraud and enhancing their understanding
of business risk and add value to their clients. Audit firms collaborate with educators and
professional bodies to seek pragmatic legitimacy by introducing the BRA as a solution for
auditing problems. The collaboration with researchers and educators helps the audit firms
to develop BRA and gain exchange legitimacy from their clients through their publication
hence pragmatic legitimacy. Audit firms apparently gain moral legitimacy by engaging
standard setters and regulators in BRA (Salijeni, 2019, p. 69).
In relation to BRA, another study conducted by Curtis and Turley (2007 cited by Salijeni,
2019, p. 69) showed that audit firms in their effort to gain pragmatic legitimacy must
convince their auditors of the positive effect of using BRA in an audit. In contrast, the
concerns of the auditor were on the impact of BRA on their professional role in the audit
process, which can raise the demand to generate moral legitimacy. In this regard, O’Dwyer
et al. (2011, p. 41) showed in his study that audit firms should appeal to a wide range of
their main stakeholder for legitimacy. He also adds that the production of legitimacy is
needed in some cases, to manipulate stakeholders to achieve this purpose. Further, if firms
fail to maintain pragmatic legitimacy, moral legitimacy can be gained through
manipulation. Based on the aforementioned, adopting new technology in audit required
producing both moral and pragmatic legitimacy as a cultural value. For instance, audit
technology can enhance the effectiveness and efficiency of the audit (Salijeni, 2019, p. 69).
Salijeni (2019, p. 70) argues that adopting new audit ideas or using new technology from
other fields in an audit may cause challenges and, therefore, will require careful
consideration of context and key audiences.
2.4. Audit processes

The audit process is a series of interrelated activities an auditor has to perform during an
audit engagement. This process starts from the pre-engagement to the point where the audit
is completed and review, and opinion is reviewed and eventually signing the audit report
(Leung et al., 2015, p. 238). During the audit process the auditor gathers information to
understand the client’s internal environment, assess the client-specific industry, and gather
enough audit evidence that will subsequently enable the auditor to form an informed audit
opinion (Eilifsen et al., 2014, p. 20). Eilifsen et al. (2014, pp. 17- 20) and Messier Jr, et al.,
(2017, pp. 18- 20), divide the audit process into seven stages and are highly interdependent
and some iterated. Furthermore, PwC (2013, pp. 8- 9), summed up these seven stages into
five key interrelated stages. Below, these five key stages are visualized and further
discussed.

15
Figure 3: Phases of the audit process
Source: PWC (2013, p. 9)
2.4.1 Pre-audit engagements

The initial phase of the audit process; it involves assessing the viability of the audit clients.
Audit standards require an auditor to establish appropriate policies and procedures for
accepting or rejecting potential audit clients or retaining current clients (PwC 2013, p. 8).
The reason for establishing such rules is to minimize the risk of an auditor being involved
with a client that lacks integrity, as this compromises the independence of auditors
(Eilifsen et al., 2014, p. 68; Messier et al., 2017, pp. 71-72). Before accepting an
engagement with a client, the audit firm should consider key issues regarding the
engagement. The auditor should perform a background check about the integrity of the
client’s shareholders, senior management charged with governance (Eilifsen et al., 2014, p.
17; Messier Jr et al., 2017, p. 18).
Further, the scope and objective of the engagement, duration of the audit, the
responsibility of management and the auditors and the audit fee all should be stipulated in
the engagement letter (Eilifsen et al., 2014, pp. 71-72; Messier Jr et al., 2017, pp. 74-
75). As part of the pre-audit engagement activities, the auditor should also consider value-
added services such as tax planning, transaction support, IT consulting, and internal
reporting process (Leung et al., 2015, pp. 318- 320). This will give the auditor opportunity
to provide additional service and hence, benefit from a holistic view of the client’s internal
environment. This will be a chance to understand the entity's business risk and provide a
recommendation to mitigate those risks (Duh et al., 2007, n d).
2.4.2 Audit plan

Auditing standards ISA 300 require the auditor to develop a proper plan to carry out the
audit activities (Leung et al., 2015, p. 327). Proper planning is essential in fulfilling the
audit objective and meeting the terms and conditions of the engagement agreement (PWC,
2013, p. 8). The planning process involves all the aspects the auditor should consider in
devising a strategy and plan for implementing the audit. This phase of the audit involves
resource planning, risk assessment, and setting materiality (Messier et al., 2017, p. 78).

16
Upon reaching an agreement to carry out the audit on a client’s financial statements, the
auditor is going set out all the resources that will be required to perform the audit
(Fukukawa et al., 2011, p. 85). This involves assembling the right audit team with the right
blend of skills and expertise and setting the time frame to conduct the audit (Eilifsen et al.,
2014, p. 75).
ISA 315 instructs auditors to perform a risk assessment of clients to identify any risk
material misstatement that may be contained in the financial statements (PwC, 2013, p. 8).
Auditors apply their gathered knowledge of the client during the initial phase of the audit,
understanding of the industry and the environment in which the client operates to identify
and assess the risk that could lead to a material misstatement (PwC, 2013, p. 8). A key
process of the planning phase is setting materiality for the financial statement. Materiality
is making decisions about what needs to be considered significant on performance, and
disclosure in significant accounts (Messeir et al., 2017, pp. 78 -79). The auditor uses the
set materiality level to determine if misstatement in an individual account or collectively
misstate the financial statements (Eilifsen et al., 2014, pp. 83-85). The auditor applies
professional judgment and experience in calculating the material level. Hence, auditors use
both qualitative and quantitative mechanisms to assess materiality. Materiality is very
important in the audit process; it also helps auditors form their audit opinion.
2.4.3 Assess internal control over financial reporting

A company’s management and board of directors build a system of internal control

purposely to enable the company to keep track of required financial reporting reliability,
achieve operational efficiency, and fulfill compliance with legal requirements (Eilifsen et
al., 2014, p. 176). In this phase of the audit, the auditor utilized the COSO framework for
internal control to assess the internal control environment of the client (Eilifsen et al.,
2014, p. 179; Messier et al., 2017, p. 180). This assessment is geared to determine whether
a company's internal control system is reliable, produces timely reports, and transparent in
terms of external financial reporting (Messier et al., 2017, p. 180). The auditor performs a
test of control for mainly two reasons, to be able to assert that the system is reliable or to
establish level substantive procedures required to generate enough sufficient audit evidence
(PwC, 2013, p. 8).
2.4.4 Perform a substantive test on financial transactions

Regardless of the strength of the internal control over financial reporting, the auditors
could not give a hundred percent reliance on the system (PwC, 2013, p. 9). International
auditing standards require the auditor to perform substantive tests on some, if not all the
financial transactions. This is particularly relevant in the process of obtaining relevant and
appropriate audit evidence to guide the audit opinion (Messier et al., 2017, pp. 191- 192).
When carrying out substantive procedures, auditors can either utilize manual procedures or
make use of Computer-Assisted Auditing Techniques (CAATs) (Eilifsen et al., 2014, p.
139). During this process, auditors carry out an analytical review of documents, make
inquiries to company personnel, customers, suppliers, also observation, and inspection to
company warehouses (PwC, 2013, p. 9). The professional standard requires auditors to use
professional judgment and skepticism to gather evidence through the review documents,
and reports (PwC, 2013, p. 8). Sufficient and appropriate evidence must be obtained for
17
auditors to justify their conclusion on management assertion on financial statements
(Leung et al., 2015, p. 440).
2.5.5 Report the findings

The last stage of the audit process is to evaluate the finding and choose the appropriate
audit opinion to produce the report to the users of financial statements (Leung et al., 2015,
p. 784). The final product of the audit process is the auditors’ report on the findings. The
audit report communicates to management and other users of financial statements how true
and fair the financial statements are (Leung et al., 2015, p. 254). The audit report usually
contains information such as the scope of the audit, the right and responsibility of auditors
and management, key accounting policies, and the audit opinion (Messier et al., 2017, p
599). According to Eilifsen, et al. (2014, pp. 590- 595) there are four main audit opinions
available for auditors to choose from. These are qualified, unqualified, disclaimer or
adverse audit opinion.
2.5. Competence and skills of auditors

2.5.1 Competence

Competence has a variety of definitions, and the word competence is used in everyday
language, and it has different meanings referring to ability, capability, and effectiveness
(Weinert, 1999, p. 3). Weinert (1999, p. 4) defined the competence concept as a
combination of individual prerequisites such as abilities, proficiency, and skills that are
significant to achieve specific objectives. In their study, Hager & Gonczi (1996, p. 2)
identified competence as a comprehensive and integrated approach that goes beyond the
tasks, skills, and knowledge. This approach contains a variety of attributes required to
perform key tasks. These attributes have cognitive attributes such as (knowledge, problem-
solving strategies, communication, pattern recognition, and critical thinking), interpersonal
skills, effective attributes, and technical skills. Training and assessment in the generic
approach are considered as a strategy to train and assess candidates' attributes. These
attributes also are assessed in isolation from any achieved work. Consequently, it may raise
concerns for future performance evaluation (Hager & Gonczi, 1996, p. 2).
The aforementioned is competence in general. International Organization for
Standardization (IOS) issued a guideline (ISO 19011: 2011) to manage auditing systems,
and it includes the required competencies for auditors. Achieving the objectives of audit
processes depends heavily on the competences of individual auditors and the audit team
who are involved in all stages of the audit process. The evaluation process of auditors
should consider personal behavior, knowledge, and ability to implement what they gained
through education, work experience, and training. This process should include an audit
program and its goal. Further, the audit team should have the diversity of competences that
is sufficient to achieve audit goals; on the other word, each auditor in the team should not
have the same competences. Based on the outcome of the evaluation audit process, the
audit team should be selected, additional training is required to enhance competence, and
evaluate the auditor’s performance continuously (BSI, 2011, pp. 24-26).

18
2.5.2 Skills

The term skills are used by an educational psychologist and human capital theorists. Skills
refer to individual worker properties instead of the propriety of a job (Vallas, 1990, p.
379). This perspective considers skills as an independent variable in the prediction of the
variation in salary levels. In contrast, a sociological point of view focuses on the property
of the job. It considers skills as a dependent variable in explaining the variation of skill
levels over time (Vallas, 1990, p. 380). Currently, technology has a major impact on
auditor work, and it could make their job easier and more efficient. However, technology
cannot substitute specific skills like human intellect, judgment, and leadership. In relation
to auditor skills, KPMG, in collaboration with Forbes Insight, determine in their study five
essential skills. These skills are required from auditors to enhance audit quality and should
be used in accompanying innovative technology (Forbes Insights, 2018; KPMG, 2017a,
pp. 5-12).
The first skill is strong communication skills which is considered the most important one.
It helps auditors to spread their ideas, suggestions, and thoughts during the meeting and
interview and negotiation with their clients. Emotional intelligence is the second skill.
Even though financial client records might be disorganized and contain the possibility of
fraud, auditors must always maintain patience to ensure everything is accurate. The third
skill is critical thinking and business acumen. Auditors can analyze objectives and evaluate
information and fact. Further, the auditor must gain insights into the client’s business and
industry to pose the right questions. The fourth skill is professional skepticism. It has a
significant role in designing and executing an audit engagement as well it helps auditors to
recognize and overcome bias and make an appropriate judgment. The last skill is
interpersonal skills. Since the auditor deals with different clients in different situations,
therefore, the success of the auditor requires exceptional people skills, for instance,
empathy which helps auditors to gain a better understanding of the client’s point of view
(Forbes Insights, 2018; KPMG, 2017a, pp. 5-12).
2.5.3 Effect of Advanced Technology on Skills and Competencies

Advanced technology played and continues to play an influential impact on the skills and
competencies required for future work. Recently, advanced technology like artificial
intelligence and robotics calls individuals to think in different ways to adjust their skills
and competences to adopt these technologies (David, 2015, p. 5). David (2015, p. 5) argues
that advanced technology like artificial intelligence and robots can replace workers in
performing routine and repetitive work, whereas, in some work that requires judgment and
flexibility, advanced technology is still so far to replace workers. Therefore, individuals
should consider creativity, flexibility, and judgment as the most significant competence to
avoid any risk of technology substitution in the future (The Institute for the Future, 2011,
pp. 8-10).
The Institute for the Future (2011) identified the ten main skills and competencies that may
be required in the future. These skills are sense making, social intelligence, novel and
adaptive thinking, cross cultural competency, computational thinking, new media literacy,
transdisciplinary, design mindset, cognitive load management and virtual collaboration.
Social intelligence refers to making a connection with people directly and deeply to
19
encourage and sense the reaction and required interaction. Socially intelligent employees
can evaluate emotion, tone, and gesture very quickly. Social intelligence also is considered
an essential skill for workers to collaborate and build a trust relationship. Based on the
study of Institute for Future, in the last three decades, the low and middle-skills work
which required manual tasks declined due to automation. Thus, high skills work that
required abstract tasks are growing. Therefore, skills such as novel and adaptive thinking
determine a premium for future work, which refers to thinking in a new way to come up
with solutions beyond the rule based. Working in diverse cultures and environments will
be more significant in the future, which can lead to innovation (The Institute for the Future,
2011, pp. 8-10). Moreover, skills like computational thinking and cognitive load
management are very significant in preparing future workers. These skills refer to the
ability to use huge amounts of data and the ability to filter relevant data to extract value
(The Institute for the Future, 2011, pp. 8-10).
In relation to BDA, Earley (2015, p. 497) determined the main challenges for current and
future auditors are lacking skills like pattern recognition and evaluating anomalies since
these kinds of skills are not within the education accounting curriculum or audit firms’
training programs. As well as, these skills require many years of experience to be acquired
(Earley, 2015, p. 497). Having the mindset and interest is a prerequisite for developing
skills, particularly in data analytics and data science. Tschakert et al. (2016, p. n/a) argue
that in fast advancing technology, auditors need to adjust quickly to new tides skills
requirements in audit engagements. It is imperative to point out that such skills are
developed through several years of learning and practicing.
2.6. Professional Judgment

Professional judgment is complicated to define as it is a subjective issue (Anca & Andra,

2014, p. 218) dependings on professional experience and personal qualities (Adrian &
Viorica, 2015, p. 141). In their article, Danescu & Chira (2015, p. 1258) determined
professional qualities as independence, accuracy and confidence, objectivity and technical
skills, professional knowledge, and ethical behavior. In relation to auditor judgment, there
are many definitions in the literature (Anca & Andra, 2014, p. 218). However, according to
Canadian research (CICA, 1995; cited in Danescu & Chira, 2014, p. 1254) it could be
defined as “the application of relevant knowledge and experience, within the context
provided by auditing and accounting standards and Rules of Professional Conduct, in
reaching decisions where a choice must be made between alternative courses of action’’.
Professional judgment considers the essence of stakeholder trust in financial information.
Further, professional judgment constitutes a prerequisite for an auditor to provide specific
quality work and to meet the expectation of financial statements users (Oleksandr, 2016, p.
35).
Moreover, international auditing standards guide auditor’s activities without determining
specific steps and situations, therefore, professional judgment “is the purpose of the
professional to act in the virtues of the professional judgment in order to solve the
particular issues of every entity, not stated in the professional standards” (Popa et al.,
2012; cited in Adrian & Viorica, 2015, p. 141). Several auditing standards include a
particular paragraph related to professional judgment, and it leaves room for auditors to

20
exercise their professional judgment in different stages of audit procedures (IAASB
Handbook, 2015). For instance, ISA 200 stated that professional judgment is very
important to conduct the audit properly. Judgment should be exercised in making particular
decisions such as risk audit and materiality; the used audit procedures meet the ISAs
requirement in terms of time and nature; in evaluating the sufficiency and appropriateness
of obtained audit evidence; in assessing management judgment in compliance with entity's
financial reporting framework and expressing auditor opinion based on gathered evidence (
IAASB Handbook, 2015. p. 93; Wedemeyer, 2010, p. 321).
Furthermore, evaluation of professional judgment is very important to ensure that reached
judgment complies with auditing and accounting principles and is consistent with the
existing fact and circumstances in the date of the auditor's report. Additionally, using
professional judgment does not represent any kind of justification of decisions that lack
supporting facts and circumstances (IAASB Handbook, 2015. p. 94).
The auditor judgment is the most critical element in auditing financial reports and
producing an appropriate audit opinion. Therefore, many concerns have been raised
regarding professional judgment, and recently the salient concern has been expressed in the
Sarbanes – Oxley Act of 2002 (SOX) due to Arthur Andersen's failure in detecting risk
(Wedemeyer, 2010, p. 321). According to the Center for audit quality report on
professional judgment resources (2014, p. 1), they found the main challenges for auditor
judgment are that auditing, and accounting standards are developed on principle based.
Similar circumstances require consistency in the decision; business transactions on a global
basis become more complex. Auditors should consider different plausible alternative
approaches in applying some standards. Thus, increasing the complexity of these
standards. Financial statement users require more disclosure related to controversial
accounting policies, estimates, and other highly subjective components.
In relation to BDA, Richins et al. (2017, p. 73) argue that the automation of financial
statements audit is not an easy process as part of audit engagement includes significant
subjectivity. For instance, making a going concern decision that includes a high space of
subjectivity cannot be fully automated. Hence the BDA can provide support for the auditor
to make this decision by providing financial statements analysis. Therefore, BDA is not a
substitute for professional judgment while it can be a compliment. Brown-Liburd et al.
(2015, p. 454) argue that BDA has a potential impact on enhancing professional judgment
and decision making. However, some issues like information overload, information
relevance, pattern recognition, and ambiguity might have adverse effects on professional
judgment (Brown-Liburd et al., 2015, p. 455). According to Krahel and Titera (2015; cited
by Salijeni et al., 2019, p. 99), eliminating the uncertainty of judgment in relation to
mentioned big data limitations could be achieved by incorporating BDA informal auditing
standards. Further, BDA has an important role in closing the perspectives of clients and
auditors to issues that include a huge amount of judgments (Salijeni et al., 2019, p. 103).
Moreover, using BDA provides an opportunity for auditors to outsource some repetitive
and routine work; hence, auditors have more time to focus on complex issues that need
their judgments (Salijeni et al., 2019, p. 106).

21
2.7. Audit efficiency

The efficiency can be generally defined as the production of goods or services without any
waste by using the required input. It also describes the relationship between output and
input. An efficient operation is characterized by using minimum input to produce a given
number of outputs in high quality or producing the maximum number of outputs by using
the given number of inputs. Therefore, management always aims to cut the cost per unit
and improve productivity in order to enhance efficiency (KPMG, 2013, p. 1).
Organizations seek to increase their efficiency by automating their business process, which
can minimize the degree of effort and eliminate unnecessary processes and procedures.
(Sirohey et al., 2012, p. 2). The increased use of automation in the audit process will raise
audit efficiency in several ways. The new advanced technology like BDA makes it possible
to test the whole population. Instead of sampling and provides auditors more certainty and
precision about transactions and more evidence regarding deficiency of control. Further,
automation can highlight any transaction that deviates from the standard process; thus,
increases the pressure on the auditor to detect fraudulent transactions (ICAEW, 2018b, p.
7).
The increased use of advanced technology and digitalization in the audit may make a huge
shift in auditor work. As machine learning and artificial intelligence enhance, there is a
possibility for auditor judgment to be digitized. This crucial part of the audit process that
can be executed by machine instead of auditor; thus, the audit efficiency becomes higher.
The ongoing development of technology in the audit may increase time efficiency and
eliminate the need for human clerical to perform routine and repetitive work. Further, data
analytics provide wide access to client data. However, filtering the data and providing
advice to clients still need human intervention. Therefore, it is very early in this stage for
machine learning and artificial intelligence to replace human inputs (ICAEW, 2018b, pp.
7-8).
Due to economic uncertainty, high competition among audit firms, and the continuous
requirements of audit clients to reduce the audit fees, audit firms are always under pressure
to manage their efficiency (Shimamoto, 2010, p. 2; Bierstaker & Wright, 2001, p. 28).
Audit efficiency can be managed by using technology tools to automate the audit process
(Shimamoto, 2010, p. 2). In his research Shimamoto (2012, p. 2) found that firm success in
adopting automation tools did not depend upon the firm size and degree of technology.
Since all firms with different levels of technology have derived value from ‘automation’.
Further, this study determined the main keys that increase audit efficiency. The effort of
automating the audit process should be organized and structured. Therefore, an
appointment of an excellent audit officer is considered very important to ensure the success
of this effort. The excellent audit officer is responsible for choosing proper automation
tools at the right time; a deep understanding of the audit process is required prior to any
automation and using new technology in the audit process. Moreover, the audit process
should also be mapped out in order to focus on necessary tasks and workflows to perform
and manage this process. Critical paths in the audit process should be determined by
understanding the tasks and workflows and their interrelations (Shimamoto, 2010, pp. 4-5).

22
The full rollout of automating the audit process should be divided into logical stages by
adopting the incremental approach. This approach helps firms to assess each stage and to
make appropriate correction prior to firm-wide rollout. It can also reduce the change
effects in staff and firm budget and mitigate the failure risk of a wide implementation.
These logical stages depend on different factors like type and complexity of engagements,
phase of the engagement, engagement staff, and tool functionality. Furthermore, the stages
vary among firms based on the intensity of these factors in each firm (Shimamoto, 2010,
pp. 5-6). The next key is making the pilot test of a particular tool on a sampling of audit
engagements to gather information and evaluate the use of new tools prior to rollout to the
entire firm (Shimamoto, 2010, p. 6).
Additionally, adopting the new tool slowly by making the staff technology-friendly to use
the tool in one engagement, which leads them to use it again in the upcoming engagements
(Shimamoto, 2010, p. 7). Further, staff training is very important to ensure the proper use
of new tools and to reduce their resistance to change and to raise their degree of adoption.
Proper staff training also reduces wasting time in their trials to determine the best way of
using this tool. The time-saving benefits from automation cannot be translated to cost-
saving since most of the staff costs are fixed. However, the benefit of timesaving can be
realized once this time can be used in additional work to generate revenue (Shimamoto,
2010, pp. 7-8). A survey conducted by EY (2017, p. 2) on the internal auditing
departments of twelve financial institutions showed that AI and data analytics increase the
audit efficiency and the coverage of audit.
The International Auditing and Assurance Standards Board have pointed out that the
increased use of advanced technology will improve audit efficiency. The technology has
an important impact on audit efficiency in different aspects, such as more effective
interaction between the internal and external auditor, which provide the external auditor
with deep insight about risk assessment and business understanding. Further, accurate
planning and meeting the timetable and reporting deadline also contribute to audit
efficiency. Organizing the communication channels and coordination between audit team
engagement and management to avoid any duplication requirement of the same issue from
different team members (IAASB, 2013, p. 38-39).
2.8. Audit Quality

The efficiency of the world’s financial market and economic growth depend highly upon
the confidence of investors. The investors' decision on how to allocate their financial
resources is based on the reliability and credibility of financial information. Therefore,
audit quality is very crucial and essential to investors and other financial statements users.
Independent auditors contribute to improving the reliability and credibility of financial
information by expressing their opinions on financial statements prepared by management
(IFAC, n.d., p. 1). Independent auditors also provide financial users with a high degree of
trust in the financial statement by gathering the sufficiency and appropriateness of audit
evidence (IAASB, 2013, p. 15).
Further, international standards of auditing and financial reporting contribute to this trust
(IFAC, n.d., p. 1). The audit quality is a subjective issue that is affected by the knowledge,
expertise, skepticism, and competence of the auditors in making their professional

23
judgment based on the fact of the engagement. Moreover, describing and assessing audit
quality is challenged by a number of factors. Material misstatement in the audited financial
statement either exists or not contributes partially to audit quality. The sufficiency and
appropriateness of audit evidence rely upon the judgment. Stakeholders have different
perspectives about audit quality, the performed work, and findings of audit lack of
transparency (IAASB, 2013, p. 15).
Many studies have been conducted on audit quality to arrive at a framework and consensus
definition of audit quality (DeAngelo, 1981; Francis, 2011; Knechel et al., 2013).
However, no consensus definition exists, and audit quality is a complex concept, and it
cannot be defined simply (Francis, 2011, p. 127; Knechel et al., 2013, p. 385). Francis
(2011, p.127) stated that audit quality has different grades that graduate from high to audit
quality, and these grades are affected by audit quality drivers. These drivers or elements of
audit quality are input, process, output, interaction, and contextual factors (Francis, 2011,
pp. 125-126; IAASB, 2013, p. 18; Knechel et al., 2013, p. 389). These elements are
illustrated in figure 4 below.

Figure 4: IAASBs’ Framework for Audit Quality

Source: IAASB (2013, p. 8)

Input is divided into three sub-categories, which are called factors. These factors are
auditor values, ethics, and attitude; auditor knowledge, expertise and adequate time

24
assigned to audit engagement; audit process implemented and quality control procedures
(IAASB, 2013, pp. 24-25; Knechel et al., 2013, p. 390). The attributes that support the
audit quality are applied at three different levels, namely, engagement, firm, and national
(IAASB, 2013, p. 24). Auditor values, ethics, and attitude have key attributes that support
the audit quality and performing the audit engagement. These attributes are integrity and
objectivity of the audit team, conducting an audit in the interest of all stakeholders,
independence, competences, and professional skepticism of the audit engagement team
(IAASB, 2013, pp. 24-25; Knechel et al., 2013, p. 390).
Knowledge, expertise, and time of audit refer to a combination of competences that the
audit engagement team has, and these competencies should be appropriate to perform the
audit. In addition to the competences, the audit team should have adequate time to gather
sufficient and appropriate audit evidence to express their opinion (Francis, 2011, p. 126;
IAASB, 2013, pp. 31-33; Knechel et al., 2013, p. 390). Audit process and quality control
procedures are the third factors of input. This factor includes; the compliance of audit team
with auditing standards, regulation and laws and quality control procedures of audit firms;
using appropriate information technology by audit team; audit team interaction with the
internal auditor; achieving audit efficiency by making an appropriate arrangement with
management and appropriate audit documentation (Francis, 2011, p. 126; IAASB, 2013,
pp. 37-38; Knechel et al., 2013, p. 390).
In relation to the appropriate use of information technology, advanced technology has a
high effect on the audit process. It has brought opportunity in efficiently gathering audit
evidence to make better coverage of transactions and risk testing (IAASB, 2013, p. 38). In
this regard, BDA, as an advanced technique, can contribute to auditing quality by
increasing the coverage of testing transactions to include the whole population of data
(Earley, 2015, p. 495; ACCA, 2016). Advanced technology has an impact on auditor work,
and it has changed the way auditors conduct their work. Audit software facilitates the work
of the engagement team and provides help to audit firms to implement their
methodologies, which in turn can increase the efficiency and quality of audit.
Further, advanced technology facilitates auditor communication within the team and with
stakeholders. For instance, email and other communication tools are widely used, and they
provide access internationally (IAASB, 2013, p. 38). However, these communication tools
cannot be without pitfalls. For example, audit evidence obtained via email exchange is less
useful than the interaction and open discussion with management, and it is easier for
management to provide inaccurate and incomplete evidence (IAASB, 2013, p. 38).
The audit process and methodology are affected by the audit’s firm policies and
procedures. These methodologies have an influential impact on audit quality by helping
auditors in exercising their professional skepticism and implementing their professional
judgment. The audit methodology should always be subject to adjustment to be fit with
professional standards as well as external and internal quality control procedures. Further,
audit methodology helps in information sharing as most documents are available in
electronic form. Audit methodology also brings the possibility to perform the work from
anywhere by staff who have limited experience. Therefore, experienced staff needs to do
the supervision and review of work to preserve the audit quality (IAASB, 2013, p. 40-41).
A survey was conducted by KPMG in collaboration with Forbes Insight to assess the
25
impact of advanced technology such as Data Analytics and AI on audit quality. The survey
finds that audit quality can be enhanced by the implementation of advanced technology.
Using advanced technology helps auditors perform routine work very quickly and free up
their time to be able to make better decisions and professional judgment, which can lead to
improved audit quality (Forbes Insights, 2019).
Output factors refer to audit reports and other information issued by auditors to different
stakeholders. Stakeholders see these reports as a part of audit quality, and they assess them
based on the value and timeliness of the reports. Moreover, the main purpose of the
auditors is to express their opinion in order to raise the reliability and credibility of audited
financial statements. An unqualified report is considered a significant indicator for
financial users, and this indicator is affected by the reputation of the audit firm. Recently,
the audit report has become standardized, and it includes more information. Thus, its value
might be increased (IAASB, 2013, pp. 46-47). Further, other indicators that affect the audit
quality are accounting restatement and litigation against the auditor, and both have a
negative association with audit quality (Knechel et al., 2013, pp. 397-398).
The interaction factor that affects audit quality refers to interaction within the supply chain
of financial statements. IFAC determines this supply chain as “the people and processes
involved in the preparation, approval, audit, analysis, and use of financial reports.” The
link in the chain should be the close and high quality within the supply chain in order to
provide high audit quality, and all links have a significant role in assisting audit quality.
Further, formal, and informal communication has an effect on the attitude and views of
supply chain members; thus, audit quality can be improved. The objective of people within
the supply chain and interaction context can affect the nature and scope of interaction
(IAASB, 2013, p. 50). On the other hand, Francis (2011, p. 134) states that audit quality
could be impaired by the long-term relationship between auditor and client due to
objectivity, and skepticism of the auditor might decline.
Contextual factors are the last factor that affects audit quality. The effect of these factors
on financial reporting vary among countries and become more complex based on their
developments and degree of a business growth. These factors are law, financial reporting
requirements, and corporate governance processes, and they become more developed as a
result of complexity (IAASB, 2013, p. 55). Further, these factors all have an effect on audit
risk, audit evidence, and audit efficiency (IAASB, 2013, p. 55; Knechel et al., 2013, p.
393). On the other hand, Knechel et al. (2013, p. 403) found that these factors have
different impacts on audit quality due to changes in audit input, and the audit process
cannot be observed (Knechel et al., 2013, p. 403).

26
3. METHODOLOGY
In this chapter, the authors describe their choice of methodology and the motivation for
choosing these methodologies. The chapter is divided into the main section. The first
section is the philosophical assumptions the researchers choose to guide their research
process. The second section contains the practical steps taken to conduct the research.

3.1. Theoretical Methodology

In this section, present the theoretical and philosophical standpoints that we deemed
important and appropriate to answer our research question. The arguments for our
research philosophy and choices are provided. The research approach, choice of methods,
and design are explained in detail with strong arguments to clarify our options.
3.1.1. Preunderstanding
Pre-understanding in research refers to the understanding of the critical issues about the
topic, and experience research is about the topic or phenomenon being studied.
Preunderstanding is a prerequisite in the process of embarking on research activity.
Researchers should have an overview of the topic or phenomenon under study. Due to the
fact that researchers mostly lack sufficient information regarding a particular subject,
preunderstanding enable them to develop a guideline on how to approach the study.
According to Nyström & Dahlberg (2001, p. 342), our pre-understanding is shaped by
biological, social factors; knowledge is built upon the foundation of preunderstanding. The
achievement of good research work depends on the researcher’s pre-understanding of the
topic and how to approach the research process, how the philosophical standpoints will
affect the knowledge being created, and how the research could contribute to the body of
knowledge. These points are key inputs for conducting good research work. Nyström &
Dahlberg (2001, p. 339), warn that pre-understanding could negatively impact a research
work.
To conduct good research, a researcher should approach the project with good
preconception and an open mind. Qualitative research is geared toward gaining more
insights into a particular subject (Bryman & Bell, 2011, p. 410). A researcher should be
aware of prejudices that come with pre-understanding in research. Inasmuch as pre-
understanding is essential in accomplishing the research, its negative impact could
jeopardize the research results and conclusions (Bryman & Bell, 2011, p. 30). Therefore, a
researcher should approach a research work by combining pre-understanding with open
mind and objectivity, particularly in the data collection, interpretation, and analysis.
During the course of their study, the authors gain an understanding of major accounting
theories and practices such as auditing, financial, and management accounting. This
coupled with theories from Business research methodologies and Current Trends in
business administration. The insights gained in the period of study guided their choice of
topic; they (authors) acquired a good preunderstanding of conducting a qualitative study
and of the chosen topic. However, it should be noted, the knowledge gained during these
periods on the topic is limited. This limitation in knowledge motivated the authors in

27
choosing to explore this topic to acquire more insight. To achieve this objective, they will
endeavor to limit prejudices during the research process. They will approach the research
with an open mind and without adversely influencing the data collection and analysis.
They will make the best use of their knowledge and experience in executing the research
process objectively utilizing relevant research philosophy and theories.
3.1.2. Research philosophy

Research philosophy can be referred to as the composition of principles and assumptions

about the nature of reality and how knowledge can be created or developed (Saunders et
al., 2016, p. 124). The philosophical standpoint of research will influence how the study is
approached, designed, and the method employed in answering the research question (Long
et al., 2000, p. 190). However, researchers should note the most important point is the
reason for choosing one philosophical standpoint over another; there is no wrong or right
philosophy. Researchers should adequately argue for their choices based on the research
question. Therefore, the appropriateness of the philosophical standpoint, approach,
method, and research design all depends upon the research topic and question. Cognizant
of the above argument, the authors set to discuss their research philosophy anchored on the
research questions. The ontological, Epistemological, and axiological standpoint influence
their choice of research approach, method, and research design.
3.1.2.1 Ontology:

Ontology is defined as the assumption of a researcher about the nature of social realities
(Long et al., 2000, p. 190; Saunders et al., 2016, p. 172; Bryman & Bell, 2011, p. 20).
There are two main dimensions of the ontological standpoints: objectivism and
subjectivism. Objectivism refers to the point that researchers perceive social realities are
distant and independent from the social actors. This implies that a social phenomenon is
natural and cannot be influenced. Objectivism supports realism, in which social entities are
considered to be like physical factors of the natural world completely independent and
detached from the influence of social actors’ beliefs and values (Saunders et al., 2009, p.
114). Long et al. (2000, p. 192) looked at objectivism as a social organism where social
realities operate according to the dictates of natural laws. Therefore, under the objective
ontology, researchers may carry out the research as an observer and cannot influence the
object being studied. On the other hand, subjectivism is the opposite of objectivism.
Subjectivism assumes social realities are created through the interaction of nature and
social actors. It is the assertion that social reality is made from the action and perception of
actors. Therefore, realities are constructed by individuals in their interaction with
phenomenology. Long et al. (2000, p. 192) see the objective ontology as the pro-natural
science view of the social world whereas, the subjective ontology as the anti-natural
science view of the social world and reality varies according to beliefs and social values.
Further, it is important to note the different views and understand the implication of the
research approach, but the two continuums should be seen as two mutually exclusive
viewpoints.
Reflecting on the topic and the research question of this thesis, the authors choose to adopt
the subjective viewpoint of ontology. The authors assume social reality is continuously
influenced by the social actors; this is based upon the view that both authors and
28
participants in this research are social actors. These social actors have a different view of
reality, depending on their beliefs and social values. All contributors of this research have
knowledge and experience of varying degrees. Therefore, they all have different views and
perceptions of social issues.
On the other hand, adopting an objective view will compromise the findings of the
research. The primary data collection is through interviews; answers to interview questions
are subject to the understanding of the interviewees and the interpretation of the response
by the researchers. As they set out to explore the impact of big data on audit quality and
process, maintaining an open mind that allows the processing of different viewpoints from
the respondents of the interview process. Furthermore, audit quality is a subjective term
and depends upon the perception of the different actors. Based on the above argument, the
authors assume that adopting a subjective approach will be more suitable to adequately
answer the research questions and gain more insight into the topic.
3.1.2.2 Epistemology:

Epistemology is a branch of philosophy concerned with the creation of knowledge. It

discusses what constitutes acceptable knowledge, how such knowledge can be augmented,
and how such knowledge can be transferred (Saunders et al., 2016, p. 127; Long et al.,
2000, p. 192). It is the epistemological viewpoint that researchers will use to measure what
they will consider as adequate knowledge contribution in research work (Saunders et al.,
2016; Bryman & Bell, 2011). Long et al. (2000) identified three dimensions of
epistemological standpoints, namely, Interpretivism, Positivism, and Realism. These three
epistemological positions help measure how the principles of natural science can be
applied to social reality. Positivism views knowledge as solid and lean towards natural
science studies. In this regard, knowledge is gained through rigorous observation and
testing facts. Saunders et al. (2011, pp. 112- 113) describe the researcher assuming the
positivist continuum as a resource researcher, where the object being observed represents
reality and is independent of the researcher.
Another epistemological stand that is close to the Positivism view is Epistemological
Realism. Realism beliefs that both natural and social sciences should employ similar
techniques in collecting data and interpretation of such findings (Bryman & Bell, 2011, p.
17). There are two divides of Realism, Naive realism, and critical realism. Naive realism
state that “what you see is what you get,” it did not recognize the interaction of social
actors. On the other hand, Critical realism argues that knowledge is a result of the
experience of the social actors and not entirely on the object itself (Saunders et al., 2016, p.
138- 139; Bryman & Bell, 20011, p. 17).
In contrast to positivism and realism, interpretivism considers social actors as a key player
in the creation of knowledge and to understand social reality an approach different from
the natural science approach is required (Saunders et al., 2016, p. 140; Bryman & Bell,
2011, p. 17). Thus, Interpretivism is linked to ontological subjectivism. Interpretivism
argues that to understand a social phenomenon, one needs to understand the beliefs and
values of those involved in creating phenomena as social actors. Interpretivist researchers
are considered as feeling researchers, and social realities are influenced by emotion and

29
attitudes of researchers and other social actors (Saunders et al., 2011, pp. 112- 113). Hence,
interpretivism is more appropriate in management and business research.
This study relies on data collected through interviews of different participants involved in
the auditing process. Participants also utilize BDA in executing their duties as auditors to
understand how the technology is changing the process and the outcome. The authors have
no predetermined supposition that backs the data to be collected. Instead, the data will be
analyzed to recognize patterns and some verbal cues in the interview conversation. The
authors understanding of using an interpretive approach will guide the interpretation of the
collected data. It is imperative to consider the personal experience and opinion will affect
the information interviewees will give to the researchers. Hence, they are social actors, and
the interpretation of social realities are subjective to personal experience and opinion
(Saunders et al., 2016, p. 141). These responses will reflect individual values and attitudes
as well as culture and other social factors.
Furthermore, the authors are not independent of the subject being studied. Therefore, they
adopt an interpretive viewpoint. This position is being considered more suitable and
appropriate for the search question and objective. On the other hand, the pre-understanding
of the authors about the topic and processes involved is another factor. The authors
consider these issues to affect the study or research being conducted. It matches the
chosen ontological standpoint. It is also suitable for the choice of research approach and
design. Further, this will not only present the subjective view of the authors, but it also
factors in the collected data from the interviewees and the subjective interpretation of their
ideas. The authors will act more like translators and disseminators of information from the
interviewees, interpreting their responses to derive meaning.
3.1.3. Research Approach

An essential element of academic writing is the concept of theory. Before undertaking a

research project, it is very important for one to be clear about the concept of theory in
business research methodologies. The researcher should have a strong background in what
role theory plays in the research project. Theory plays an influential role in choosing the
research design and method that would enable the researcher to fulfill his/her research
objective. The two main approaches to theory development in research are the Inductive
and Deductive approach (Morgan, 2007, p. 71). The difference between these two
approaches is that in the deductive approach, the researcher will use a hypothesis to test the
theory while in the Inductive approach, the researcher uses findings of research to develop
a theory (Saunders et al., 2016, p. 143). Morgan (2007, p. 71), indicates that the deductive
approach is theory-driven while the Inductive approach is data-driven.
The Deductive tactic consists of setting a research design that generates a theory as well as
developing a hypothesis to test the theory to either accept or reject the theory (Saunders et
al., 2016, p. 146; Bryman & Bell, 2011, p. 150). According to Saunders et al., (2016, p.
146) and Bryman & Bell, (2011, p. 150), Deductive approach is widely used in the natural
science field; that is more ontological objectivist and epistemological positivist.
Alternatively, the Inductive approach is the direct opposite of the deductive approach. The
process involves using this alternative, to begin with, qualitative data collection such as
conducting an interview with research participants. The collected data is then analyzed to

30
identify patterns and themes (Saunders et al., 2016, p. 147; Bryman & Bell, 2011, pp. 13-
14). The purpose here is to use the result to formulate a theory. In contrast to the deductive
approach, the inductive approach derives its origin from the social science where research
is more subjective and interpretive in nature (Saunders et al., 2016, p. 147; Neuman, 2013,
p. 70).
Furthermore, an alternative available for researchers different from the two common
approaches discussed above is the Abductive approach (Saunders et al., 2016, p. 148;
Morgan 2007, p. 71). The abductive reasoning requires both Inductive and Deductive
approaches, it means a researcher alternates between the approaches at the various stages
as the research progresses (Saunders et al., 2016, p. 148; Bryman & Bell, 2011, p. 13).
Abductive reasoning is widely utilized in situations where the findings lack sufficient
pragmatic proof to support the theory (Bryman & Bell, 2011, p. 27).
The purpose of this study is not to test existing theories; the authors will adopt an inductive
reasoning approach. It is important to point out that the researchers relied on previously
conducted studies on the topic to formulate the theoretical framework and steer the
research. Therefore, it is fair enough to say there are elements of deductive reasoning as
well. Hence, there is no intention to build new theories purely. As mentioned above, hardly
a researcher depends on a single approach in research. The dominant approach will depend
on the research topic and question. The choice research philosophies from section 3.1.2.1
and 3.1.2.2 above guided our choice of reasoning. The goal of this study is to explore how
BDA is affecting auditors’ skills and competence requirement, professional judgment,
perceived audit efficiency and audit quality. The empirical data will be gathered using
interviews with auditors. This is bound to put the perception and experience of participants
into consideration and also contextualize their response.
3.1.4. Research Method

There are two main types of research methods at both ends of the research continuum
(Yilmaz, 2013, p. 323; Long et al., 2000, p.). In one extreme is the quantitative method,
while at the other opposite end is the qualitative method. The philosophical assumptions,
theoretical background, and methodological assumptions will affect the choice method
adopted in research (Yilmaz, 2013, p. 312; Neuman, 2013, p. 167). The difference between
the two paradigms does not necessarily mean they are mutually exclusive (Long et al.,
2000, p.195; Neuman, 2013, p. 167; Yilmaz, 2013, p. 315). The combination of the
paradigms is what Morgan (2007) called the Pragmatic approach.
The quantitative method is the process of using statistical data to investigate the correlation
between variables (Saunders et al., 2016, p. 166; Yilmaz, 2013, p. 311). The researcher
utilizes mathematical models to analyze the collected data to establish patterns and
relationships. According to Yilmaz (2013) and Neuman (2013), generally, this quantitative
method uses standardized data collection tools such as questionnaires, experiments, and/or
surveys. The collected data is transformed numerical data as input to the mathematical or
statistical models. Usually, the sample size in a quantitative method is large enough to
enable them to generalize his/her conclusion (Yilmaz, 2013, p. 314).

31
A qualitative method, on the contrary, focuses on the process, environmental settings for
interpretation of meanings to understand and gain more insights into a social phenomenon
(Saunders et al., 2016, p. 168; Bradley, 1993, p. 433). The data used in a qualitative study
is more descriptive and explanatory in nature and non-standardized (Yilmaz, 2013, p. 313,
Saunders, et al., 2016, p. 168). The data is collected through interview or observation of
research participants. There is a relationship between the researcher and the research
participant. This relationship enables the researcher to gain more insights and to clarify
ambiguous situations. Generally, qualitative research is subjective and interpretive in
nature and followed an Inductive approach (Saunders et al., 2016, p. 168).
This research is aimed at investigating the impact BDA on audit quality, and the skills
auditors need in executing their responsibility using BDA in the process. BDA in auditing
is a new phenomenon and fully exploited. Therefore, the researchers are convinced that the
qualitative method would be more appropriate. There is no adequate empirical data that
could support a quantitative approach. Furthermore, drawing questionnaires or conducting
a survey to collect quantitative data will not be sufficient enough to fulfill the research
objective.
On the other hand, the researcher’s pre-understanding and the fundamental philosophical
assumptions required the use of the qualitative method. Additionally, conducting
interviews to collect qualitative data will permit the authors to gain more insights and a
better understanding of the topic. Therefore, the authors will conduct an interview using a
semi-structured interview with open-ended questions. Along with the semi-structured
interview, the authors will ask additional questions if required. This will allow
interviewees to elaborate and provide in-depth information on the topic.
3.1.5. Research Design

Research design is the strategy upon which a researcher will base the research to answer
the research question. The research design is anchored on the objective of the research
project, the philosophical assumptions, and it is the foundation on which the data collection
technique is determined (Neuman, 2013, pp. 38-39). There are research strategies at the
disposal of researchers to choose from. These are Explanatory, Descriptive, and
Exploratory (Saunders et al., 2016, pp. 174-176; Neuman, 2013, pp. 38-39).
The explanatory design focuses on dilemmas and settings to understand the relationship
between cause and effect to provide a detailed explanation of phenomenon (Saunders et al.,
2016, p. 176; Neuman, 2013, p. 39). The emphasis of this design is on the researcher
having in-depth knowledge to provide a sufficient explanation of the relationship between
variables. A descriptive research design tries to describe certain elements, such as
characteristics and social phenomena. This design may be used to extend exploratory
research and therefore be more structured in nature than exploratory research (Saunders et
al., 2016, p. 175). Neuman (2013, p. 38) indicates the descriptive research is
comprehensive and highly accurate, logical and may refute previous explanations, and the
exploratory and descriptive research confuse together in practice. Finally, the third
research design is exploratory research. This research design is geared towards providing
researchers with more insights, understand new social trends (Saunders et al., 2016, p. 174;
Neuman, 2013, p. 39). Exploratory research is applicable in a situation where the

32
researchers only acquire a limited knowledge about the topic. This design is flexible, it
allows the researchers to incorporate new information, and unlike the explanatory design,
the exploratory is not structured (Saunders et al., 2016, p. 174). This type of design purely
makes use of qualitative data (Neuman, 2013, p. 39).
In this thesis, the exploratory design is adopted; based on the topic and the research
question, this design is more appropriate to answer the research questions and fulfill the
research objective. The authors have limited knowledge on the topic; it is also considered
that interviewees may pose limited knowledge about the subject. The authors also noted
that there is limited research previously conducted on the topic. Therefore, this research is
aimed at contributing to the development of knowledge on the subject. Finally, due to the
flexibility of the exploratory design, it is possible for the authors to reformulate the
research approach and adjust the research questions as the interview progresses and new
information becomes available.
Table 2: Summary of philosophical assumptions

Philosophy Authors’ choice

Ontology Subjectivism

Epistemology Interpretivism

Research Approach Inductive Approach

Research Design Exploratory

Research Method Qualitative

3.2. Practical Methodology

In this section, the authors describe how the research was conducted to fulfill the research
objective. Description of how and motivation for data collection, sampling method,
participants selection, the interview guide, interview process, data analysis, and the
literature search was carried out is provided in detail in this section.

3.2.1. Data Collection

To achieve high-quality research, preparation to collect the appropriate data is necessary.

Preparation for this phase of the research can be tedious and complex, and if not well
executed, it can have negative consequences on the research purpose and the findings (Yin,
2018, p. 130). Accordingly, the skills and knowledge gain during the course of their study,
the authors possess the prerequisite criteria for data collection. These include asking
relevant questions, listening attentively, etc., these techniques were acquired through
various mechanisms ranging from smaller research projects to reviewing previous research
papers. Accordingly, the authors set criteria for choosing the target group of respondents.
This was to ensure the appropriate group of respondents that can provide relevant data are

33
contacted. BDA requires a substantial amount of investments. It is deemed that smaller
firms may not possess the necessary financial muscle to acquire such technologies (Cao et
al., 2015, P. 427). Therefore, the respondents from the big four and mid-tier firms were
targeted for the formal data collection. In addition, the authors continuously review
scientific literature relating to the topic to formulate the interview questions.
Many literatures in business research methodology identify two categories of data, that is
primary and secondary data (Hox & Boeiji, 2005; Saunders et al., 2016; Bryman & Bell,
2011). Accordingly, the authors of this thesis follow the same categories or sources of data.
In this thesis, the primary data collected is used purposely to answer the research
questions, whereas the secondary data is used in constructing the research questions and
building the theoretical framework and to supplementing the primary data. These
categories of data collections are discussed below.
According to Hox & Boeiji, (2005, p. 593) primary data is the data collected specifically
for the purpose of answering the utilize research questions. The authors the procedures that
best suit the research design and approach. In this research, the authors make use of an
interview as the main primary data collection method. Yin (2016) and Saunders et al.
(2016) highlight the importance of using the interview as primary data collection in
qualitative research. The use of interviews enables researchers to gather data that is
relevant and valid towards fulfilling the research objective and answering the research
question objectively. Using interviews in data collection, allow the research to address not
just the “How” questions but a researcher is able to ask the “Why” question as well (Yin,
2016, p. 142; Saunders et al., 2016, p. 391; Rowley, 2002, p. 17). Interviews allow
researchers to understand how a social phenomenon occurs but goes further to gain more
insights by knowing why such a phenomenon occurs in such ways.
Interviews are classified into three categories, structured, semi-structured, unstructured
interviews (Bryman & Bell, 2011, p. 467; Saunders et al., 2016, p. 390). A structured
interview is where researchers used standardized questionnaires with pre-coded answers.
As there may be an interaction between the researcher and the respondents during the
explanation process, the questions would be asked in the same way as they set with the
right tone to minimize biases (Saunders et al., 2016, 391). These types of interviews are
more suitable for collecting quantitative data. While in semi-structured interviews, the
researcher sorts the interview questionnaires according to themes with some key questions
known as the interview guide. This type of interview provides the interviewees an
opportunity to respond in detail, providing more insights for the researcher (Bryman &
Bell, 2011, p. 467). The questions asked may change as the interview progresses,
depending on the response provided by the interviewee. The researcher is able to pick up
on key issues and to ask to follow up questions (Bryman & Bell, 2011, p. 467). This type
of interview begins and ends with some comments; the researchers make use of audio-
recording devices to record the interview for better transcription (Saunders et al., 2016, p.
391).
On the other hand, an unstructured interview is an informal interview process with no
preset themes or questions. The interviewee has the freedom to talk freely; a researcher
explores more in-depth about the topic and the research question (Saunders et al., 2016, p.
391). Both semi-structured and unstructured interviews are appropriate for qualitative data
34
collection. Collecting data through interviews has the advantage of flexibility (Bryman &
Bell, 2011, p. 467).
Based on the above classification of the interview, the authors of this thesis use semi-
structured interview types. The authors choose this approach based on its ability to allow
the interviewees to express an open situation than the structured interview. It allows the
authors to control the thematic discussion during the interview process. It also allowed the
authors the opportunity to ask the most relevant questions to answer the research questions.
According to Saunders et al. (2016) and Bryman & Bell (2011) the authors prepare an
interview guide based on the thematic discussion of the research topic. However, the
interview guide was only used as an instrument to direct the interview discussion to key
issues pertinent to the research objective.
Furthermore, several journal articles and textbooks were reviewed and are used in this
research. Both kinds of literature were adequately analyzed in relation to the suitability of
this research project. The textbooks were accessed through either the University library or
from colleagues, while others were accessed online. Journal articles were retrieved from
the University database and major journal websites as well as google scholar. Both
textbooks and journal articles are selected based on the relevance of answering the research
questions.
3.2.2. Sampling

Regardless of the research question or the research topic, a researcher may not be able to
study an entire population. That could provide information in understanding the topic or
answering the research questions (Saunders et al., 2016, p. 272). Like any other research
project, the authors are aware of the challenges presented in sampling participants.
Accordingly, there are two major sampling approaches; probability and non-probability
sampling (Saunders et al., 2016, p 275). Probability sampling is the technique that gives an
equal opportunity to each member of the target population being selected, and this
approach is more suitable for surveys and experimental research (Saunders et al., 2016, p.
276). This implies that the samples will be randomly selected. Non-probability sampling,
on the other hand, the possibility of choosing any member of the target population is
unknown, and it is not possible to make statistical assumptions (Saunders et al., 2016, p.
295; Cooper, 20, p. 358).
The objective of selecting a sample is to be able to answer the research question and fulfill
the research purpose by understanding the dynamics of the phenomenon. Saunders et al.
(2016, p. 296) further argued that, in business research, particularly in market surveys and
case studies, it is difficult to use probability sampling. Accordingly, the authors of this
thesis adopted a non-probability sampling approach. The decision was based upon the fact
that it was not possible to know the number of participants in the study. In addition,
probability sampling is more appropriate for quantitative research. hence this research
qualitative probability sampling will not be suitable. Furthermore, the sample size is
another factor. The probability sampling technique uses a relatively large sample size
compared to non-probability sampling. The reason being in a quantitative study, the
objective would be to study a population as large as possible in order to generalize the
results. Whereas, in a qualitative study, the objective is to explore and provide more

35
insights into a social phenomenon. Hence, generalization is not the objective and the
population are relatively smaller.
Bryman & Bell (2011, p. 190), identify three types of non-probability sampling strategies.
The first strategy is quota sampling, which practitioner’s belief is very much like
probability sampling. It is widely used in commercial research, such as market research
and political opinion (Bryman & Bell, 2011, p. 194). Quota sampling uses structured
interviews as a data collection strategy (Saunders et al., 2016, p. 299). Therefore, it could
be concluded the data collected are more similar to statistical data. The second non-
probability sampling is convenience sampling. This is a sampling technique where the
researcher selects participants based on accessibility (Bryman & Bell, 2011, p. 190). This
type of sampling may not provide the appropriate respondents to answer the research
questions. Researchers should consider the relevance of the population in providing
relevant input to the research work. The third strategy is snowball sampling or volunteer
sampling. As the name suggests, it is a sampling base on volunteerism. Participants will
volunteer to take part in the research process (Saunders et al., 2016, p. 303). It is
commonly used if it is difficult to identify the target group. The most challenging part of
using snowball sampling is establishing the first contact. The researcher then asks the
contact person to recommend others who could possibly take part in the research. Then,
these contact persons will select themselves to participate in the study (Bryman & Bell,
2011, p. 193).
Cognizant of the above discussion and putting all the advantages and disadvantages of
using each sampling strategy into consideration, this thesis adopts snowball sampling.
Snowball sampling is considered to be the most suitable in this research. This process is
employed to locate people with adequate, relevant information. There is no preset number
of participants; the number of people increases as more informants are found. The period
during which this research is conducted is considered a busy period for auditors.
Technically, the end of the fiscal year marks the beginning of the audit season. Auditors
have busy schedules trying to deliver the report on time to relevant stakeholders. During
this period, it is hard to find auditors willing to spare time for non-work-related
engagements. Therefore, the authors relied heavily on the initial contacts for referrals to
possible participants with relevant information.
3.2.3 Participant Selection

To fulfill the objective of the research, the authors intended to interview auditors at
different levels in the audit profession who use big data analytics in executing the audit
work. The authors want to interview participants from associates’ level to partner level.
The aim of interviewing individuals at different levels of the audit team is geared towards
having a deeper understanding of which levels are more affected by the tools. Since not all
audit firms fit in the purpose of the research, a thorough evaluation, and a review of audit
firms was conducted. To effectively achieve the purpose of the research, data from audit
firms that are not utilizing big data tools in the process may not be useful in the data
analysis, and discussion of this study. Big data requires a substantial amount of investment
in both the auditor and client part (Cao et al., 2015, P. 427). Audit firms with small audit
clients may not need the tools. Besides, auditors may only make use of the tools if the

36
client takes the initial step to invest in the collection and storage of big data (Cao et al.,
2015, P,427). Therefore, the authors targeted participants from the big four and mid-tier
audit firms around the world. It was not possible to access individual contacts of auditors, a
request for an interview was sent using the contact form to specific countries (See
appendix 1). Contact forms were submitted to the big four firms to many countries across
Asia, Africa, Europe, Middle East, North, and South America (See appendix 4). In efforts
to reach participant for this thesis, the authors tried to contact more auditors through
LinkedIn. One hundred and seventy auditors were contacted through LinkedIn. Although
the initial responds were positive but prior to the data collection most those who responded
declined or never replied to the authors. However, the few interviews conducted were
achieve as results personal contact or LinkedIn. Those who responded were forwarded to
all necessary documents. However, most of the companies declined the request citing busy
schedules. The interviews conducted were mainly obtained through personal contact with
the researchers.
The authors explore all available means to secure more interviews, but all efforts were
futile. In the end, only seven interviews were secure from the big four within Sweden.
Malterud et al. (2016, p. 1759) argue that qualitative research sampling should not focus on
the sample size but rather on the quality of the information generated. On the other hand,
Marshall et al. (2015, p. 21), argue that sample size in a qualitative study is influenced by
cultural factors, the field of study and geographical factors. However, both Marshalls et al.
(2015) and Malterud et al. (2016), argue that interviews in qualitative studies should reach
a saturation level. Considering the period of writing this thesis and audit season, the
researchers cannot influence the number of participants. Although the number of
interviews is small, the authors are confident the information generated during the
interview process is relevant and adequate to answer the research questions. Besides, the
interviewed participants are at different levels on the organizational structure. They have
different experiences, qualifications; this, coupled with all the participants, work with the
big four.
After participants agreed to take part in the interview, an interview guide was developed
and sent to all participants. Along with an information sheet containing additional
information (See appendix 2) pertaining to the right of the participants and ethical
consideration of the research process. Further, participants were asked to confirm if they
can answer the questions related to those indicated in the interview guide. After confirming
they are comfortable discussing questions related to the interview guide, each participant
was asked to choose a date between the end of March and the beginning of April. For
convenience purposes, all the interviews were scheduled online using video conferencing.
One interview was conducted in March while the rest were conducted between the first and
third week of April.
3.2.4. Interview Guide and Interview Process

As highlighted in the previous sections, the interview method of this thesis followed a
semi-structured interview with open-ended questions. Saunders et al. (2009, p. 326)
highlight reliability, validity, generalization, and biases as major concerns of data collected
through semi-structured interviews. According to Jacob (2012, p. 1) qualitative research is

37
concerned with studying people’s experience in context. Furthermore, Gill et al. (2008, p.
292) indicate the purpose of conducting an interview is to explore the experience, views,
beliefs, and motivation of others about a social phenomenon. Therefore, if experience and
the context under which the data is collected change, then findings cannot be generalized
or transferred (Saunders et al., 2009, p. 236).
To fulfill the purpose of the research, the authors read to understand the existing literature.
Quality data collection requires proper preparation, and the authors divide the topic into
themes to develop the interview questions. In each theme, a special focus was given to the
drivers of the theme. For example, when setting the possible question for audit efficiency,
the authors look at how big data analytics could affect the time and cost of an audit. The
questions were developed to serve as mechanisms to put the interview in context. The
question guide was divided into five sections. Section one contains general questions to
collect background information, section two focuses on skills and competence of auditors,
this is followed by auditor’s judgment, and then audit efficiency and the final section
contains the questions pertaining to audit quality (Appendix 3). It is important to point out
that these questions were only used as a guide, and they were not strictly followed. The
authors make on the spot adjustment to the base of the question on the information and
discussion of the interviewee. Follow up questions were asked to clarify or gain more
insights about pertinent issues highlighted during the discussion (Bryman & Bell, 2011, p.
475). The authors were able to identify key patterns and themes that were crucial to
fulfilling the research objective.
It is important when formulating interview questions, for a researcher to consider what
kind of knowledge will be essential to answering the research questions (Bryman & Bell,
2011, p. 475). The background knowledge about auditing and business, coupled with the
literature review, was very crucial at this stage of the research process. Jacob (2012) in her
research “Writing Interview Protocols and Conducting Interviews,” identified fourteen
points to consider when writing good question guides. These points were utilized in the
context of the research topic. When designing the interview guide, the authors were
interested in questions that will generate more information about the topic and contribute
immensely to achieving the research objective (Gill et al., 2008, p. 292). Contextualizing
some of the questions was a challenge, as each of the interviewees has a different
experience and utilize big data tools in a different context. A draft of the interview guide
was sent to colleagues of the authors for review in relation to the research questions and
the research objective (Bryman & Bell, 2011, p. 477). Discussions were held to improve
the questions. Before sending out the question guide to interviewees, the authors conducted
two test interviews with separate colleagues from those that reviewed the draft. The test
interview was geared towards helping the authors to prepare for the interview process
beforehand. Finally, the draft questions were sent to interviewees, and subsequently
interviews were scheduled.
Gill et al. (2008, p. 292) state that, the interview should start with the researcher reminding
the interviewee about the ethical considerations such as anonymity, the confidentiality of
the information received during the interview process. This will strengthen the relationship
between the interviewer and the interviewee. The interviewee will have the confidence to
talk freely, give an honest opinion about the phenomenon. Similarly, Jacob (2012, p. 6)

38
indicates that developing the interview guide alone does not guarantee a successful
interview. Therefore, she outlined ten steps an interviewer should follow for a successful
interview. Throughout the interview process, these guidelines from Gill et al. (2008) and
Jacob (2012) were adhered to. The authors reminded each of the interviewees of their right
to the information they provide during the interview process, how the information and their
identity will be protected. A consent to audio records of the interviews was obtained. The
interviews start with an introduction of the researchers and the purpose of the research. The
interview follows the thematic outline of the interview guide from the most basic
information and proceeds deeper into the subject matter with detailed information. Where
necessary, follow up questions to clarify certain issues that arise during the interview
process.
According to Bryman & Bell (2011, p. 481), it is important for the interview process to be
recorded using audio recording devices. To avoid the distortion of information, the
interview should be conducted in quiet and calm places suitable for the participants (Gill et
al., 2008, 292). In this case, all the interviews are conducted online due to distance and
convenience for both interviewees and the authors. Two interviews were conducted via
Skype, one through a telephone call, and the rest were conducted using Microsoft Teams
(see table 2 below). However, these requirements were fulfilled by both parties. Gill et al.
(2008) and Jacob (2012) highlight listening as the most important skill an interviewer
needs to conduct an effective interview. This is in line with Bryman & Bell (2011) an
argument for recording the interviews using audio-recording devices. Recording the
interview process allows the interviewer the opportunity to listen attentively and keenly
follow the discussion that will help him/her to ask follow-up questions. On the other hand,
Saunders et al. (2016, p. 411) suggest complementing the audio-records by taking notes
during the process. The argument for this is based upon the fact audio-records may not be
of high quality or even get lost. This is to make sure the interviewer will always have some
elements of evidence. The downside to this approach is that the interviewer may lose track
of the interview and miss key points worth following up. In this thesis, the authors use two
devices to record the proceedings. After each interview, they (authors) exchange the files
for making sure they are safely kept, to avoid completely losing the records.
After the interview, a researcher can decide to use the recordings directly or transcribe the
records into text format depending on the adopted research strategy (Saunders et al., 2016,
p. 416). Bryman & Bell (2011, p. 482) and Gill et al. (2008, p. 292) stress that it is very
important to transform the audio-recordings to text. Transcribing the recordings word for
word helps reduce biases and help the researcher interpret the information more accurately
than in the audio format. However, transcribing is a difficult process and time consuming
(Bryman & Bell, 2011, p. 483).
The process is a bit easier if the transcription is done immediately after each interview
when the discussion is fresh on the mind of the researcher (Saunders et al., 2016, p. 416).
Having these points in mind, the authors transcribed the audio-recordings into text word
for word. To ensure accuracy, the authors endeavored to do the transcription immediately
or the soonest possible time. Although this process required a lot of time instead of
splitting the transcription process, each record was transcribed by both authors separately
and reviewed both transcriptions together by reading both scripts and listening to

39
recordings several times. This was geared towards achieving accuracy and consensus about
what is said during the interview. All transcription documents saved separately, bearing the
identification titles with date, time, and duration of the interview.
Table 3: Participants interview details

Participant Position Interview Date Medium Duration

P1 Senior Auditor 2020-03-26 Skype 0:52:39

P2 Audit Manager 2020-04-03 Skype 0:50:51

P3 Senior Auditor 2020-04-04 Microsoft Teams 0:54:14

P4 Senior Auditor 2020-06-04 Telephone 0:42:44

P5 Senior Manager 2020-04-08 Microsoft Teams 0:47:43

P6 Assistant Manager 2020-04-14 Microsoft Teams 0:58:27

P7 Senior Auditor 2020-04-17 Microsoft Teams 0:48:50

.
The final sample of this study draw combination of participants from varying experience
knowledge, and expertise. The participants represent the lower and middle level auditors.
Auditors at these levels of management hierarchy utilized the BDA tools in their daily
audit activities. Furthermore, the selected sample work with big four firms. This means
also, they audit mostly large clients where the use of BDA is required to effectively
conduct audit services. As revealed by most of the participant, they audit either listed or
subsidiaries of listed companies in the largest cities in Sweden. However, some of them
audit public or government institutions. In contrast, the limitation or weakness of the
sample is the of lack representation from top management. Therefore, the thesis has not
captured the view of top management on the topic in empirical findings.
3.2.5 Data Analysis

Qualitative data is unstructured, unlike quantitative data, which is more structured in

nature. Qualitative data is complicated and difficult to analyze (Bryman & Bell, 2011, p.
571) this due to the nature of data collection (Saunders et al., 2016, p. 568). Similarly,
there are no clear-cut rules for analyzing qualitative data like quantitative data, which is
tested and measured before being archived (Bryman & Bell, 2011, p. 571). However,
according to Bryman & Bell (2011, p. 571) and Braun & Clarke (2006, p. 79) there are
three most quoted qualitative data analysis methods. These are grounded theory, analytical
induction, and thematic analysis. The grounded theory approach relied on finding the core

40
terms from the data to build a theory (Saunders et al., 2016, p. 594). In other words, the
grounded theory of data analytics disregards previous theories. While analytic inductive
involves formulating a research question from the existing theory and devising a
hypothesis for the research question (Bryman & Bell, 2011, p. 571). After the formulation
of the research question and explain the hypothesis, analytical induction then proceeds to
data collection. The researcher continues with data collection until all the possible
explanations of the hypothesis are achieved, and there is no new information to further
explain the hypothesis (Bryman & Bell, 2011, p. 571).
Furthermore, the grounded theory approach is geared towards theory building, whereas the
analytical inductive inclined more to theory testing. Therefore, these two approaches are
not suitable for this research. The intention of the authors is neither built nor test a theory.
As indicated in the research approach section, the authors will be switching back and forth
between the inductive and deductive approaches. Therefore, it is imperative to adopt a data
analysis method that will suit the research approach. According to Saunders et al., (2016,
p. 579), argue that a thematic analysis method is a generic approach to analyzing
qualitative data with features of both grounded theory and analytical induction. Hence, in
this research, the authors will adopt the thematic analysis approach.
The thematic analysis is the process of identifying classifying and reporting patterns to
contain within qualitative data (Braun & Clarke, 2006, p. 78). This process involves the
researcher coding the data set to identify patterns or themes for further examination and
explanation (Saunders et al., 2016, p. 579). Thematic analysis is very systematic and
flexible, and it can be used to analyze large volumes of qualitative data sets (Saunders et
al., 2016, p. 579). This method is systematic because it follows a logical and orderly
process; it is also flexible since it encompasses a process that is suitable for any research
philosophy (Saunders et al., 2016, p. 579). Braun & Clarke (2006) and Saunders, et al.
(2016) identify six steps in a thematic analysis approach that systematically follow each
other (see table 3). It is important to note that these steps are followed chronologically, but
research will be moving back and forth during the process (Braun & Clarke, 2006, p. 86).
However, it is more straightforward compared to other techniques used in qualitative data
analysis (Saunders et al., 2016, p. 580). Therefore, the researcher can spend more time and
energy in conducting a rigorous analysis. The researcher can present the analysis in an
understandable manner, building the reliability and validity of the data (Braun & Clarke,
2006, p. 93).

41
Table 4: Phases of the Thematic Analysis (Braun & Clarke, 2006, p. 87)

Subsequently, the authors kick starts the analysis process by transcribing the audio files to
text. The authors read the transcribed text to establish familiarity, taking notes of key
points. Further, the authors identify and select the most relevant to answering the research
questions. Next, the data was put in table form, reflecting the interview guide. This was
done to identify relevant data and to generate the initial data codes. Themes were
formulated according to the information obtained from the interview process. Following
the six steps on the table above, the authors continuously reviewed the themes to align the
themes with the research question. The continuous review of themes also avails the authors
with a better understanding of the data and themes. Finally, the authors are able to achieve
precise and logical data presentation. To keep the data presentation interesting and
coherent, the authors avoided possible repetition of the data.
3.2.6 Literature Search

In every scientific research will start with a review of the existing literature in identifying
the research gap and problematization (Boell & Cecez-Kecmanovic, 2014, p. 268). A
critical literature review involves a critical evaluation of the prior research papers ranging
from journal articles, textbooks (Saunders et al., 2016, p. 73; Cronin et al., 2008, p. 38). A
literature review is an integral part of the research process. It is the foundation on which
the research background, theoretical framework, and the findings are judged (Saunders et
al., 2016, pp. 70- 71; Rowley & Slack, 2004, p. 32). According to Saunders et al. (2016, p.
74) there are basic approaches to conducting literature review; inductive and deductive
approaches. The inductive approach is where a researcher starts with the literature review
and then builds theories from the literature. In contrast, the deductive approach begins with
theories and then collecting literature to explain the theories (Saunders et al., 2016, p. 74).
According to the above description, the authors adopted an inductive approach to the
literature review. A literature review was conducted in order to identify the research gap

42
and developed the research objective. Subsequently, theories were identified from the
existing literature, and those theories were used in the theoretical framework. The
theoretical framework was mainly developed using peer-reviewed journal articles from
reputable journal publications such as JSTOR, psycARTICLES, Business insights, Wiley
journals, SAGE, etc. This literature was accessed through the Umeå University library and
the Groningen University database. These databases were used because the universities
control the articles stored in their database. Most articles in the university database are
peer-review. However, other search engines such as google scholar were extensively
utilized as well.
According to Cronin et al. (2008, p. 40) keyword search is widely used to identify relevant
literature and faster in retrieving literature from the databases. Accordingly, the authors
make use of keywords to search for literature related to the topic. Key terms used in the
process include but not limited to the following: - Big data, Big data analytics, audit
process, audit quality, audit efficiency, professional skepticism, professional judgment,
auditors’ skills, auditors’ competence, etc. Furthermore, bibliographies from the key
articles were used in the search. Another strategy useful in assessing literature is the
combination of key terms (Cronin et al., 2008, p. 40); this strategy was also extensively
utilized.

43
4. EMPIRICAL RESULTS AND FINDINGS
In this chapter, the authors intend to present the empirical data collected through a semi-
structured interview conducted with seven auditors. The chapter first outlines the
background of the interviewees to serve as an introduction to the data. Then the data is
presented following the outline structure in the theoretical framework and methodology
chapters.
4.1. General bachground

This thesis on the impact of Big Data Analytics on audit efficiency, quality, and auditor
judgment and auditor’s skills and competence needed to work in this environment is
conducted by using a qualitative approach through semi-structured interviews to collect the
relevant data. The participants in this study were chosen from the big four audit firms in
Sweden. The big four audit firms were selected because they are dealing and using BDA in
their audit engagements in a wide range. Further, the individuals interviewed have the
knowledge and experience in BDA. Therefore, seven semi-structured interviews were
conducted between 26 March and 17 April 2020. Table 5 includes the descriptions of
interviews.
Interviews were tape-recorded, and their length ranged between 45 minutes and 58
minutes. The interview guide (see Appendix) was submitted to participants one week
before the interview date to cover all points in the given time and efficiently use the time.
The interview guide was designed to cover diverse topics to answer the research questions.
The interview guide includes a range of topics; general information about participants like
education, experience and their role in the firm they work for and how their roles are
connected to BDA; information about participant’s client such as large or small firms and
listed or not listed firms; the skills and expertise are needed for current auditors to work in
BDA environment; questions to find out the effects of BDA on audit efficiency, audit
quality, and auditor judgment. In addition to the interview guide, follow up questions and
other questions beyond the interview guide, but within the BDA environment were asked
to gain more insight about the topic.
Participant 1 is a senior auditor in a big four audit firm in Stockholm office with four years
of experience. The auditor has a master’s degree in accounting, auditing, and financial
analysis and will be promoted to assistant manager in September 2020. The auditor’s
clients are mostly Swedish subsidiaries of USA listed companies.
Participant 2 is a manager in a big four audit firm in Stockholm office with six years of
experience. The auditor has a master's degree in business administration and economics. In
addition to this role as manager, the auditor is a member of the audit quality group in the
firm. This group is responsible for enhancing work and developing audit methodologies
and ensuring that firms comply with ISA primarily and other regulations. The auditor’s
clients are primarily large listed companies within the manufacturing industry.
Participant 3 is a senior auditor in a big four audit firm in the Gothenburg office with three
years of experience. The auditor has a bachelor’s degree in business administration and
finance. The auditor’s work mainly dedicates to large and medium-sized companies.

44
Table 5: Participants background

Experience
Code Participant role Organization Educational Background
in Years
Master’s in accounting,
P1 Senior auditor Big Four 4
auditing, and financial analysis

Master’s degree in business and

P2 Manager Big Four 6
economics

Bachelor’s degree in business

P3 Senior auditor Big Four 3
administration and finance

P4 Senior auditor Big Four 4 Master's Science in accounting

Master of Science in business

P5 Senior Manager Big Four 15 administration and accounting /
CPA
Master’s in business,
Assistant
P6 Big Four 7,5 Economics, and auditing / CPA
Manager
in progress
Bachelor’s degree in business
P7 Senior auditor Big Four 4 administration and information
technology

Participant 4 is a senior auditor in a big four audit firm in the Gothenburg office with four
years of experience. The auditor has a master’s degree in accounting. The auditor’s clients
are mostly SMEs.
Participant 5 is a senior manager in a big four audit firm in Stockholm office with 15 years
of experience. The auditor has master science in business administration and accounting
and has a CPA as well. Currently, the auditor works as a data specialist in capturing data
and analyzing data especially, for large listed companies, and provides support to audit
teams. Further, the auditor provides training to auditors in the firm in big data and analytics
tools. The auditor provides auditing services for midsize clients and provides analytics
services primarily for large clients.
Participant 6 is an assistant manager in a big four audit firm in the Gothenburg office with
7.5 years’ experience. The auditor has a master’s in business, economics, and auditing and
has started in CPA as well. The clients’ auditors are big listed- companies.
Participant 7 is a senior auditor in a big four audit firm in Stockholm office with four
years’ experience. The auditor has a bachelor’s degree in business administration and
information technology. The auditor works partly on regular audit services and partly as an
analytics specialist. The clients are mostly government and public institutions.

45
Before asking the participants, any questions related to research questions, auditors were
asked how their role connected BDA, how BDA changes their role, and what kind of tools
and programs they use. The response of auditors varies due to their experience as some
auditors have not experienced any other type of audit to provide their opinion. All
participants are connected to BDA as ‘’the BDA now is part of modern society and part of
audit requirement’’ [P1]. Accordingly, auditor’s role becomes more connected to BDA:
‘’So actually, it is changed a lot as we are pushing our clients, and we are pushing
our audit to go toward this direction more for big data analysis’’ [P1].
‘’As part of the work that we have in the last few years moved more and more to big
data analytics as a part of those procedure’’ [P2]
‘’All teams are required to use data analytics as part of their audit tool’’ [P3]
The empirical findings related to how their role has changed or are going to change in the
future vary among participants. The participants 1 and 3 said their role have not changed
for different reasons:
‘’The role is not going to change, because the big data analysis is a tool. [..] only like
a tool that is going to help me to have a better audit or let us say that to understand
the client and their transactions in a better way, to have this visualization of the
transactions’’. [P1]
‘’It hasn't changed yet, because I joined as I said at the beginning around two years
and a half ago’’ [P3].
While other participants indicate that their role has been changed and are going to change
due to ongoing advanced technologies:
‘’has definitely changed, before we use big data analytics it was all about sample
testing and we're spending more time on risk assessment, we are spending more time
on understanding the data population before we actually dive in and selecting those
samples’’. [P6]
‘‘It has already changed. We have used analytics for many years on various
engagements. 10 years ago, we intensified our development of new tools and analytics
is now used on almost all of our audit engagements’’. [P5]
In relation to tools and programs, all participants, in addition to Excel as a basic tool for
analytics, use almost the same programs such as Microsoft Power BI, Qlik Sense, and
Tableau. They also use their own tools and program which designed in house or
incorporation with a specialist IT company in order to satisfy their clients need.
4.2. Skills and competences

Auditing today becomes more complicated and relies more upon analytics. As a result,
auditors should go beyond the understanding of business and industry and the sample set
of transactions (CFO Journal, 2017). Further, auditors deal with large volumes of data to
determine outliers and anomalies and to extract value from data and additional insight. In
46
order to deal with this type of data and applying new audit methodologies, auditors should
possess more analytics skill and competences according to participant 3:
“[..] it will require more analytical skills going forward to use big data in the audit.
Like in traditional audit, it has been usually done using tick and tie methods, like you
obtain some samples, and you take them to the general Ledger data, and it is a very
simple approach. But now considering using big data, you have more insights from
the analyzed population, and therefore, you would require some analytical skills and
mindset to understand this distinction [...]” [P3].
While other participants see that dealing with BDA in some stages like capturing and
downloading the data, coding, and preparing appropriate tools to use, they need some kind
of skills. These skills are necessary for the auditor to have:
“The data capture, data transformation and loading of tools is normally handled by a
team of specialist, so it´s nothing we expect from our auditors to handle themselves.
Instead, each audit team should have analytics skills as end-users in order to interact
with the various tools and be able to draw conclusions based on the data loaded […]’’
[P5]

‘’It requires a lot from our team actually like when it comes to power BI and this kind
of data analysis program. So here it requires a special kind of skill. It has a lot of
coding; it has a lot of very specific IT skills. For example, I do not have this kind of
skill, I can read the data, but I cannot put this information and make it readable. So, I
cannot do the coding myself, so we have a team we called an analytics team; they help
us to code the data and to make it readable to us ‘’[P1].
In addition to analytical skills, skills, and competencies related to accounting and finance
are still very important to understand the business and industry. These skills are more
significant than IT skills in this stage as BDA in its infancy:
“I would say the way that we operate right now, the understanding of the data is, of
course, more important than an understanding of the systems itself. But from where
we are right now, I would say that the competence is related to accounting or finance
as well, of course analytical competence as the general mindset and understanding of
the businesses is more important than actual IT understanding” [P2].
Further, analytical specialists are more required now than before to provide support to
audit teams. The analytical specialists also become more crucial for audit with larger
clients. They provide support and perform data input and assess IT system of clients.
‘’Usually, they operate more as a support function, supporting the audit team and
assessing that the quality of IT controls, quality of data and then the teams themselves
with the actual in-depth understanding of the business in conducting more analytics
and specialists are supporting us in designing the analytics or designing our
appropriate test of the input data’’ [P2].
‘’…. we have a special department for the biggest clients. They usually perform most
of the analytics procedures. They are always with the team to do the analysis’’ [P4].

47
 ‘’… It would be more helpful as well I think for these big clients to have a special
person in the engagement which they do the analytics for each engagement, and I
think this is going to be the future. So, we should have a special team to do the
analytics because it requires a lot of work, and it is not easy like big data analytics is
not very easy and simple work’’ [P1].
To find out if there are any outsourced analytical activities in the absence of analytical
specialists, all participants argue that they get support within the firm due to all of them
working for the big four audit firms. They get support from the headquarter or might be
from other offices.
‘’I do not know how it is actually done in smaller audit firms, to be honest. But since I
am working with a big audit firm, we have different functions. It is always possible to
find one or two persons within the audit firm who can help you out. I mean, we have
all departments; we have IT departments and so many different levels of expertise or
requisite expertise. So, I have only worked with outsourcing internally to other
functions within the audit firm’’. [P6]
‘’…. within the company, we can bring in people to perform certain tasks but not from
outside the company’’. [P7]
To develop and enhance the skills and competencies of the current auditor to be able to
utilize the BDA. Big four audit firms have periodic training programs. Currently, these
programs focus more on analytics and provide training on using some useful analytical
programs such as Qlik Sense, Power BI, and Tableau.
‘’We have quite detailed analytics training as part of our approved curriculum. it is
increasing in size or proportion with regards to other aspects such as just basic
accounting & finance and tax, etc.’’. [P2]
‘’To utilize big data analytics tools and programs, the skills required could be like
using Excel data analysis, Qlik sense that could also be useful for presenting the result
of different analysis. Therefore, skills or training in these programs is important.’’
[P4]
One participant argues that current auditors also need knowledge in the programming
language in order to create and develop programs.
‘’The first part is like creating the data itself creating the programs, and I think for
doing that it will require light data transformation courses in programming or such,
like programming languages like Python or SQL and so on and so forth’’ [P3].
While participant 6 believes that the best way for the current auditor to improve their skills
to utilize BDA is learning through practice and implementing client data.
‘’ What helps me is best practice examples. Because you can get educated in many
ways and you lot of people can tell you that this tool can give you this answer, we can
do XYZ. But what makes it easier for me is when somebody can say yes, we have this
tool, we have power BI, we have tableau, we have excel or other programs; but then if

48
 somebody can show me that we actually built this with this data, with this tool and see
actually see what the outcome of it is[... ] So yes, it might require some technical
training, but it also requires training in how we think and how our mindset is with
performing an audit’’. [P6]
With the growing use of BDA and relying more on analytics even in small engagements,
the audit firms may require more qualification in hiring new auditors. Therefore, in this
part, the authors try to find out the potential changes that academic and professional
curriculum could make to prepare future auditors. Two participants proposed that
education should gear more toward IT to understand the increased complexities of IT.
‘’ more IT-focused training and education are probably needed to understand the
complexities in our client’s organizations and IT environment. I think that the current
sort of accounting and finance background is important too, to interpret and
understand the analytics’’. [P2]
‘’The future is toward modern audit; modern audit means more digitalization and
more data analysis. This is the future. So, we have to prepare them as well before that
to add these courses into their curriculum, not only theories. But add these kinds of
tools because the future is towards modern audit’’. [P1]
Participant 5 suggests that education curriculums should have theoretical knowledge about
analytics in order to prepare future auditors to gain an analytical mindset. Because each
audit firm has its own analytical tool, therefore it is very difficult for future auditors to
train on specific programs. They should also learn and understand basic Excel as key for
any further development:

‘‘I think it would be good if the education contained theoretically knowledge and
examples of how analytics can be embedded in the different phases of the audit. It
might be hard to have education on specific analytics applications since these differs
between auditing firms and over time, but I would at least say that Excel is a very
good start for most people’’ [P5].
While participant 6 sees that some Universities have already made some changes and have
started teaching big data courses for business students to be able to meet job market
requirements:
‘’We already see support in terms of education [ ...] they have like courses now in
big data how you should work and understand it. So, I think from an education
perspective; we are doing a lot already now’’. [P6]
4.3. Professional Judgment

Professional judgment is the most critical issue in any audit engagement and is most likely
to be used by auditors in all audit stages to make sound decisions. It is a subjective issue
and depends highly on the auditors’ skills, competences, and experience. Further, the
international auditing standards encourage professional judgments in implementing them
because these standards are principles-based that provide general principles and guidance
and give professionals more room for judgment. In this part of this thesis, the authors

49
would like to explore the impact of BDA on auditor judgment by asking the participant
what the effect of BDA on auditor judgment and decision-making is? All participants said
that BDA has a significant impact on professional judgment:
‘’The professional judgment aspect has definitely increased. Again it's all about how
you understand, how you read, and how you assess the outputs based on the analysis
that you have built.[..] I think big data analytics has definitely increased the
importance of skepticism as an auditor’’. [P6]
‘’Big data analytics will help the auditor to be more comfortable to make the
judgment. Because you can use the data analysis in the planning phase of say and
when you do the planning phase of the audit and then you do data analysis, and then
you would know really where do you have risk, and then you will do like the focus
area more efficiently. [...] You will be more comfortable because you are looking up at
the whole population just to summarize it for you’’. [P1]
‘’it has a very significant effect in looking at what is judgment and decision-making.
We can now make a more fact-based risk assessment and really do a more focused
audit. We are looking at trends and unusual patterns. Rather than just looking at a few
samples of invoices, for example, in revenue. We can now audit the entire population
and focus on the outliers and items and questions that stand out from the norm’’. [P2]
‘’I think it is a lot much different than a traditional audit. It also helps to understand
which accounts risky and which accounts are limited risk or less risky than the others.
It will create less subjective judgment because you have the results in front of you. [...]
It has a lot of effect on judgment now and also in the future’’. [P3]

"Professional judgement is always and key and very important when evaluation the
results of the audit procedures, both with and without the use of BDA analytics. BDA
have the potential to highlight new findings and risks that was not identified before
when using traditional audit procedures and these will have an effect on the
judgments and decision making the auditor will have to make.” [P5]

Auditors make their decisions and professional judgment based on verified data; therefore,
in this regard, auditors were asked how they verify the reliability and veracity of data. The
answers are similar as they check the completeness and accuracy of data and the extent and
scope of procedures depend on engagement size:
‘’Working with data analytics, we are required to check the completeness and
accuracy of the data we are using. So we are always checking the data that we are
using regardless if you are using data analytics or not and that can be checked
against data output or trial balance that we have received from the client, also check
it against other obtained data sources which we have received….’’. [P6]
‘’…. you have to always check the accuracy and the completeness of the information.
So, this is, of course, based on what type of data you are requesting from the client,
but every part we received from the client, everything is checked for accuracy and
completeness tests’’. [P1]

50
 ‘’We do a completeness check for the data we obtained from the client itself. If we
obtain general Ledger data, trial balance with all Journal entries, everything that has
happened during the year. We do a completeness check by reconciling the data too to
the documents we obtained from the client….’’. [P3]
‘’There are some prerequisites or expectations from my side, either regulations that
require us to ensure the reliability of important information basically obtained by our
clients. Depending on what information we receive, of course, we tailor our testing
based on that. But if we get a massive report with a lot of data, we will usually test it
through either reconciliation to the general ledger or sample testing against invoices
or contracts.’’. [P2]
When then asked auditors, what are the effects of large volumes of data on auditor
judgments in relation to information overload and irrelevant information? Three
participants responded to this question as follows:
‘’it depends on the size of the client, because again if you are working on large clients,
you can get so much information. I mean, you are using the word information
overload that is definitely the case. Sometimes you receive so much, and, in the end,
you end up spending too much time on understanding other parts of the data that are
actually irrelevant for your audit procedure’’. [P6]
‘’Before you do analytics, the key thing is what I would say, upfront to know what
procedures you are going to perform because it is very easy to get lost in the details.
[...] We need to decide this is what we are looking for, these are the risks we are going
to look for, skip the rest’’. [P5]
‘’...the information overload, if you don't know what you need or what to look at or
understand, I believe that the most common thing that happens especially to newly
hired persons that are working with analytics. [..] Of course, if you do not use the tool
and analysis right, you get irrelevant information, you are looking at wrong things. I
think that is the things that could happen or potential mistakes that would happen’’.
[P7]
Another question, the authors asked how auditors reach the conclusion that the gathered
audit evidence is appropriate and sufficient by using big data tools. All the participants
point out that this conclusion is typically based on professional judgment and experience
alongside what evidence is gathered during the process.
‘’This is somehow complicated, of course, and based on to some extent on
professional judgment. However, there are guidelines on how to perform analytical
reviews as part of the audit and how to determine those thresholds of expectations in
our analytical procedure. So basically, based on those guidelines in combination with
professional judgment, you formulate some sort of expectation, and then you compare
that to actual’’. [P2]
‘’…. but it is again all about professional judgment. It is very hard to give you one
answer that when you do this, then it is enough. It is all about a specific case and how
that output is aligned with the other outputs you have’’. [P6]

51
 “We have various supporting documents for what auditor should do, what the best
practice is when performing different tasks etc. But everything will at the end of the
day come down to professional judgment. [..] and I think that is one of the challenging
parts for auditors when applying analytics, to be able to decide when enough is
enough”. [P5]
Further, follow up questions were asked to some participants about the impact of non-
financial data and semi-structured data as a part of big data on auditor judgment. 3
participants responded that they make use of non-financial data in their audits, two believe
non-financial data is not relevant for normal audit procedure but mostly use in fraud
investigation, while 1 participant argues that the tools don’t incorporate non-financial
information as of now.
‘’It is also based on the client and the engagement we have. Some engagements we do
follow up like for a newly listed company, we usually use such information to check
for going concerned’’. [P4]
‘’We can say one very relevant situation; the situation we are in now with the Corona
situation (COVID-19). That also is something that you have to bear in mind when
performing risk assessment’’. [P6]
‘’The effects are definitely that we can be more fact-based in our audit, and we can
focus on unusual trends or patterns that are unexpected for some reason. Also, to see
if there is other information available which we can use either as contradictory
evidence or evidence supporting our expectations or the trends that we are observing
in the data obtained directly from our clients’’. [P2]
‘’COVID-19 now, that will affect a lot of companies. That is a kind of information that
you bring into the audit process. Basically, not yet, but I think we will be there in a
couple of years, that we will use more non-financial data into our analysis’. [P7]
Moreover, the authors asked the participant if the BDA eventually can replace or
standardize the auditor's judgment. All auditors said that in the near future BDA or other
advanced technology like AI and robotics are not expected to fully replace or standardized
the auditors’ judgment
‘’I do not think data analytics or other robotics etc. will completely replace the need
for auditors. Because there are a lot of areas in which there are a variety of needs for
management assessments, assumptions and estimates and so on[...], We will for sure
see more analytics-based or potentially even robotic base audits, but auditors will
instead focus on the bigger judgments and estimates’’. [P2]
‘’Of course, you can do standard things, you can do standard checks and writing
things is fine. Then maybe it's correct, but at the very end of the day, I would say it
still needs humans to decide like does it make sense depending on other information
we know about the client or what we heard’’. [P5]
‘’when we get more used to working more with big data analytics when it gets more
standardized, then I think that you can then conclude faster on information received.

52
 But again, I do not ever think that we will get away from this profession adjustment’’.
[P6]
‘’I think it is very early to say that the robots and the machines will replace auditors
at least not in the five upcoming years’’. [P3]
4.4. Audit Efficiency

In order to explore the effects of BDA on audit efficiency, the questions were asked on the
impacts of BDA on time spent on engagement activities, audit cost, audit fees as the main
drivers of audit efficiency. The auditors gave similar responses to the time spent on an
engagement audit that will increase in the first year of implementing BDA.
‘’In the first year, I would definitely say it is an increase in time. When implementing
big data analytics, there are quite many considerations that an auditor must take when
making this decision to move from sample base procedures to analytics, and it usually
takes some time’’. [P2]
‘’The first year of your audit, especially if the audit team is not very used to it because
we are in the transition phase. Now we have been doing it for several years, but in the
beginning, it takes more time. Because the audit team was not used to doing this kind
of audit’’, [P5]
‘’The first time you start working with big data analysis, it will take longer than the
first two or three years’’. [P4]
‘’In the beginning, the first year when using big data, it takes more time to do the
audit, then it seems more ineffective because you need to implement all of the things
that you need to understand the analysis, etc.’’. [P7]
‘’But with the big data analysis, it will help you, maybe it is painful at the beginning
because you have to ask the right questions to receive the right data’’. [P1]
Further, most auditors see that the implementing BDA in next years will be more efficient
and it decreases the time spent on audit engagements
‘’ ...once you have done it once or twice, then it would be much easier, and it will be
much more time-efficient for us’’. [P1]
‘’[...] then going forward, it will reduce significantly. Because you know how to use or
update these analytics tools and how to shift from traditional audit to analytics audit.
Then over the next few years, the number of hours spent in the engagement will
reduce’’ [P3]
‘’...after this, it would be really easier and more efficient to perform the data analytics
and do auditing than performing the traditional procedures’’. [P4]
While participants P2 and P5 observed that implementing BDA in the coming years may
have the same time as traditional auditing, and the main purpose of BDA is to improve the
audit quality instead of reducing the time spent on audit engagements.

53
 ‘’Hopefully, the years after, they will be maybe the same time maybe just a bit more
efficient. But I think the main goal with analytics as we apply it is not to reduce the
time, I would say it is more about the quality and to bring something more to the
client’’. [P5]
‘’So far we haven't really seen the big effect on time but rather a big quality increase,
I would say’’. [P2]
Participants were asked to determine the impacts of BDA on time spent in terms of
percentage. The answers came very diverse, and the participants found it very difficult to
determine a specific percentage, and, in some cases, they gave a very rough percentage:
‘’...it is depending on case by case and the risk level you have in a specific
engagement. So, it is very hard to say. I do have some cases, wherein this specific
case, when using analysis, I think that we can save like I do not know 30% or 40%,
but in other cases might be 10%. In some cases, you might spend more time because
there are areas that we were not aware of when not performing data analysis. So, it is
very hard to say’’. [P6]
‘’I cannot really give you a specific percentage because it really depends on the
engagement. It varies from engagement to engagement. So, I cannot really give you an
exact percentage. Maybe between 10% to 40% or so’’. [P4]
‘’I would say the first year, it will increase probably by 10% to 15%. But then down to
normal it will decrease by 15% to 20% when you have everything in place’’. [P7]
When asking the participants how BDA affects the audit cost and audit fees, answers were
more diverse. The diversity is based on their position and ability to reach out to some
information like audit fees.
‘’The audit cost at the moment is increasing due to all the investments that we are
putting in our audits. [..] The fees are a bit harsh for me to comment on, that is above
my level, to be honest’’. [P7]
‘’I think right now, we do not really see an increase or decrease in audit costs or audit
fees. I do believe, however, in the future that there might be an efficiency gain, and
then cost might go down a little bit when we streamline the big data analytics
procedures even further’’. [P2]
‘’I would say the audit fees would be the same; I know some clients told us that if you
do things more efficiently with data like this, then we can reduce cost and time. But
often it is not that much efficient doing the thing, but we can do a better audit with
more quality. Since it ends up with the same things at most times, it is not more
expensive. [...] Initially, I think it will be more costly because you will invest in a lot of
things, you develop tools, and you may buy more licenses for those tools’’. [P5]
‘’It would not make it cheaper, definitely. I would say the same cost or a bit more,
and it depends on the market’’. [P1]

54
 ‘’Audit costs basically will be less. [...] These tools help create more insights for the
client, and the client will be willing to pay for the insight. Thus, increase your audit
fees because they are creating more insights to their clients compared to what clients
or companies’ employees are doing. I think the audit fee will increase over time if
analytics are used; also, the number of hours in engagements will maybe be less; so, it
would be more profitable I think for the audit firms’’. [P3]
Participants were asked if the executing full population audit is more expensive than
sampled-based auditing.
‘’When they spend time, and if you spend more time, then, of course, it would be more
expensive. But hopefully, in the end, it would be the same or a bit better depending on
where we went off’’. [P5]
‘’I think the big change so far is in quality. We can look at the whole population in a
very much more efficient way than prior samples. When doing the analytics, we focus
on the outliers and exceptions. This means that we take back that initial efficiency gain
in actual complex audit practices’’. [P2]
4.5. Audit Quality

The quality of audit reports has been heavily criticized in the last few decades. The
reliability of audit reports is crucial for decision making. To ascertain the impact of BDA
on audit quality, the authors dissect audit quality into its three drivers: input, process, and
output. This section will discuss the processes and output to measure the impact of BDA
on audit quality. Among the questions asked relevant for this section include, the impact of
BDA on the audit process, risk and material misstatement identification, effectiveness of
engagement team in using BDA tools during the audit process, reliability of the audit
report by using BDA tools, and how BDA affect the interaction among stakeholders during
the audit.
When asked how BDA affects the audit process and procedures, participants gave similar
responses. Participant 1, 5, and 7 indicate a test of internal control are not relevant in
performing analytical procedures:
‘’For controls, we have to exclude that, it has nothing to do with big data analysis.
Controls, we never use it’’. [P1]

‘’Our analytical solutions are currently not designed to test and evaluate controls,
instead analytics is more to be seen as a substantive testing approach”. [P5]

‘’…. but not that much on control testing, because that is another approach. Then you
don't use that much data analytics’’. [P7]
While other participants did not mention the effect of BDA on internal control testing, they
did share the view that data analytics differs in many ways with traditional auditing
methodology. These differences include sampling, substantive testing approach, and risk
assessment. All the participants indicated that BDA reduces the chances of selecting huge

55
unnecessary samples that will require a lot of time both from the auditor and client to
process:
‘’….it affects the audit procedures but basically, the sampling […] when we use data
analytics, I would say it will get you less sample testing’’. [P7]
‘’It impacts the audit in a way that we can test the full population instead of just
looking at samples, and we can identify trends and unusual transactions where we can
spend more time and focus on identifying outliers that are due to errors either by
fraud or mistake. I would say that the larger impact that we can get is a bigger
coverage in our audit procedures’’. [P2]
“Data analytics procedures provides us with alternative ways of gathering audit
evidences and to identify outliers and it reduces the need of the random sampling that
in the past could end up with very large sample sizes.’’. [P5]
‘’...the number of samples will definitely decrease, but again, we are spending more
time in the earlier stages in our audit than we did before’’. [P6]
One participant argue that they are not using sampling along with the analytics tools; they
either use the analytics approach or sampling approach:
‘’I do not recall performing big data analysis and sampling for the same audit. I
usually do perform either the big data analysis or the sampling’’. [P4]
This indicates that sampling processes differ because now auditors can identify exactly
transactions that look abnormal and select those to perform further investigation.
The views that surface relating to the substantive procedure represents a consensus. All the
participants agreed on the use of analytics in substantive testing procedures
‘’…...in data analytics it is substantive testing that is what we do……’’ [P5]
‘’...If you are using data analytics, you will make a more substantive approach; Which
is testing 100% of the population.[....]...if you are using analytics as your question
suggested you will do substantive testing….’’. [P3]
‘’… again, we are using data analysis in risk assessment and substantive testing’’.
[P6]
To confirm which phase of the audit is affected more, a follow-up question was asked at
which phase of the audit process are the BDA tools used more. This question received
mixed answers, some indicating that tools are used in the planning phase of the audit
process, while others responded that they use tools both in the planning and execution
phase. However, none use the tools at the completion and reporting phase.
‘’I would say it is still mostly used in the risk assessment process and then in the
planning stage of the audit. But for areas like revenue, for example, it's getting more
and more common to have substantive analytical procedures as well as in the

56
 execution part of the audit where we move away from sampling procedures to
leverage on analytics for the actual conducting of the audit’’. [P2]
‘’...We use it in two phases, planning and execution [..]. However, the real work is in
the execution phase’’. [P3]
‘’I would say it might be more in the executions phase because that is where you
spend more time, but it helps you in the planning as well…’’. [P5]
‘’We used the analytics tools throughout the whole year, but I will say it's more
common during the year-end audit’’. [P7]
Risk assessment is one of the most difficult and important aspects of auditing. To ascertain
if BDA could help auditors in risk assessment and identifying material misstatements. All
the participants confirmed a positive impact. According to the participants, BDA helps in
identifying risk and material misstatements, citing a full population test, which means all
transactions are tested. It is easier to identify risky areas and misstatements. However, one
of the participants argues that if the individual misstatements are not material, they can slip
through even if the aggregate is material. Nevertheless, they employ traditional methods to
identify material misstatements depending on planned materiality:
‘’ ...looking at the first part to identify risk, big data analytics is nowadays really big
and important aspects of our risk assessment procedures. Where we can disaggregate
the population into subpopulations based on homogeneity... ‘’. [P2]
‘’You can definitely use it to identify risk and material misstatements’’. [P6]
‘’Data analytics is about insights, and insights may lead to misstatements or
discovering something that the client may not be aware of’’. [P3]
‘’I would say it has a really high probability of identifying any existed material
misstatements because, as I said, you are testing the whole population, not always the
sample’’. [P4]
‘“Looking at full population of data help us with risk identification and to identify
outliers that accumulated could lead to material misstatements, outliers that we
potentially could have missed when using traditional auditing procedures”. [P5]

Participant 5 was asked if they are still performing materiality planning in there; he
responded that part of the audit has not changed. The only thing that in that aspect is the
metrics used.
‘’This is not any difference; the only changes are on the metrics depending on the
client or what's going to be our thresholds of things; that depends on the client’’. [P5]
Users of financial statements rely heavily on auditors' reports for decision making when
asked how BDA is improving the reliability of the auditor report and information therein.
‘’I am a firm believer that big data analytics improves the audit quality, in terms of
conducting a more thorough and fact-based risk assessment. Which basically outlines

57
 what we are looking at and why and also in what way we are doing it. Secondly, also,
I think we will see a higher number of misstatements identified in the end when
leveraging on big data analytics throughout the audit process’’. [P2]
‘’it is true that going from a sample mindset to testing the full population, in the end,
might give more confidence of saying that the population is correct…. looking at the
internal organization of the clients, it definitely increases the quality, and they also
believe it increases the quality’’. [P6]
‘’it has a positive impact on audit quality. Because as I said before, it gives you a
really good picture, and you look at it from the above, and then you see it and then
you know here is where I need to focus’’. [P1]
‘’It's just different ways to end up with the same conclusion in most cases, but then, of
course, if you use analytics, you might see things that you did not see in the past,
depending on the situation. So maybe you can identify things that slipped through in
the past, but we do not hope for it because it is supposed to be good in both ways
also’’. [P5]
‘’I would say the audit report is not affected by the analysis. Because to sign the audit
report, we perform the big data analysis or do the other substantive audit procedure.
So, I would say the reliability of the report is not affected by big data’’. [P4]
Participant 3 could not adequately comment on the issue relating to the audit report.
The effectiveness of the engagement team is a vital contributor to audit quality. Changing
the approach and tools is bound to impact the effectiveness of the engagement team. To
examine how implementing the BDA tools in modern audit contributes to audit quality, the
authors asked how effective team members are in using the BDA tools. Basically, the
responses link these questions to team experience, client ability to provide the required
data, and skills & competence possessed by various engagement team members.

‘’Our auditors have used analytics for several years and are getting used to it in their
day to day work, but of course there are always some differences between teams and
offices.” [P5]

‘’We're getting better and better. [..] But again we should also keep in mind that not
everybody should have the role of working with big data, not everybody should be an
IT specialist when performing an audit we still need to have an old mindset, somebody
to focus on areas rather than just doing to extract understanding data and such’’. [P6]
‘’We are getting more and more effective. We are getting more used to it, but still, it is
fairly new in the audit process, although we always used it to some extent…. It is
improving every year’’. [P2]
‘’ We really try to be more effective, and everyone pushes for it these days, even
partners...we are effective at using it...I would love to use it every time, but the thing is
the client effect gives us this data. This could be very challenging for some clients’’.
[P1]

58
 ‘’For new team members, it might be really hard to understand the analytics, but after
one or two years, you have more experience and more understanding of the data
analytics tools’’. [P4]
‘’I don't think we have enough people that are good at using the tools’’. [P3]
‘’One thing that we're trying to do is to have more training that is mandatory going
forward, and we have started with it now; it's mandatory for first-year auditors to do
specific data analytics courses’’. [P7]
To understand how the tools, change the composition of the engagement teams, the authors
asked the question if the engagement team members are required to possess similar
knowledge or skill on BDA tools. The response received indicates that it is not necessarily
required for all the team members to have the same level of skills. However, it is important
for everyone to have basic knowledge and understanding of how to work with tools. Most
participants highlighted the auditing process is teamwork, each member of the team will
perform a different task; therefore each member is required to possess different skills and
expertise that related to their role but for the purpose of coherent efforts and effective
communication, having background knowledge of data analytics is important.
‘’Definitely not, like you don't need to teach all the team members the same skill’’.
[P1]
‘’Usually, the team members complement each other’’. [P4]
‘’Sometimes I compare it to like a football match, somebody is the striker, somebody is
the goalkeeper, somebody is the midfielder; you all have different roles on the field’’.
[P6]
Another output to measure audit quality is the interaction between auditors and key
stakeholders such as management, audit committee, and those charged with governance.
To gain insights into how BDA affects the interaction between the various stakeholders
during the audit process, the authors asked participants how BDA improves the interaction
and communication between the stakeholders such as management and audit committee.
‘’I would say the big data will affect the relationship with these different stakeholders.
I mean, when you perform the analysis, you also request fewer supporting documents,
which also means less work for our clients’’. [P4]
‘’Clients can upload the documents; they could see the due date of obtaining these
documents and so on and so forth…… We use interactive tools to obtain documents
from clients instead of obtaining them by email’’. [P3]
The use of analytics can in some cases be a game-changer where we can bring
additional insights to management. E.g. by looking at the relatively simple GL data,
we can bring out much information that clients may not think about themselves’’.
[P5]

59
It is easy and also more interesting for the stakeholders to be provided with insights
when it is visualized in a smart and cool way, and that is what big data can do for you
[P6]
‘’Of course, they want to know how we are using big data analytics, and how we
structure our analytics to ensure the quality of their financial processes. So, it's
definitely a big topic on the agenda and based on the analytics we are conducting and
the visualization tools in particular which provide more value-adding information to
the business management and audit committees’’. [P2]
‘’It gives a very good improvement, because when you go to the audit committee and
show them like graphs from their own data. [...]All this gives a very good impact, then
the client is impressed. This is a modern audit which is like I said before, everyone is
trying to go through this direction’’. [P1]
‘’I don't think it will improve the interaction with all the stakeholders actually; it
improves more for management in the company’’. [P7]
‘’Another thing could be like when we perform the big data analysis the result could
have diagrams or charts, or visualizing could help to understand their business. So,
our results could be like a tool for them’’. [P4]

60
5. ANALYSIS AND DISCUSSION
In this chapter, the empirical data from the semi-structured interview interpreted analysis
and discussed it in detail. As mention in the methodology chapter, the analysis follows a
thematic approach, as suggested by Braun & Clerk (2006). The data is further compared
with theories in applying pattern matching. This chapter also followed a similar outline to
the theoretical framework and empirical data. At the end of the chapter, a summary
finding is provided in table 4.

5.1. Skills and competence

5.1.1 Skill sets required to perform an audit in a BDA environment

Advanced data analytics is considered a disruptive technology that has a huge effect, such
as a paradigm shift on the way of conducting a financial statement audit. As a result,
incorporating big data analytics in financial statements audit requires many changes,
among these changes, skills, and competence of auditors (Alles & Gray, 2016, p. 46). The
participants share a similar opinion that increased analytical skills and IT knowledge are
the most important required skills to work within the BDA environment. According to
Earley (2015, p. 497) analytical skills and pattern recognition to evaluate anomalies
represent a challenge for auditors as these skills have not been in accounting education; as
well they need many years of experience to be acquired. Moreover, all participants added
that Excel is the basis for using analytical programs and developing their capability to use
more advanced programs such as Power BI and Qlik sense. This observation of
participants is consistent with Alles & Gray (2016, p. 52) that auditors should have skills
and competence beyond the usual skill of Excel to use BDA appropriately. Further,
participants mentioned that the auditor should have advanced IT skills like programming
language to create and develop accounting programs.
However, economic knowledge such as accounting, company law, IFRS and auditing,
critical thinking, and business acumen is fundamental and more weighted than IT skills to
understand business environments and interpret anomalies and patterns according to one
participant. This finding aligns with (Richins et al., 2017, p. 76) that accountants’ financial
knowledge and strategic planning accompanied by their business acumen are the most
important skills to interpret unstructured data. In this regard, ICAS (2016, p. 9) determined
the key skills that auditors need to improve to cope with challenges of today and future;
business acumen; technology and data interrogation; and soft skills as the ability to
implement professional judgment and skepticism. As a result, it is observed from the
literature and empirical results that auditors require to have a skillset. This may include
essential economic knowledge, critical thinking, and professional judgment as to be the
key skill sets the experience as well with technology and analytical applications that
become to be more significant in a BDA environment.
Audit firms in their efforts to apply BDA widely, the main challenge is the lack of skills
and competences of their auditors. Therefore, audit firms assign rigorous training programs
to improve the skills and competences of their current auditors. Most interviewed auditors

61
leave the same feedback that audit firms have training programs for all auditors with
different positions to improve their skills in basic accounting, IFRS, and auditing.
However, the current training program focuses more on IT knowledge and analytical
application with a noticeable increase in training in terms of size and time in BDA,
according to some participants. Adequate training played a significant role in developing
the skills to deal with traditional data, and it will continue to play the same role in adopting
BDA (Brown-Liburd, 2015, p. 459). Moreover, consistency with this finding, the IAASB
(2016, p. 11) states that: “From the most experienced to the least experienced auditor, and
from the largest accounting firms to the smallest, changing the auditor’s mindset to
gathering audit evidence from the use of Data Analytics compared to traditional
techniques will require time and investment in training.” In contrast, one participant sees
that learning through doing and tracing the outcome is more important than theoretical
training.
5.1.2 The need for a data analytics specialist

Today, auditing becomes more complicated and requires skills and competencies beyond
the accounting and traditional auditing methodologies. It observed the increased use of
specialist staff in supporting audit teams in capturing data (FRC, 2017, p. 16). Further,
FRC (2017, p. 16) stated, ‘’There is no doubt that the firms have individuals and
specialists with extensive technology skills’’. Accordingly, the participants indicate that
incorporating BDA in auditing required more specialists in the audit team to provide
support to auditors in specific areas related to data. As well as, the auditor mentioned that
technical expertise and data specialists become an essential part of the audit team to
provide the support in capturing and downloading the data, coding, and preparing
appropriate tools to use in conducting audits. This finding in line with FRC (2017, p. 16);
having a data specialist in the audit team is very important to make the audit team more
efficient in obtaining entity data and more motivated to use analytical tools. Furthermore,
using the expertise from other fields as a part of the audit team to make professional
judgment has been started historically. However, a concern raised that tension could arise
between auditors and data specialists related to their role in the daily audit process (Salijeni
et al., 2019, pp. 109-110).
In addition to the task as mentioned above of data specialists, some participants indicate
that data specialists could organize workshops and contribute to teaching auditors how
they can utilize BDA and provide support in large client engagements. Further, it can be
understood that data analysts can provide training to audit teams within firms. This finding
is consistent with ICAS (2016, p. 15) that not all audit team members should have the
same knowledge. Instead, specialists could be offered with tailormade training on areas
such as professional ethics and judgment to enable them to integrate within the audit
profession fully. Thus, they can upgrade their contribution and effectiveness on audit
engagements. Moreover, all participants report not being aware of data analytics specialists
outside the audit firm being brought in to perform any specific task. This shows that audit
firms are able to train auditors within the firm to perform all the required analysis. On the
other hand, it can also be considered that the tools have not developed to a level beyond the
scope of auditors, or audit firms are conscious of the high cost involved in recruiting data
specialists on a temporal basis.

62
5.1.3 Potential barriers to Future Auditors

Several reports have shown that BDA in auditing is at the infancy level, and it will
continue to evolve. The environment on which audit is carried out is rapidly changing, and
this change means the role performed by auditors will change as well (BDO, 2016, n/a).
The complexity of the audit environment will require audit firms to recruit talent from
diverse backgrounds such as mathematics, engineering, technology, etc. Furthermore,
understanding advanced excel, statistical packages, and computer programming languages
will be crucial in the recruitment process (Salijeni, 2019, p. 175). Most of the participants
highlight that advanced excel is a vital tool in the auditing process, and audit firms will
seek to employ candidates with a prerequisite in excel and other analytics tools for
employment. Consistent with Salijeni (2019), one participant indicates the demand for
more IT technicians in the audit field will grow to a level where knowledge in business,
accounting, or finance will not be enough for future auditors. A participant argued before it
was not required for an auditor to have any knowledge about technology, but the trend has
changed over the few years. What the future hold is for technology in audit but to what
extent is not clear, because the development is just starting. Another Participant also
mentioned the importance of understanding programming language and coding as a huge
part of data analytics. This is in line with Salijeni (2019) that language programming and
data science are considered an advantage in data analytics.
Similarly, Earley (2015, p. 499), finds that future auditors need to be better data scientists
to be able to fit in the audit environment. Having an analytical mindset will make future
auditors ideal for the job and more suitable with the tools (Tschakert et al., 2016, n/a). The
participants shared similar views. An emphasis on the analytical mindset was manifested
throughout the interviews. Audit firms will be only willing to recruit those with a keen
interest in technology and are up to date with the happenings around the technologies
relevant to the industry. According to one of the participants, their firm has already started
hiring those with an analytical mindset and an interest in dealing with the technology. This
is just the starting point, as the evolution of data analytics in audit is just at the initial stage.
Things will get more challenging for those already taking a career in the profession but is
more likely to hinder those accounting and finance backgrounds with no mindset or
interest in the technology. Therefore, it is important for those aspiring future auditors to
take note of the development and the potential barriers to their goals.
Furthermore, the lack of technical skills such as data visualization tools is a cause for
concern for current auditors, as indicated above. However, this should serve as an alarm
for future auditors to invest in developing their technical skills ahead of time. According to
Tschakert et al., (2016), technical skills are required to understand the data and know-how
to manipulate to produce the desired results. Accordingly, the information received from
the interviewees asserts that technical knowledge and competence in visualization tools
will serve a great advantage for future auditors. Participants are of the conviction that
having a record of having the skills and knowledge on applicants CV gives an edge over
other candidates, “it will be really good for the person if he has on the CV before he starts
working as an auditor” (P1). This is also in line with Earley (2015, p. 499), data
visualization tools are among the most desired skills and training on pattern recognition,

63
and critical thinking should be encouraged among auditors particularly those looking
forward to entering the industry.
Considering the argument above, it can be assumed that these points, if not mitigated, will
serve as an impediment for accounting and auditing students to make it to the auditing
field. These findings are significant as it outlines the combination of skills and competence
required to perform an audit in the future. It is clear both from previous literature as well as
this research that auditors with business and accounting knowledge will continue to play a
significant role, but there is going to be a combination of different sets of skills, and the
above-outlined skills will take center stage. Therefore, to break the barriers, a shift from
the myopic view of the auditing field is required.
5.1.4 The need to redesign the academic curriculum for Accountants

Findings from the empirical result suggest that academic curriculums are not keeping pace
with technological advancement and the requirements of the current job market. This
finding aligns with that of Cao (2015, p. 428) and Earley (2015, p. 499) that the academic
curriculum has not made room for recent development technology to prepare students
enough for the job market. All interviewed participants said that when their firms hire new
graduates, they have to go through a series of training because they have no background in
the tools being used by the firms. A majority of the participants believe the academic
curriculum is not up to date with what is required by employers. Most importantly,
students are not fully aware of the trending issue around the auditing field, making it more
challenging for graduates to adapt to the complexity of the environment and the tools being
used. However, one participant argues that some universities have already introduced
analytics programs in their curriculum; “At least in the school that I was studying, I know
that they have like courses now in big data” (P6). Traditional auditing test books and
related literature need to be up to date to be able to address the skill gap (Early, 2015, p.
499). Nonetheless, participants said the knowledge in accounting and finance would
continue to be relevant in the foreseeable future. Still, there is a paradigm shift that needs
to be addressed in academic institutions.
Academic institutions play a major role in the propagation and dissemination of
knowledge. Alles & Gray (2016, p. 57) call research on how the accounting curriculum
could be changed to address the shortage skills required to perform audits using analytics
tools. Earley (2015, 499), warns that the accounting curriculum does not need a total
overhaul but rather redesign to accommodate changes in technology. Participants in this
study share similar opinions; all participants highlight that the current curriculum only
requires limited adjustment to make room for the technologies being used today. This is
also in line with the findings of Rachins (2017). According to data received during the
interview process indicates that similar analytics tools are being used across the industry as
well as audit clients. Programs such as excel analytics, Power BI, Qlik Sense, and tableau
are widely used. According to Earley (2015, p. 499), training on pattern recognition,
critical thinking, and more analytical procedures are needed along with the accounting
curriculum. Therefore, academic institutions need to consider the implication of the current
wave of technology in skills and knowledge requirements.

64
5.2. Professional judgment

Auditors rely heavily on professional judgments and professional skepticism in the

decision-making process. The reasons behind the emphasis on professional judgment due
to the diversity and complexity of the business, different industries, and international
locations prevent auditors from making preset audit rules (Appelbaum et al., 2017, p. 8).
Therefore, auditors exercise their judgments in all stages of the audit process from audit
planning and performing an audit to reaching a conclusion (KPMG, 2018, p. 71). In this
part of the study, the authors examine the impact of BDA on the auditor's judgment as a
part of the research question. The empirical findings suggest that BDA has a positive
impact on auditor judgment. It assists auditors to make fact-based decisions and to focus
on areas that may include potential risks like anomalies and outliers.
5.2.1 Understanding the client Business environment and industry

Risk assessment is a significant part of audit planning, which includes the understanding of
the client's business environment and industry. Advanced technology such as BDA has
increased firms’ ability to collect, manage, and analyze more data effectively. As a result,
auditors have high potential to make a better judgment or enhance the way of making a
judgment and decisions like risk assessments (Brown-Liburd et al., 2015, p. 452). There is
a tendency among participants that BDA assists auditors in making a better judgment in
risk assessment based on facts. Looking at the full population to determine the outliers and
anomalies and focusing in these areas that may include high potential risk. Instead of
looking at a limited number of samples. For instance, BDA provides information that
prepared and posted the entries to GL in case any of these entries posted by CFO, and there
might be a chance for fraud (P3). This finding is consistent with Brown-Liburd et al.
(2015, p. 452) that using big data analytics can assist auditors in assessing client business
risk and other judgmental decisions like fraud risk, internal control, and going concerns.
Further, using BDA provides the auditor an opportunity to have a profound and robust
entity’s understanding and its environment, thus improving the quality of risk assessment
(IAASB, 2016, p. 7). Using BDA also generates additional insights through visualization
and dashboard, which need to be interpreted by the auditor, in other words requiring a
professional judgment (Salijeni et al., 2019, p. 113). Thereby, the authors conclude that
using BDA provides more insightful information about the whole population, which
facilitates auditors to exercise their judgment and make decisions based on fact and
objective means.
5.2.2 Quality of evidence

Auditors should test the quality of any evidence before exercising their judgment and make
decisions. Salijeni (2019, p. 17) mentioned that BDA could enhance the quality of
gathering evidence by improving the technical efficiency of the audit. All participants
indicated that audit evidence is subject to accuracy and completeness tests before
performing any further step in the audit process. For example, completeness tests by
making reconciliation between trial balance and balance sheet; accuracy tests by making
substantive tests. Further, another participant argued that evidence within big data
environment subjects to cross-check with trial balance and with other obtained sources data

65
that may be received from other engagement audits. This finding is consistent with Brown-
Liburd et al. (2015, p. 452) that the main concerns of auditors are to extract value from big
data and make sure that professional judgment and decisions rely on trustworthy and
relevant information.
In their efforts to express their opinion, auditors follow the standards which provide them
with general guidance and rules. However, the auditing process ends with judgment related
to many decisions; for instance, reaching a conclusion that gathered evidence is
appropriate and sufficient (Brown-Liburd et al., 2015, p. 452). The empirical findings
related to audit evidence provide different perspectives, which may reflect the fact that
most participants in this thesis are senior auditors. As a result, they do not have sufficient
experience to see the whole picture and provide more insight related to some issues like
professional judgment. Nonetheless, some participants mentioned that the decision of
gathered evidence is appropriate and sufficient to rely on some part of professional
judgment and specific cases and its complexities. Further, one participant indicates that
BDA itself is a tool that provides visualization and graphs that assist the auditor in making
sound decisions. These findings are consistent with Salijeni et al. (2019, p. 113) that
increased the use of BDA in audit as technique-based approaches may reduce the need for
professional judgments.
As well known, big data constitutes financial data and non-financial data like emails and
social media comments. The empirical results came very diverse that non-financial
information has limited or no effect on financial statement audit based on some
participants. However, two participants argued that non-financial information is used by
the auditor to support them in making some decisions like risk assessment by giving an
example that Covid-19 should be presented in most audit engagements in 2020. Moreover,
two participants mentioned, based on their observation that non-financial data support
auditors in making some decisions like fraud risk and going concerned.
On the other hand, two participants indicate that big data, in some cases like information
overload, has a negative effect on professional judgment. Auditors can also very easily get
lost in the details, especially if they do not have a good plan about what they are going to
perform and which data they should have. This finding is in line with Brown-Liburd et al.,
(2015, p. 453) using big data may concern the auditors in professional judgment and
decision making in relation to dealing with huge amounts of data from different sources.
Furthermore, auditors are not familiar with collecting and analyzing this kind of data
within an engagement.
5.2.3 Professional experience

With the growing use of BDA in financial statements audit, concerns have been raised
about the role of auditors (Salijeni et al., 2019, p. 113). Further, pushing more BDA in an
audit may increase the reliance on technique-based approaches hence lead to reduce the
need for auditor’s judgment (Salijeni et al., 2019, p. 113). However, using BDA can assist
auditors in conducting audits, but it could not automate some parts that need professional
judgment or subjective assessment (Richins et al., 2017, p. 73). In this regard, empirical
findings indicate that BDA has a positive impact on audit, and it improves the ability of the
auditor to make decisions based on facts and insightful information. As well as, BDA and

66
disruptive technology like AI and robotics cannot wholly replace the need for auditors due
to some areas in audit needing management assumptions and estimates. Thereby, BDA
provides support to auditors in making a professional judgment, but it cannot substitute
their judgment. Therefore, experience and knowledge will continue to play a vital role in
auditors' judgment.
5.3. Audit Efficiency

Empirical evidence received in this study suggests that efficiency is not achieved through
time and cost savings; in the first two years of using the BDA tools in the audit
methodology, more time is required. Interviewed participants indicate that time spent on
engagement activities increase due to the implementation of the new tools. This is due to
the fact that it takes a lot of time to identify the appropriate techniques and approaches to
collect relevant data. However, after two or three years, timesaving can be achieved when
all the tools are fully integrated into the methodology, and auditors understand how to use
the tools effectively. ICAEW (2016, p. 6) contrarily suggests that BDA is not likely to
lower time during the audit process, and auditors will have enough time to evaluate risky
areas to provide more fact-based evidence. Two participants shared this view. The
development of these tools was geared towards improving audit quality; therefore, it is
reducing the sample of the population to only areas that require more time for further
examination. However, it is anticipated that there could be a gain in timesaving in the
future. Three of the participants mention that in the future, time spent in engagement
activities when the engagement teams are more effective and familiar with the tools will
decrease. This argument is based on auditors' understanding of the client’s core activities
and risky operation, which will make it easier for auditors to evaluate these areas faster.
Others argue reducing the time may compromise quality, which is not the objective of the
new methodology.
Evidence and previous research show that implementing new technologies will require a
substantial amount of investments (IAASB 2016, p. 11). Obtained evidence is not
different, and audit firms are undertaking massive investment in the acquisition of new
technologies and training programs. This evidence contradicts the claim made by Alles
(2015, pp. 440- 441), that big data will lower audit costs and increase profitability. The
cost of an audit is also associated with numbers of personnel taking part-time spent on
engagement activities. As indicated above, more focus should be geared towards
improving audit quality. It was discovered that the number of personnel carrying out audit
engagement is not reduced. Therefore, audit firms are yet to realize significant cost-saving.
On the other hand, fees are affected by the overall cost of the audit. Since audit firms are
incurring more costs in terms of investment, audit clients are expected to pay a little bit
higher fee. All participants said that audit fees are not expected to be lower due to the
technology, because audit firms are investing in new technologies; clients are expecting
more insights from the audit reports. It will cost more or the same cost for more quality
and insights. It is possible in the future to realize lower audit costs and fees when BDA
tools are fully integrated into the audit methodology, and we see a positive experience and
learning curve of auditors in using the tools.

67
5.4. Audit Quality

Part of the research questions of this study is to explore how BDA in auditing affects audit
quality. Data collected regarding this question is based on the IAASB framework for audit
quality. However, the framework 2013. The framework outlined four key indicators to
measure audit quality. These are input, process, output, and interactions. The empirical
data collected during the interview suggest that there is an improvement in audit quality.
Through input by employing analytics tools and combination of various talents and
competences in the audit team; performing rigorous tests and procedures; producing fact-
based audit evidence and ease of communication with the audit committee and
management (Knechel et al., 2013, p. 390) These findings are supported by the claim
made by the FRC (2017) that data analytics through the input, process, output, and
interactions. These findings are discussed further in the subsequent subsections.
5.4.1 Input

Input reflects on the competence of the audit engagement team and the techniques
employed to conduct the audit. According to the data gathered during the data collection
process of this study shows that the engagement team comprises individuals with different
areas of expertise. According to the FRC (2017), BDA contribution to audit quality
depends on the engagement team to utilize the tools to understand the client entity.
Findings in this study are that using analytics tools contribute immensely to the auditor's
understanding of the client environment; “we use the tools to get more information and
understand the client industry” (P6). As the data suggest, using the tools in the planning
phase of the audit gives auditors more insights about the nature of the entity's IT
environment, business operation, and areas of high risk. Although the information received
reveals that the engagement team may not be effective in using the tools at the initial stage
of incorporating the technology in the audit. However, all participants indicate they get
more effective over time. Therefore, experience using the tools will improve the focus on
identifying key issues worth focusing on and subsequently improving the quality of the
evidence.
Furthermore, FRC (2017) claims that the engagement team knowledge of the client system
is key to achieving high quality. ICAEW (2016), argues that data analytics are developed
purposely to improve audit quality, although the achievement audit quality does not rely
solely on the tools but cannot be achieved without tools as well. This is confirmed in this
study. The information received in this study is that the tools are helpful in achieving high
audit quality, but the most important quality of the analysis is hugely dependent upon the
skills and competence of the engagement team members to perform the required test and
interpret the results. Accordingly, one participant said, “I am a firm believer that big data
analytics improves the audit quality but is based on the team's ability to carry out an
appropriate analysis [P2].” This proves that the tools and competence of the audit are
vital input to achieve audit quality. The tools complement the effort of the audit team.
5.4.2 Process

The second factor that influences audit quality is the processes such as risk assessment,
control test, and substantive test carried out during the engagement (Knechel et al., 2013,

68
p. 390). Many authors stress the ability of BDA to test full populations instead of the
traditional sampling method (Earley, 2015, p. 495; ACCA, 2016). All participants confirm
this ability of the BDA tools to perform analytical tests on the entire population of all the
transactions regardless of the size of the population. This allows auditors to detect
misstatements no matter the magnitude. Therefore, there is less detection risk. Testing the
entire population enables the auditor to identify abnormal transactions due to misstatement
or the nature of the transaction. These transactions sampled out for further investigation.

Figure 5: Data analytics – Impact on audit quality

Source: IAASB (2016, p. 7)

Risk assessment is a crucial part of the audit process. In traditional methodology, the
extent of the substantive procedures, sampling, and materiality planning depends on the
level of identified risk. FRC (2017) found that data analytics can enhance audit quality by
focusing on more risk assessment and producing more fact-based evidence. Participants in
this study echoed similar sentiments. Data analysis assists auditors in identifying risk areas
and transactions; then, auditors can focus on these areas studying the patterns displayed
through the visualization tools. “Data analytics is a big part of the risk assessment
procedures; we can disaggregate the population into subpopulation based on
homogeneity” (P2). Which means auditors can assess a single class of transactions at a
time. This is able to give a holistic view and understand the nature of all the transactions.
The description given is depicted through the diagram above (IAASB 2016, p. 7).
However, there is a contradiction when it comes to internal control, some participants said
they are not performing tests of control since they are testing the full population whereas
other participants indicate that the tools are very helpful in understanding the internal

69
control and IT environment of the client. There are two possible explanations of this
deviation from the other participant. First, these two participants mention that they are
SMEs as their major client. It could be that these clients are small and do not maintain a
complex internal control IT system. Therefore, using analytics tools can enable auditors to
wave the test of control. Secondly, the audit of large clients with complex internal control
and IT systems that is beyond their scope may require more experienced team members
with more skills and expertise to conduct the internal control test.
5.4.3 Output

The empirical results show that the output element of the audit quality is positively
affected by employing analytics tools in the audit methodology. According to three of the
participants, data analytics improves the quality of the audit report and enhances the
client's financial reporting system. Which implies that auditors are able to perform
thorough check and balances of the client records that will help the client improve the
reporting framework; at the same time enable the auditor to gather enough insightful
evidence to guide the formulation of the audit opinion. Another participant highlights how
they (auditors) leverage the tools to get a holistic view of the client system; in return, they
generate solid and fact-based evidence. These findings are in accordance with Knechel et
al. (2013, p. 390) that the quality of the financial reporting framework of the client and the
reliability of the auditor report influence audit quality. Hence the reliability of the audit
report depends on how the clients can trust the information contained in the report for
decision making.
Furthermore, the findings include using non-financial data to perform further investigation.
However, the use of non-financial data is not fully integrated into the process. One of the
participants indicated non-financial data use mainly used in fraud investigations. This
implies that if there is any suspicion of irregular activities, auditors can make use of data
analytics to further investigate the activities. In light of the above, in an interview with the
World Street Journal, Lara Abrash, a managing partner at Deloitte (2017), argued that audit
of the future would evolve to include more than a financial statement audit, the participants
of this study share a similar view. Most of the participants foresee the full utilization of
non-financial information to help clients understand their business better and gain more
insights into key performance areas that require improvement. Therefore, it is anticipated
that in the future, we are more likely to see an unparalleled improvement of clients'
reliance on the audit report. Similarly, the information contained in the audit report
increased in value (IAASB, 2013, pp. 46- 47). Participants also highlighted that errors are
minimized if not eliminated at all.
5.4.4 Interaction

The relationship between the various stakeholders of the audit process is an important
element of achieving audit quality (Francis, 2011, p. 134). Depending on the nature of the
interactions, quality can be impaired. The empirical findings indicate data analytics
produce a positive impact on the interaction among the key stakeholders such as
management, audit committee, and those charged with governance with the engagement
team. The tools facilitate the timely transmission of information between these
stakeholders. Using the tools provide auditors the opportunity to effectively communicate
70
with management and the audit committee about the discoveries made during the audit
process through the use of visualization tools to display such information in graphs and
charts. This builds confidence in the audit report and the approach the auditors
implemented to conduct the audit. One of the participants mentioned using the
visualization tools allow management to understand how the audit is conducted and can
request more information or insights about certain areas of activities in the business.
Similarly, Salijeni (2019) found that using the tools can ease communication, improve the
reliability of the audit report for decision making.
On the other hand, it was also discovered that tools not only enhanced the interactions
between the audit team and other stakeholders but improved communication among the
engagement team members. Engagement team members are able to share vital information,
and all participants can access the relevant information and view the progress of the audit
process. As Francis (2011, p. 134) suggests, interactions are influential factors of audit
quality, so is the interaction and coordination of the engagement team. The empirical data
indicate the tools coherently, and more focus is possible among team members. Which
means all the activities required to complete the audit is clearly available for the team.
In summary, the BDA have a positive effect on the factors of audit quality. The evidence
suggests that auditors are able to use analytics and visualization tools to gather sufficient
and fact-based audit evidence. The tools are very effective in identifying risk. Besides
auditors can test an entire population, making it easier to identify discrepancies contain in
financial statements. Auditors can assess the completeness and accuracy of data before
performing the analysis. Furthermore, Visualization tools enhance auditors understanding
of the client business processes and environment. Auditors can have a very robust and
holistic view of client business and industry. In turn, this help improves the professional
skepticism and professional judgments. Additionally, the results indicate that audit quality
will increase more, because auditors are getting better at using the tools. Audit firms are
continuously providing training for their auditors. These training will increase time
efficiency, which means auditors can focus more on improving the auditor quality.
Moreover, the visualization capabilities of the tools improve the interactions between
auditors and key stakeholders such as management and those charged with governance.
Overall, the audit quality is improved, and clients can benefit more the tools in the form
getting more insights from audit reports.

71
6. CONCLUSION
This chapter contained the final words about the findings, given a summary of the key take
away from the paper and answers to the research questions. Included in this chapter is
how the research contribute to the existing literature, practical and societal implication of
the study. Limitations of the study and area future research are delineated in this section.
The last two sections of the chapter are the quality criteria, and the ethical consideration
of the study is discussed.

6.1. General Conclusion

The aim of this research is to explore how the use of BDA in the audit process may affect
professional judgment, audit efficiency, and perceived audit quality. Secondly, this thesis
aims to look at how the use of BDA in the audit methodology will affect the minimum skill
and competence requirement for auditors. It was found that the tools are used at the two
most crucial phases of the audit process, planning, and execution phase. Auditors are able
to perform an extensive risk assessment, identifying higher risk areas to focus on, using a
full population rather than a traditional sampling method, lowering the possibility of
detection risk. The result of this study shows that the skills and competence requirement
for auditors is getting more complex and diverse. It encompasses many areas of expertise
in addition to economic knowledge such as accounting, business law, IFRS, and auditing;
IT and data science are getting more prominent in the audit field. The findings particularly
point out that auditors need to advance their IT skills and competence with the main focus
on excel. Excel is found to be the foundation of all the analytics tools being used in the
audit process. Auditors are also required to upgrade their skills accordingly. These findings
suggest that academic institutions should also redesign their curriculum to make room for
the needed competence and skills in the job market.
Furthermore, professional judgment is highly subjective and depends hugely on
professional experience and knowledge, understanding of client business environment, and
quality evidence obtained during the audit process. The findings show BDA tools
enhanced professional judgment in terms of risk assessment and understanding of the
client business environment. Further, the decision that the gathered evidence sufficient and
appropriate is not clear cut due to the fact that most of the participants are senior auditors.
They may not have a holistic overview of the whole picture in decision making. However,
it was discovered the tools are not likely to replace professional judgment in the near
future.
Moreover, the findings show that efficiency has not been achieved yet with the tools.
Instead, implementing the tools for the first and second year, time spent on engagement
activities will increase, thereby increasing audit costs. This can be explained by the fact
that implementing a new methodology. It will require more time for the user to learn and
adapt to the new methods. Auditors need to learn how to use the tools when they are
initially introduced in the audit process. Moreover, efficiency has not been realized with
the tools because the main objective of using the tools in auditing is to improve audit
quality. It is argued that quality should not be compromised to save cost or time. However,

72
it is anticipated that efficiency can be achieved in the future when auditors are more
comfortable, and the experience of using the tools is higher. The empirical findings suggest
that incorporating BDA tools into the audit methodology has an impact on the perceived
audit quality. This impact will continue to increase as auditors know how to use the tools
effectively and efficiently. Auditors use the tools to create more value for audit clients and
users of financial statements. The visualization tools improve communication between key
stakeholders such as management, audit committee, and the audit team; also,
communication among team members is enhanced through BDA tools.
6.2. Contribution & Implication of research

6.2.1 Theoretical contribution

Data analytics in audit is a trending topic and generates a lot of interest from auditors,
clients, regulators as well as users of financial statements. In this era of digitalization and
fast-paced innovation, this research contributes to the scantily available literature,
complementing some of the previous research conducted on this topic by providing more
insights into how BDA can affect perceived audit quality from the practitioners
perspective. The empirical findings provide an understanding of why efficiency has not
been achieved using BDA tools. Prior to this study, it was belief implementing data
analytics will automatically improve efficiency and perceived quality. However,
surprisingly it was discovered efficiency had not been realized. Considering that this is the
first research on this topic conducted in Sweden, the paper contributes to the ongoing
debate on the topic in the Swedish context by providing evidence on the implication BDA
in the audit methodology. Further, the paper uncovers areas for future research, thereby
providing reference points worth considering in the development of the topic in
academics.
6.2.2 Practical implication

The findings of this thesis are of practical use for various stakeholders. Audit firms can
rely on the findings of this study to benchmark resource allocation to engagement teams.
The study provides more insight into how the data analytics affect the audit process and
what skills and competence combination is required to accomplish the engagement
activities. Furthermore, the research identifies the key competencies and skills required to
perform an audit in a big data environment. This will help audit firms in the recruitment
process for new auditors. It will help to identify and tap the appropriate combination of
talents and skills. On the other hand, the finding can be useful to audit clients and users of
financial statements as well. It was found that BDA improves audit quality and creates
more insights for clients. This avail audit clients how BDA is affecting the audit process,
what value can be generated, and how they can benefit from an audit process using BDA
tools. The result will also be useful in helping clients to collaborate with auditors to gain
more information about their business process for better strategic planning.
6.2.3 Societal implication

Socially, this study is relevant for academic institutions and regulatory bodies. The
empirical findings provide an understanding of the need to redesign the academic

73
curriculum to prepare future auditors. Therefore, academic institutions alongside audit
firms and regulatory bodies can use the findings in this paper to determine the required
adjustment in the current academic curriculum. This research can also be used by
regulators to understand the impact of BDA on audit quality. Thus, it can be helpful to
understand how to regulate the implementation of the new audit methodology and monitor
auditor's qualifications and evaluate the experience. Individual auditors and those aspiring
to work in the audit industry can use the findings of this paper to develop their skills and
monitor the technology advancement affecting the profession and the industry at large.
6.3. Limitations

Firstly, the period in which the study is conducted is during the busy season for auditors. It
was difficult to obtain participants to participate in the research. This led to the fact that the
authors could not bring in more participants. On the other hand, time factors on the side of
the researchers limit their ability to conduct more interviews, resulting in a less diverse
perspective in the empirical evidence. Secondly, the mix of the interviewed participant
does not include participants from top management. It is thought that participants from top
management would have provided more valuable information, particularly on the aspect of
professional judgment. However, the authors invested a lot of effort to reach partners and
more senior managers for interviews, but all efforts went in vain. The above factors
contributed to the limited sample size and lack of more senior management representatives
in this study.
Moreover, the initial objective of the authors was to study the topic from diverse
perspectives; the aim was to interview auditors from different countries and regions around
the world. The authors sent emails to big four firms to many countries around the world
(see appendix 4), most of them did not respond, and the few that responded declined to
take part in the research. All the interviewed participants are based on Swedish big four
audit firms. Therefore, this research is limited by the fact that all the obtained evidence
only focused on the Swedish market. Therefore, the findings do not represent other
companies or firms, particularly not those that operate outside Sweden. Furthermore, based
on the sample size and composition of the interview auditors, it is impossible to draw any
kind of generalizable conclusions from it, more research is needed on the topic.
6.4. Avenues for future research

First and foremost, as mentioned in the previous section, persons interviewed in this study
do not include higher-level staff in the audit profession. Hence the evidence collected in
regard to professional judgment is not sufficient. In the future, research that includes top
officials in the audit field should be conducted to better understand how BDA affect
Professional judgment and decision making in auditing. Secondly, the empirical evidence
suggests that BDA has an impact on audit quality. This opinion is based on the perspective
of audit practitioners. It can be argued that clients, regulators, and other users of financial
statements may have a similar or different opinion. Therefore, the authors of this research
thought it would be interesting for a similar study to be conducted with a focus on gazing
at the perspective of audit clients, regulators, and other users of the financial statements.
The findings of such studies can be compared with the finding of this study. It was found
IT-related skills are becoming more salient in the audit field, warranting a call to redesign
74
the academic curriculum. Research is needed to look at how best academic institutions and
professional regulatory bodies can collaborate to identify the required adjustment to the
current curriculum that fits with the skills required in the field.
Moreover, for the fact that this study was focused only on big four audit firms in Sweden,
this study can be repeated with firms outside Sweden with different regulatory
requirements. Such a study will offer an overview of how regulatory requirements affect
the use of BDA in auditing and the impact it will have on audit quality. Finally,
quantitative, or longitudinal research can be conducted to investigate the impact of BDA
tools on audit efficiency. This will provide more insights into the time spent and the cost of
audits from both client and practitioners’ perspectives.
6.5. Quality Criteria

In order to assess qualitative research work, there is no consensus among researchers in

which quality criteria should be used (Bryman & Bell, 2011, p. 43; Leung, 2015, p. 325).
However, reliability and validity are most common in evaluating qualitative research work
in business and management research (Bryman & Bell, 2011, p. 41). Reliability is
concerned either if the outcome of the study is repeatable or whether the measures
designed for a concept in management and business are consistent (Bryman & Bell, 2011,
p. 41). Therefore, research seems to be reliable if the researcher can repeat former research
designs and reach the same results. Further, reliability should be considered in a distinction
between internal and external reliability. Internal reliability refers to consistency should be
ensured during research. External reliability refers to the consistent finding that can be
gained by using the same data collection method and analytical procedures either by the
same researcher on another occasion or by different researchers (Saunders et al., 2016, p.
202). Despite all that, reliability is essential and critical criteria to ensure good research
quality, but it is not sufficient, and quality of research also requires validity (Saunders et
al., 2016, p. 203).
Validity refers to measures used are appropriate, the analysis of results is accurate, and the
findings are generalizable (Saunders et al., 2016, p. 202). Further, validity considers the
most significant criterion in research, and it concerns ‘’the integrity of the conclusions that
are generated from a piece of research’’ (Bryman & Bell, 2011, p. 42). Moreover, validity
and reliability are related to each other for instance, the measure of concept should be
stable, which means reliable in order to be a valid measure (Bryman & Bell, 2011, p. 42).
Validity is discussed in terms of internal validity and external validity. The internal
validity is related to causality, which refers to a causal relationship between two variables
is presented accurately by research (Saunders et al., 2016, p. 203). External validity or
generalizability questions ‘’whether the results of a study can be generalized beyond the
specific research context’’ (Bryman & Bell, 2011, p. 43). Saunders et al. (2016, p. 205)
argued that concerns had been raised for using external validity in qualitative research due
to generalizability that cannot be achieved by conducting a study on small samples.
However, other forms of generalizability are indicated by qualitative researchers in
demonstrating the value and quality of a qualitative study. As well as, using a qualitative
study is most likely to explore and provide insight in order to develop theory instead of
providing generalizability (Saunders et al., 2016, p. 398).

75
On the other hand, some qualitative researchers propose other criteria to be used in
evaluating the qualitative study. Lincoln and Guba (1985, cited by Bryman & Bell, 2011,
p. 43) suggested alternative or similar criteria in assessing qualitative research. The
alternative relies upon trustworthiness, which composite of: Credibility, transferability,
Dependability, and Confirmability.
Dependability is a parallel criterion to reliability in quantitative research (Saunders et al.,
2016, p. 206). It means that researchers should ensure that the record of the whole research
process (problem formulation, fieldwork notes, interview transcripts, selection of research
participants, data analysis decisions, and so on) should be kept in an accessible way.
Dependability can be achieved by following an auditing approach by assigning peers as an
auditor. Therefore, peers are responsible for ensuring that proper procedures are followed
during the research (Bryman & Bell, 2011, p. 398). However, Concerns raised related to
reliability dependability of semi-structured interviews due to issues of bias. This bias
should be considered in three types: interviewer bias, interviewee bias, and participant
bias. Another concern about semi-structured interviews is the cultural difference between
interviewee and interviewer (Saunders et al., 2016, p. 397). In this research, the authors
followed all recommended measures to achieve dependable conclusion through keeping
proper documentation of all relevant information gathered during the course of the research
such as the interview guide, recordings of all the interviews, transcriptions of the
interviews as well as participants information. These records were kept in a shared file and
individually by the authors to ensure safekeeping and availability.
Credibility is parallel to internal validity in quantitative research (Saunders et al., 2016, p.
206). It refers to the ability of researchers to gain access to the experience and knowledge
of a participant and their ability also to explore the intended meaning from the language
used by the participant. The credibility of collected data relies upon the scope of exploring
meaning during an in-depth or semi-structured interview (Saunders et al., 2016, p. 398).
Further, a high level of credibility can be achieved in a semi-structured or in-depth
interview by using clear questions, proper meaning, and gaining responses from different
perspectives. The credibility can also be achieved by using qualitative interviews, which
may assist researchers in building trust and relationships, gather sufficient data, and verify
these data by asking participant permission (Saunders et al., 2016, p. 400- 401). However,
the credibility of the outcome of qualitative research may be impaired due to different
types of bias and cultural differences (Saunders et al., 2016, p. 398). The credibility of this
research was ensured through the semi-structured interviews. Participants had the chance
to elaborate and provide a detailed explanation of issues that were relevant during the
interview process. The authors also use the follow-up question to clarify the pertinent issue
that was deemed crucial in answering the research questions. The authors as well provide
the data collected during the interview with the participants for verification to make sure
participants' ideas are adequately represented in the research. A draft of the empirical
results was sent to each of the participants. The authors received five feedbacks; three of
which required minor adjustments while others express satisfaction. Two participants did
not respond; therefore, the authors conclude that their silence means acceptance.
Transferability is an equal criterion for external validity in a quantitative study. It defines
that a full description of studies like research questions, context, design and findings, and

76
interpretations should be provided by researchers to readers in order to make a judgment to
transfer study to another setting (Saunders et al., 2016, p. 206). Confirmability refers to
researchers should act on good faith in order to ensure the objectivity of research.
Considering that full objectivity in business research is kind of impossible. However, it
should be obvious that research finding has not been extracted from researcher
theoretical preconception or personal value (Bryman & Bell, 2011, p. 398). This research
is based on an exploratory study, which means if the context and situation on which the
research is being conducted change, the findings may differ as well (Yin, 2003, p. 38).
Base on this argument, the authors do not intend to generalize or transfer the findings of
this research to other settings or contexts. The paper is aimed at contributing to the existing
literature on the topic.
6.6. Ethical consideration

Ethical behavior should be presented by all parties in research. The main purpose of ethical
consideration is to ensure that research activities do not expose anyone to harm or adverse
consequences like pain, embarrassment, and loss of privacy. Therefore, designing any
research should consider the protection right of all participants or subjects. Further, the
personal integrity of researchers must be present in ethical research (Cooper & Schindler,
2014, p. 28). The researcher protects participants in research by following and explaining
the guideline, study benefit; participants' rights and protections; and gaining informed
consent (Cooper & Schindler, 2014, p. 29). The researcher should inform the participants
of the benefit of study and should also be very careful to avoid any overestimate or
underestimate of the benefits. At the beginning of each interview, the interviewer should
introduce him or herself to participants and provide the name of the research. They should
also provide very clear information about the objective and purpose of the study. This
information help participants to know their speaker and encourage them to provide truthful
answers. In some cases, the interviewers conceal the purpose and benefit of the study to
avoid introducing bias; hence the deception problem will arise (Cooper & Schindler, 2014,
pp. 29-30).
Deception takes place when part of the truth is presented by participants or when there is a
full compromise about the truth. There are different perspectives on deception. On the one
hand, deception should never take place. On the other hand, deception can occur in two
cases: prior to any survey or experiment, participants should be protected from biasing;
confidentiality of the third party such sponsor should be protected. However, using the
deception should not be an endeavor to enhance the response rate. Further, a balance
should be considered between the gained benefit from deception and the risk against
participants (Cooper & Schindler, 2014, pp. 30-31). Finally, informed consent must be
taken from participants prior to their participation in the research. In order to secure the
informed consent of participants, the procedures of the interview or any research design
should be fully disclosed before taking part in the study. Moreover, informed consent
should be signed in some cases, like dealing with children and medical research. However,
the oral consent of participants in business research is considered sufficient (Cooper &
Schindler, 2014, p. 31). According to the AoM code of ethical conduct, researchers have a
duty to protect and maintain the dignity, freedom, well-being, and privacy of participants.

77
The participants also have a right to refuse or terminate their participation at any time
(Bryman & Bell, 2011, p. 133).
Once the research is complete, debriefing participants should occur in case of their
deception intentionally or accidentally (Cooper & Schindler, 2014, p. 28). The debriefing
aims to provide information about the nature of the study to participants, its result and to
provide assistance if the participants are exposed to any adverse consequences from their
participation in the study (Saunders et al., 2016, p. 257). Lastly, the right to privacy is a
right for all individuals to have privacy, and researchers should respect this right. Further,
the right to privacy should be guaranteed to validate the study and to provide protection to
participants as well. The salience aspect of the right to privacy is confidentiality, and its
protection is essential. Researchers can protect the confidentiality of participants in many
ways, such as signing non-disclosure documents and restricting access to information
about participants like name, address, phone number, and other identifying features by
researchers who have not signed the non-disclosure. Confidentiality is just a part of the
right to privacy; it also involves the right of participants to refuse to take part in an
interview or refuse to answer any question in an interview (Cooper & Schindler, 2014, p.
28).
Cognizant of the above concepts and ethical considerations requirement in business
research, the authors put some of the recommendations that are relevant for this study.
Such consideration includes but not limited to following informed consent, invasion of
privacy, confidentiality, anonymity, honesty, and transparency. Prior to scheduling the
interview with participants, the authors sent a document to all participants to obtain
consent to record the interview. They were informed that the information provided during
the interview will only be available to the authors, and the supervisor of this thesis upon
request. The information will be used purposely to answer the research question (See
Appendix 1 & 3). Thus, the information will not be disclosed to a third party. At beginning
of every interview, the authors reiterate the above information to all the participants to
ensure informed consent is obtained. To avoid an invasion of privacy, all participants were
asked if it is okay for the interview to be conducted via video conferencing.
Furthermore, information in this thesis does not contain any personal data that could be
linked to individual participants or the organization they work for. For honesty and
transparency, the authors sent the complete empirical results to the participant to ensure the
information is appropriately utilized in this research. Feedback received from participants
were accordingly adjusted in the final draft of the thesis. Therefore, this research was
ethically conducted in accordance with the USBE thesis manual requirement.

78
7. Reference:
Abbott, L. J., Daugherty, B., Parker, S., & Peters, G. F. (2016). Internal Audit Quality and
Financial Reporting Quality: The Joint Importance of Independence and Competence.
Journal of Accounting Research, 54(1), 3–40
Abrash, L. (2017). Gaining Insights from Innovations in a Financial Statement Audit. CFO
Journal. (CFO insights & analysis written and compiled by Deloitte.)
ACCA. (2016, July 01). data analytics, and the auditor. [Online].
https://www.accaglobal.com/in/en/student/exam-support-resources/professional-exams-
study-resources/p7/technical-articles/data-analytics.html [Retrieved 2020-03-28].
Adrian, S., & Viorica, M. (2015). Using the professional judgment within the financial
auditor's profession. Global Economic Observer, 3(1), 141-143
Alles, G. M, and Gray, G. L. (2016). incorporating big data in audits: identifying inhibitors
and a research agenda to address those inhibitors. International Journal of Accounting
Information Systems, Vol. 22, pp. 44-59.
Alles, G. M. (2015). Drivers of the use and facilitators and obstacles of the evolution of big
data by the audit profession. Accounting Horizons, Vol. 29, No. 2, pp. 439-449.
Alles, M., & Gray, G. (2018). The pros and cons of using big data in auditing: a synthesis
of the literature and a research agenda. Rutgers. pp.1-37.
Anca, O., & Andra, M. (3/20). Professional judgement. The key to a successful audit. Sea:
Practical Application of Science, Ii (5 (3/2014), 217-222.
Appelbaum, D. A. (2016). securing big data provenance for auditors: the big data
provenance black box as reliable evidence, Journal of Emerging Technologies in
Accounting, Vol. 13, No. 1, pp. 17-36.
Appelbaum, D. A., Kogan, A., and Vasarhelyi, M. A. (2017). Big data and analytics in the
modern audit engagement research needs, auditing: A Journal of Practice and Theory, Vol.
36, No. 4, pp. 1-27. 248
Appelbaum, D. A., Kogan, A., and Vasarhelyi, M. A. (2018). analytical procedures in
external auditing: A comprehensive literature survey and framework for external audit
analytics, Journal of Accounting Literature, Vol. 40, June, pp. 83-101.
Ashton, R., & Ashton, A. (2007). Judgment and decision-making research in accounting
and auditing (Cambridge series on judgment and decision making). Cambridge: Cambridge
University Press. (2007).
BDO United States (2016). Data Analytics & Enhancing Audit Quality. (online) [Available
via] https://www.bdo.com/insights/assurance/corporate-governance/data-analytics-
enhancing-audit-quality [Retrieved 2020/04/17]

79
Bell, S. K., & Cecez-Kecmanovic, D. (2014). A Hermeneutic Approach for Conducting
Literature Reviews and Literature Searches. AIS Journals. Vol. 34(12), 257-286

Bierstaker, J. L., Burnaby, P., & Thibodeau, J. (2001). The impact of information
technology on the audit process: an assessment of state of the art and implications for the
future. Managerial Auditing Journal, 16(3), 159–164.

Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative
Research in Psychology. Vol. 3 (2), 77-101.

Brown-Liburd, H., & Vasarhelyi, M. A. (2015). Big Data and Audit Evidence. Journal of
Emerging Technologies in Accounting, 12(1), 1–16.
Brown-Liburd, H., Issa, H., & Lombardi, D. (2015). Behavioral implications of big data's
impact on audit judgment and decision making and future research directions., 29(2), 451–
468.
Bryman, A. & Bell, E. (2011). Business Research Methods. 3rd edition. Oxford, UK:
University Press Inc.
BSi. (2011). Guidelines for Auditing Management Systems (ISO 19011:2011). [PDF].
Available via: http://qic-eg.com/wp-content/uploads/2015/08 /BS-EN-ISO-19011-
2011.pdf. [Retrieved 2020-03-10].
Cannon, D. L. (2017). Audit Process: Certified Information Systems Auditor Study Guide.
4th edition. New Jersey: John Wiley & Sons, Inc.
Cao. M., Chychyla, R., and Stewart, T. (2015). big data analytics in financial statement
audits, Accounting Horizons, Vol. 29, No. 2, pp. 423-429.
Cooper, D. R., & Schindler, P. S. (2014). Business Research Methods. Twelfth Edition.
McGraw-Hill. New York.

Cronin, P., Ryan, F., & Coughlan, M. (2008). Undertaking a literature review: a step-by-
step approach. British Journal of Nursing. Vol. 17(1), 38- 43

Danescu, T., & Chira, A. (2014). Professional judgment and reticence to apply sampling
techniques. Procedia Economics and Finance, 15, 1253-1258.
Deegan, C. M. (2019). Legitimacy theory. Accounting, Auditing & Accountability
Journal, 32(8), 2307–2329.
Drira, M. (2013). Toward a General Theory of Client Acceptance and Continuance
Decisions. Journal of Comparative International Management, Vol. 16, No. 1, 37-52
Duh, R., Lee, W., & Hua, C. (2007). Non-audit service and auditor independence: an
examination of the Procomp effect. Springer Science+Business Media, LLC.
Earley, C. E. (2015). data analytics in auditing: opportunities and challenges. Business
Horizons, Vol. 58, No. 1, pp. 493-500

80
Eilifsen, A., Messier, W. F., Glover, S. M. & Prawitt, D. F. (2014). Auditing and
Assurance Services. 3rd edition. New York: McGraw-Hill
Ernst & Young. (2018). Advancing Analytics and Automation within Internal Audit.
[PDF]. Available via: https://www.ey.com/Publication/vwLUAssets/ EY-advancing-
analytics-and-automation-within-internal-audit/$FILE/EY-advancing-analytics-and-
automation-within-internal-audit.pdf. [Retrieved 2020-03-30]
Financial Reporting Council (2017). Audit Quality Thematic Review: The Use of Data
Analytics in the Audit of Financial Statements. (online) Available via:
https://library.croneri.co.uk/miscaqtr0117 [Retrieved 2020/04/17]
Forbes Insights. (2019, March 4). How Advanced Technologies May Improve Audit
Quality. [Online]. Available via: https://www.forbes.com/sites/insights-kpmg
/2019/03/04/how-advanced-technologies-may-improve-audit-quality/#. [Retrieved 2020-
03-25].
Francis, J. R. (2011). A Framework for Understanding and Researching Audit Quality.
AUDITING: A Journal of Practice & Theory, 30(2), 125–152.
FRS. (2017). Audit Quality Thematic Review: The Use of Data Analytics in the Audit of
Financial Statements. [Available via] https://www.frc.org.uk/getattachment/4fd19a18-
1beb-4959-8737-ae2dca80af67/AQTR_Audit-Data-Analytics-Jan-2017.pdf
Fukukawa, H., Mock, T. J., & Wright, A. (2011). Client Risk Factors and Audit Resource
Allocation. A Journal of Accounting, Finance, and Business Studies, Vol. 47, No. 1, 85-
108
Gandomi, A., & Haider, M. (2015). Beyond the hype: Big data concepts, methods, and
analytics. International Journal of Information Management, 35(2), 137–144.
Gepp, A., Linnenluecke, Martina K., O‟ Neill, Terrence J., and Smith, T. (2018). Big data
techniques in auditing research and practice: Current trends and future opportunities,
Journal of Accounting Literature, 40, pp.102–115.

Gill, P., Stewart, K., E. Treasure, E., & Chadwick, B. (2008). Methods of data collection in
qualitative research: interviews and focus groups. British Dental Journal. Vol. 204(6),
291-295

Hager, P. & Gonczi, A. (1996). What is Competence? [Online]. Available via:

https://www.researchgate.net/publication/232061493_What_is_competence. [Retrieved
2020-03-10].

Hox, J. J., & Boeije, H. R. (2005). Data Collection, Primary Vs Secondary. Encyclopedia
of social measurement. Vol. 1, 593- 599

IAASB [International Auditing and Assurance Standards Board]. (2013). A Framework for
Audit Quality Key Elements that Create an Environment for Audit Quality.

81
[PDF]..https://www.ifac.org/system/files/uploads/IAASB/Framework-for-Audit-Quality-
Outline.pdf [Retrieved 2020-03-24].
IAASB [International Auditing and Assurance Standards Board]. (2013). A Framework for
Audit Quality. [PDF]. Available via: https://www.ifac.org/system/files/publications
/files/A%20Framework%20for%20Audit%20Quality.pdf. [Retrieved 2020-03-24].
IAASB. (2016). Exploring the Growing Use of Technology in the Audit, with a Focus on
Data Analytics. [Available via] https://www.ifac.org/publications-resources/exploring-
growing-usetechnology-Audit-focus-data-analytics
ICAEW. (2018b). Understanding the Impact of Technology in Audit and Finance. [PDF].
Available via: https://www.icaew.com/-/media/corporate/files/middle-east-
hub/understanding-the-impact-of-technology-in-audit-and-finance.ashx. [Retrieved 2020-
03-30].
ICAS & FRC. (2016). AUDITOR SKILLS IN A CHANGING BUSINESS WORLD
[Available via] https://www.icas.com/__data/assets/pdf_file/0010/259516/Auditor-skills-
ICAS-FRC-20.9.16-Final.pdf
IFAC. (n.d.) Audit Quality and International Standards on Auditing. [PDF]. Available via:
http://www.ifac.org/system/files/downloads/Audit_Quality_Fact_Sheet.pdf. [Retrieved
2020-03-25].
Institute of Chartered Accountants. (2006). Audit Quality Fundamental - Audit Purpose.
[PDF]. [Availablevia] https://www.icaew.com/-/media/corporate/files/technical/audit-and-
assurance/audit-quality/audit-quality-forum/meeting-notes-2006/july-2006-audit-
purpose.ashx?la=en. [Retrieved 2020-03-25].
International Accounting, Auditing & Ethics, (2016). Data analytics for external auditors:
international auditing perspectives. International Accounting, Auditing & Ethics, pp. 1- 26

Jacob, S. A. (2012). Writing Interview Protocols and Conducting Interviews: Tips for
Students New to the Field of Qualitative Research. The Qualitative Report. Vol 17(6), 1-
10

Julia Kokina and Thomas H. Davenport (2017) The Emergence of Artificial Intelligence:
How Automation is Changing Auditing. Journal of Emerging Technologies in Accounting:
Spring 2017, Vol. 14, No. 1, pp. 115-122.
Knechel, W. R., Krishnan, G. V., Pevzner, M., Shefchik, L. B., & Velury, U. K. (2013).
Audit Quality: Insights from the Academic Literature. AUDITING: A Journal of Practice
& Theory, 32 (Supplement 1), 385–421.
KPMG. (2013). Performance Auditing. [PDF]. [Available via]
https://assets.kpmg/content/dam/kpmg/pdf/2016/05/Performance-Auditing.pdf. [Retrieved

82
Leung, L. (2015). Validity, reliability, and generalizability in qualitative research.
(Research and Audit) (Report). Journal of Family Medicine and Primary Care, 4(3), 324–
327.
Leung, p., Coram, p., Cooper, B. J., Richardson, P. (2015). Modern Auditing & Assurance
Services. 6th Edition. Australia: John Wiley & Sons
Long, R., White, M., Friedman, W. & Brazeal, D. (2000). The ‘qualitative' versus
‘quantitative' research debate: A question of metaphorical assumptions? Journal of
international value-based management, 13, 189-197.

Malterud, K., Siersma, V., D., & Guassora1, A. D. (2016). Sample Size in Qualitative
Interview Studies: Guided by Information Power. Qualitative Health Research. Vol.
26(13), 1753- 1760

Marshall, B., Cardon, P., Poddar A., & Fontenot, R. (2013). Does Sample Size Matter in
Qualitative Research? A Review of Qualitative Interviews in IS Research. Journal of
Computer Information Systems. Vol. 54(1), 11- 22

Messier, W. F., Glover, S. M., & Prawitt, D. F. (2017). Auditing & Assurance Services:
Systematic Approach. 10th Edition. New York: McGraw-Hill Education.
Miklos A. Vasarhelyi, Alexander Kogan, and Brad M. Tuttle (2015) Big Data in
Accounting: An Overview. Accounting Horizons: June 2015, Vol. 29, No. 2, pp. 381-396.
Morgan, D. L. (2007). Paradigms Lost and Pragmatism Regained: Methodological
Implications of Combining Qualitative and Quantitative Methods. Journal of Mixed
Methods Research, 1(1), 48–76.
Neuman, W. L. (2014). Social Research Methods: Qualitative and Quantitative
Approaches. 7th Edition. Edinburgh: Pearson New International Edition.
No, W. G., Lee, K., Huang, F., & Li, Q. (2019). Multidimensional Audit Data Selection
(MADS): A Framework for Using Data Analytics in the Audit Data Selection
Process. Accounting Horizons, 33(3), 127–140.

Nyström, M. and Dahlberg, K. (2001). Pre‐understanding and openness–a relationship

without hope? Scandinavian Journal of Caring Sciences, 15(4), pp.339-346.

O’dwyer, B., Owen, D., & Unerman, J. (2011). Seeking legitimacy for new assurance
forms: The case of assurance on sustainability reporting. Accounting, Organizations, and
Society, 36(1), 31–52.
Oleksandr, S. (2016). Judgements in the implementation of tasks of audit of financial
information. European Cooperation, 5(12), 35-44
Phil D Wedemeyer. (2010). A discussion of auditor judgment as the critical component in
audit quality – A practitioner's perspective. International Journal of Disclosure and
Governance, 7(4), 320–333.

83
Power, M. K. (2003). Auditing and the production of legitimacy. Accounting,
Organizations, and Society, 28(4), 379–394.
PwC (2013 January). PricewaterhouseCoopers 2013. Understanding a Financial Statement
Audit. [Online]. Available via: https://www.pwc.com/gx/en/audit-
services/publications/assets/pwc-understanding-financial-statement-audit.pdf [Retrieved
2020-03-09].
Richins, G., Stapleton, A., Stratopoulos, T. C., & Wong, C. (2017). Big Data analytics:
Opportunity or threat for the accounting profession? Journal of Information Systems, vol.
31, no. 3, pp.63-79.
Robson, K., Humphrey, C., Khalifa, R., & Jones, J. (2007). Transforming audit
technologies: Business risk audit methodologies and the audit field. Accounting,
Organizations, and Society, 32(4), 409–438.

Rowley, J. (2002). Using Case Studies in Research. Management Research News. Vol.
25(1), 16- 27

Rowley, J., & Slack, F. (2004). Conducting A literature review. Management Research
News. Vol. 27(6), 31- 39

Salijeni, G. (2019). Big Data Analytics and the Social Relevance of Auditing: An
Exploratory Study (Doctoral dissertation, The University of Manchester (United
Kingdom)).
Salijeni, G., Samsonova-Taddei, A., & Turley, S. (2019). Big Data and changes in audit
technology: contemplating a research agenda. Accounting and Business Research, vol. 49,
no. 1, pp.95-119.

Saunders M., Lewis P. & Thornhill, A. (2009). Research Methods for Business Students.
5th edition. Essex: Pearson Education Limited. E-book.

Saunders M., Lewis P. & Thornhill, A. (2016). Research Methods for Business Students.
7th edition. Edinburgh Gate: Pearson Education Limited. E-book.

Shimamoto, D. C. (2010). Automation is Key to Improving the Efficiency and

Effectiveness of Audits. [PDF]. Available via: https://www.confirmation.com/media
/1309/white-paper_automation-is-the-key-rev.pdf. [Retrieved 2020-03-29].
Sirohey, S. A., Hunjra, A. I. & Khalid, B. (2012). Impact of Business Process Automation
on Employees’ Efficiency. Bulletin of Business and Economics, 1(1), 1-12.
Suchman, M. C. (1995). Managing Legitimacy: Strategic and Institutional
Approaches. The Academy of Management Review, 20(3), 571–610.
Tschakert, N., Kokina, J., Kozlowski, S., & Vasarhelyi, M. (2016). The next frontier in
data analytics. Journal of Accountancy (online) Available via

84
https://www.journalofaccountancy.com/issues/2016/aug/data-analytics-skills.html
[Retrieved 2020/04/20]
Vallas, S. P. (1990). The Concept of Skill: A Critical Review. Work and Occupations,
17(4), 379–398.
Weinert, F. E. (1999). Concepts of Competence. [Online]. Available via:
https://pdfs.semanticscholar.org/8b88/efa9dd5e0a4b605aea6e5e3b9ec640beb089.pdf.
[Retrieved 2020-03-10].
Yilmaz, K. (2013). Comparison of Qualitative and Quantitative Research Traditions.
Epistemological, Theoretical, and Methodological differences. European Journal of
Education, 48(2), 311-325.
Yin, R. (2003). Case Study Research Design and Methods. Third Edition. CA: Sage
Publications.

Yin, R. K. (2016). Qualitative Research from Start to Finish. Second edition. New York.
(Ebook)

Yin, R. K. (2018). Case study research: design and methods. sixth edition. United States
(Ebooks).

Yoon, K., Hoogduin, L., and Zhang, L. (2015). big data as complementary audit evidence,
Accounting Horizons, Vol. 29, No. 2, pp. 431-438.
Zhang, J., Yang, X., & Appelbaum, D. (2015). Toward effective big data analysis in
continuous auditing. Accounting Horizons, 29(2), 469-476.

85
8. Appendix
8.1. Appendix 1: Email to prospective participants

To whom it may concern,

We are graduate students in accounting at Umeå University in Sweden. Currently, we are
conducting research on the impact of Big Data Analytics (BDA) on audit and audit
professions. To date, BDAs are a trending phenomenon that affects several fields. One of
the salient components of BDA involves auditing, which is affected by advanced
technology. Therefore, the purpose of our thesis is to find out the impacts of BDA on audit
procedures and both required knowledge and expertise from auditors to work in a new
environment.
The thesis will require interview processes to gather relevant data. Therefore, we would
like to interview auditors in your firm to provide us some insights. Due to the distance and
time considerations of doing face to face interviews, we would like to conduct the
interview via Skype or other possible means. The interview process will be undertaken in
English and taken for roughly 30-45 minutes in duration. Being cognizant of the auditors’
work schedule, the schedule for the interview will follow and be determined by the
auditors, respectively. However, due to the given time frame of writing the thesis, we
would be happy when the interview can be conducted at the beginning of April 2020.
Importantly, during the interview processes will include recording for transcription
purposes. Therefore, ethical aspects in academic research are considerably applied,
including the storing and usage of the data Information collected during the interview
processes will be accessible both for the authors and supervisor upon his request. Note,
however, all information collected is used for academic research purposes.
Anonymity will be fully granted. Personal information and potentially sensitive
information of the participants and the firm they work with will not be disclosed.
Participants will be represented by initials by the researchers in the draft and final copy of
the research paper. Participation in the interview is voluntary. With respect to the
participants’ volition, participants have the right to refuse to answer particular questions or
withdraw the consent at any time. The interview guide will be utilized during interview
processes in which it will be shared with the participant information sheet before the
interview.
We highly appreciate your valuable contribution. It helps us to complete our study and
broadly helps us to contribute to knowledge development in the area of big data and audits.
We look forward to your response. Should you need further information regarding our
research project, please let me know.

Kind regards,
Hamadou Kandeh & Mohamed Alsahli

86
8.2. Appendix 2: Interview guide

Warm-up Questions
1. Could you tell us about yourself? Education background, experience?
2. What is your role in this organization, and how is it connected to Big Data Analytics?
3. Has your role changed, or do you expect it to change as a result of using Big Data
Analytics?
4. What type of client (Large or small firms) do you audit?
Skills and Competence
1. What skills and level of expertise are required to perform an audit in this Big Data
environment?
2. Do you think more analytic specialists are required in the engagement team?
3. What kind of training programs do auditors need to utilize big data analytics?
4. What changes do you think the academic and professional curriculum could make to
prepare future auditors?
Judgment
1. What is the effect of big data on auditors’ judgment and decision making?
2. Do you check the reliability/veracity of external data (obtained from outside the audited
organization)? If so, how?
3. What are the effects of large volumes of data on auditor judgments in relation to
information overload, irrelevance information, and ambiguity?
4. How do auditors reach the conclusion that gathered evidence is appropriate and
sufficient by using big data tools?
Audit Efficiency
1. In your opinion, do you think BDA has an effect on time spent on engagement
activities? How?
2. What is the effect of big data on audit costs and audit fees? How much?
3. Do you think executing full population audits is more expensive than sample-based
auditing? Why?
Audit quality
1. How could BDA impact the audit work, i.e., audit test, audit procedures, sampling,
control, and substantive testing, etc.?

87
2. Does BDA have the potential to identify risk and material misstatements?
3. Is big data improving the reliability of the audit report and the information contained in
it? How?
4. How effective are engagement team members in using BDA during the audit process?
5. How big data improve interactions between the stakeholders in the audit process?

88
8.3. Appendix 3: Information sheet

Background
We wish to take this opportunity to thank you for accepting to participate in our thesis
project. The purpose of this research is to explore the effect of big data Analytics on the
audit processes, audit judgment, efficiency, and quality, as well as auditors’ skills and
competence. Together with this information sheet, an interview guide containing the main
questions that will be asked during the interview session is attached and sent to the
participants prior to the interview. Participants are drawn both within and outside the
Swedish audit industry; most of the participants work in the big four audit firms in the
world.
Participation
The participant will have an interview using Skype or telephone calls. The interview
session is expected to last for approximately 30- 45 minutes. We will require your
permission to audio record the interviews to enable us to fully transcribe the interview
process to ensure accuracy. The interview will be conducted in the English language
between the last week of March and the first week of April 2020.
Confidentiality and Use of collected data
The anonymity of participants will be fully granted. Names and organization of
participants may only be available for the researchers and the supervisor of the thesis upon
his request. Names will be replaced with letters or some other forms of nomenclature to
protect participants’ identities.
Data collected during the interview process will remain the right of interviews before the
final draft of the thesis is published. The content will only be available for the authors and
the supervisor and utilized purposely for the objective of the study.
The authors may make use of direct quotations in the manuscript. Should the authors use
direct quotations, participants will be notified. A copy of the draft document will be sent to
participants for review. Participants can correct, complete, or request the complete
omission of quotes from the document.
Finally, we wish to reiterate the point that participants can withdraw their consent of
participation in the research and will respect their decision. Also, participants have the
right to refuse to answer questions.
Hamadou Kandeh Mohamad Alsahli
073 6464763 072 9074003
[email protected] [email protected]

89
 8.4. Appendix 4: List of contacted big four Forms for interview

Region Country EY Deloitte PWC KPMG BDO GT

China * * *
India * * *
Korea *
Singapore *
Asia

Pakistan *
Taiwan * *
Malaysia * *
Indonesia `` * *
Japan * * * *
Ghana *
Nigeria *
Senegal *
Africa

South Africa * * * *
Rwanda *
Ivory Coast *
Egypt * *
Kenya * *
France * *
Germany * *
UK * * * * *
Netherland * * * *
Europe/ Oceania

Belgium * * * *
Denmark * * * *
Finland * * * *
Norway * * * *
Switzerland * *
Sweden * * * * * *
Newzealand * * *
Australia * *
Qatar *
UAE * * * *
Middle East
Saudi Arabia * * *
Jordan * *
United States * * * *
North/South Canada * * * *
America Brazil * *
Chile * *

90
Business Administration SE-901 87 Umeå www.usbe.umu.se