You are on page 1of 13

Journal of Accounting Education 38 (2017) 3749

Contents lists available at ScienceDirect

Journal of Accounting Education


journal homepage: www.elsevier.com/locate/jaccedu

Educational Case

Manual journal entry testing: Data analytics and the risk of


fraud
Rebecca Fay a, Eric M. Negangard b,
a
East Carolina University, Slay Hall 331, Greenville, NC 27858, United States
b
McIntire School of Commerce, University of Virginia, United States

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

Article history: Big Data is revolutionizing the business world as it enables companies to discover valuable
Received 30 June 2016 insight from within the vast volumes of data now available to companies. The data analysis
Received in revised form 21 December 2016 skills related to Big Data are in high demand, yet many accountants lack the data analysis
Accepted 21 December 2016
skills necessary to improve decision making within a company or increase the effectiveness
Available online 10 January 2017
and efficiency of an audit. This case provides you with a hands-on opportunity to utilize the
data analysis skills that are in such high demand. You will harness the power of Big Data
Keywords:
while performing a procedure that is required on all financial statement audits an anal-
Journal entry testing
Data analytics
ysis of journal entries for potential red flags of fraud. The case is completed in two phases.
Big Data First, in Phase I, you will learn how to address one of the primary challenges in the use of
Completeness testing Big Data validating the data which ensures that the files are complete. Specifically,
PCAOB inspections you will import and analyze client files in IDEA to determine if the manual journal entry
Forensic file is complete or if any records are missing from the file. In Phase II, you will validate data
Audit from a second client then perform a battery of tests aimed at identifying potential red flags
Fraud of fraud. To complete the second task, you will need to consider factors such as who
Case study
recorded the journal entries, when the entries were created, the description provided (or
lack thereof), and whether the entries were back-posted or out-of-balance. You will also
learn how to use fuzzy matching and Benfords Law.
2016 Elsevier Ltd. All rights reserved.

1. Case

1.1. Background information

Big Data is impacting all aspects of business and is quickly becoming a top focus for accountants regardless of their spe-
cialty financial accounting, management accounting, auditing, forensic, tax advising, education, or regulation both within
the United States and internationally (AAA, 2016, CGMA, 2014, Cokins 2014, Maxwell, Steiner & Stein, 2013, Murphy, 2014;
Novack, 2013). The term Big Data refers to the sheer volume and speed at which data is being generated through electronic
transactions, website clicks, texts, Facebook activity, photographs, videos, and countless other means. Enterprising compa-
nies are finding that effective use of Big Data can provide valuable information about their customers. Beyond just demo-
graphics and prior purchases, companies can now use browsing preferences, website views and clicks, environmental

Corresponding author.
E-mail addresses: fayr@ecu.edu (R. Fay), ericnegs@virginia.edu (E.M. Negangard).

http://dx.doi.org/10.1016/j.jaccedu.2016.12.004
0748-5751/ 2016 Elsevier Ltd. All rights reserved.
38 R. Fay, E.M. Negangard / Journal of Accounting Education 38 (2017) 3749

Table 1
Benefits of generalized audit software (GAS).

 Tests entire populations of data rather than just a sample


 Analyzes data from a variety of sources including report files and unique data formats from industry-specific software
 Aids in risk assessment by identifying trends and anomalies in the data
 Facilitates comprehensive analysis of general ledger transactions
 Processes large volumes of data almost instantaneously
 Maintains the integrity of the source data by storing and presenting imported data in a read-only format
 Performs common audit procedures with the click of a button
 Works with any file size and an unlimited number of records
 Creates a log of every action the user takes (Brown-Liburd et al., 2015; Murphy & Tysiac, 2015; Smith, 2015)

factors (such as weather), and social media trends to predict their customers purchasing patterns, enhance customer satis-
faction, and improve their own bottom line (Brown-Liburd, Issa, & Lombardi, 2015; Gray & Debreceny, 2010).
Big Data has provided opportunities to enhance accounting information and audit quality as well. In its earliest and sim-
plest form, accountants began utilizing Big Data when computer assisted auditing techniques (CAATs) enabled the testing of
entire populations of data, consisting of billions of records (such as the 1 million customer transactions Walmart generates
every hour), rather than relying on sampling (Data, 2010). For decades, audit firms have relied on the use of data analysis
software known as generalized audit software (or GAS) to enhance the quality of an audit. IDEA and ACL, the two most
commonly used GAS packages, provide auditors with numerous benefits as listed in Table 1. Most audit firms have members
of staff trained in the use of GAS and many expect even entry-level staff to perform data analytics.
As technological capabilities have improved, the profession has expanded to incorporate new nonfinancial sources of
information. For instance, sentiment data from Twitter and other social media sources can predict product demand, an
increase in product returns, or a decline in inventory value. Weather information can be compared with sales by geographic
location to identify unexpected spending patterns; searching the content of electronic documents such as email can help
companies identify deviations from company policies or red flags of fraud (Cao, Chychyla, & Stewart, 2015; Fay &
Negangard, 2016).
The U.S. Securities and Exchange Commission (SEC) is drawing on data analytics tools to enhance the likelihood of detect-
ing financial statement fraud in its registrants. It developed an Audit Quality Model, commonly referred to as RoboCop, that
analyzes the financial statements of public company for heightened fraud risk within 24 h of being submitted to the SEC. The
model looks for risk factors such as when companies use accounting techniques to report aggressively high earnings while
using an alternate method for tax that minimizes taxable income. More recently, the SEC has expanded the use of data ana-
lytics to assess the narrative information included in corporate filings for heightened risk of fraud. For instance, the software
can scan the Management Discussion & Analysis for a particular list of words and phrases used by prior fraudulent filers, a
negative tone, or language intended to obfuscate complex issues (Novack, 2013). Scott Bauguess, Deputy Director and Dep-
uty Chief Economist of the SECs Division of Economic and Risk Analysis (2016) cautions that the data analytics conducted by
his division of the SEC does not reduce the need for human involvement. To generate high quality results, each risk identified
through data analysis must ultimately be assessed by the SEC staff with significant experience and insights into the market
(Bauguess, 2016).
The largest accounting firms are similarly embracing cutting edge technology to enhance the traditional audit. While GAS
was the earliest, and currently the most widespread, implementation of Big Data in the accounting profession, the develop-
ment of additional Big Data tools (e.g., telematics, drones, and text and unstructured data analysis, Hadoop, Distributed File
System) has created opportunities for accountants and auditors (Applebaum & Nehmer, 2016, Brown-Liburd et al., 2015; Fay
& Negangard, 2016). Big Data techniques can enable auditors to compare data from multiple sources within the company
(such as using revenue or procurement cycle sub-ledgers to verify amounts reflected on the general ledger) and to evaluate
the data in real-time. Additionally, Big Data allows auditors to gain a more thorough understanding of the clients accounting
and processes which should result in more accurate risk assessments and ultimately higher audit quality. However, many
believe that the value of integrating Big Data and analytics into the audit will only be realized when used by auditors
to. . . develop new skills focused on knowing what questions to ask of the data, and the ability to use analytics output to pro-
duce audit evidence, draw audit conclusions, and derive meaningful business insights (Ramlukan, 2015, 12).
To that end, the largest accounting firms are using software and other cognitive technologies to identify patterns in data
indistinguishable by the human eye. Most notably, KPMG has contracted with IBM to utilize Watson, the software best
known for defeating Jeopardys two greatest champions, to enhance its audit and advisory services. Watson can process large
volumes of unstructured data and provide meaningful insights in English (and seven other languages) (Best, 2016; Claburn,
2016; IBM, 2016). KPMG Chairman and CEO Lynne Doughtie states, KPMGs use of IBM Watson technology will help
advance our teams ability to analyze and act on the core financial and operational data so central to the health of organi-
zations and the capital markets. In addition to the unprecedented possibilities for enhancing quality, the potential for cog-
nitive and related technologies to help us pursue new business offerings is extraordinary (Lee, 2016). While the
opportunities related to Big Data are virtually endless, a survey of 2000 companies found that 86% are struggling to find
the valuable information buried amidst the overwhelming amount of meaningless data (Chartered Global Management
Accountant (CGMA), 2014). Studies estimate the demand for data analytics will double in the near future, providing oppor-
tunities for professionals in information technology, finance, and accounting. McKinsey Global Institute (2011) forecasts a
R. Fay, E.M. Negangard / Journal of Accounting Education 38 (2017) 3749 39

heightened demand through 2018 stating, The United States alone faces a shortage of 140,000 to 190,000 people with
analytical expertise and 1.5 million managers and analysts with the skills to understand and make decisions based on the
analysis of Big Data.
This case provides you with a hands-on opportunity to learn and utilize some of the data analysis skills that are in such
high demand. You will harness the power of Big Data while performing a procedure that is required on all financial statement
audits an analysis of journal entries for evidence of fraudulent financial reporting.1

1.2. Case materials: Phase I verifying accuracy and completeness of electronic files

For the entire case, you will play the role of an external auditor who has been asked to analyze journal entries for a height-
ened risk of fraud. You will utilize IDEA, one of the most common GAS packages (a form of CAAT), to perform your analysis.
The case will be completed in two phases. In Phase I, you will address one of the primary challenges with Big Data vali-
dating the data prior to performing analysis by assessing clerical accuracy and testing the files provided by the client
for completeness (Ramlukan, 2015). In Phase II, you will expand your skills by performing a battery of tests to determine
which, if any, journal entries should lead to a heightened risk of fraud.
You have been working full-time for a CPA firm since your graduation eight months ago and have impressed Greg Max-
well, the manager who supervises you, with your accounting knowledge, analytical skills, technical expertise, and attention
to detail. As you walk into the office he greets you.
Greg: Good morning. When you get settled, stop by my office. I have a new assignment for you.
You: Im ready to get started. What will we be working on?
Greg: I have a new audit client, Transport Titania, that I think would be great for you. It is a multinational company spe-
cializing in transportation equipment and consulting, with operations in the US, Europe, Australia, and Singapore. Since
you have experience using IDEA, I would like you to assist me on this client. We will start by testing the companys journal
entries to help us meet our audit objectives related to the consideration of fraud risk.
You: That sounds ominous. Is there something specific you are concerned about?
Greg: No, I havent heard any allegations or rumors of fraud if that is your question. But remember, external auditors are
required to evaluate journal entries for evidence of fraudulent financial reporting on every financial statement audit.
You: Right, I remember my audit professor mentioning that. I also remember learning about a specific company that
used journal entries to intentionally manipulate its financial statements.
Greg: Yes, unfortunately there have been quite a few. WorldCom used journal entries to inappropriately capitalize
expenses as fixed assets, inflating its net income and total assets by $3.5 billion (Beresford, Katzenbach, & Rogers,
2003). Cynthia Cooper, the internal auditor, actually discovered the fraud by analyzing the companys journal entries.
Adelphia concealed various misappropriation of assets and around $2.3 billion in loans that were made to the executives
using inappropriate journal entries. Audit standards require auditors to consider the overstatement of revenue and
receivables; the two accounts most commonly manipulated. In order to detect potentially fraudulent journal entries, it
is critical that we, as auditors, understand the various types of entries posted in the accounting system in order to appro-
priately assess risks related to each type of transaction. Here is a document (see Fig. 1) that summarizes some important
information about the various types of journal entries.

You review the page that Greg hands to you before continuing.
You: So, we need to consider risk factors for the particular client, industry, and operations, and then begin testing its
journal entries for fraud.

Greg: Yes, but before that, the first thing we need to do is make sure that the data file provided by the client includes the
detailed information for all of the manual journal entries. In other words, I need you to validate the data by testing it for
completeness.
You: What would we be looking for? If the data is exported directly from the clients accounting system wont it auto-
matically, by design, be complete?
Greg: In most cases, yes. But, imagine if the client exported the wrong date range, accidentally missed a certain class of
transactions, or worse yet, intentionally omitted fraudulent entries. Before we begin our testing procedures we must
always verify that we have a complete set of transactions. It seems like a simple step, but it is one that auditors must
always be vigilant about conducting. Both the AICPA (2003) and the Center for Audit Quality (CAQ) (2008) have issued
practice alerts stressing the importance of this step.2 Despite these alerts, when the Public Company Accounting Oversight

1
This procedure is required by AS 2401 (PCAOB, 2016, 2015b, 2015c, 2015d) and AU-C 240 (AICPA, 2016) for public and private companies, respectively.
2
It is important in testing journal entries and other adjustments that the auditor be aware of and consider the entire population of journal entries and other
adjustments. The auditors ability to detect fraud is adversely affected if he or she is not assured of access to all of the journal entries posted and other
adjustments made during the audit period. (AICPA, 2003, emphasis added).
40 R. Fay, E.M. Negangard / Journal of Accounting Education 38 (2017) 3749

Automatic vs. manual journal entries

Accounting systems generally consist of two forms of journal entries. Automated journal
entries originate from the various subsidiary ledgers and post standard, recurring transactions
directly to the general ledger. Manual journal entries, on the other hand, are prepared by individuals
to capture economic activity outside of the subsidiary ledgers. Some of these manual entries
represent standard, recurring business transactions (e.g., adjusting accruals and prepaid accounts
balances) that should be addressed by the companys internal controls over journal entries.
The remainder of the manual entries, used to record unusual events or adjustments to reserves,
are called non-standard entries (also manual adjustments, or post-closing adjustments) and may slip
through the companys internal controls or even represent an instance of management override, in
which the company makes last minute adjustments to their ledger, without any legitimate business
purpose, to manipulate financials into showing the specific amounts of revenue, net income, assets,
or liabilities.
In some cases, management prepares top-side entries in which they adjust account balances
using a spreadsheet after the trial balance has been closed; this form of adjustment outside of the
accounting system results in managements desired impact on the financial statements without
recording any entries in the system, thus bypassing all internal controls. Xerox to the tune of $3
billion, Health South $2.7 billion, Keystone Bank $500 million, and Cendant $161 million, are just a
few of the companies that used non-standard journal entries to inflate income (Carmichael 2010,
DeVries & Kiger 2004, Gray & Debreceny 2010, Loraas & Searcy 2010, Weld, Bergevin, & Magrath
2004, Wallace 2000).

Fig. 1. Automatic vs. manual journal entries. (See above-mentioned references for further information).

Board (PCAOB) inspects the workpapers of registered accounting firms, one of the most common deficiencies it finds is that
the auditor failed to adequately test client-provided data for completeness.
You: Are you talking about the smaller firms that only audit one or two issuers?
Greg: The PCAOB (2015a) has issued a report summarizing the numerous deficiencies with how smaller firms address
the risk of fraudulent journal entries, but these deficiencies are not limited to small firms. As you can see, each of the Big
Four has been cited for failing to validate data that is, make sure it is complete before they conduct further testing, and
many have failed to appropriately test journal entries for fraud risk.
You review the document that Greg hands you (see Fig. 2).
Greg continues: This test of completeness is critical for the audit. I will provide guidance and review your work along the
way to ensure that we satisfy all requirements of the standards. This is an excellent opportunity for you to think through
the process and begin to envision what it involves. Once you have had a chance to digest everything and develop a for-
malized plan, I will evaluate your approach and we can discuss the strengths of your analysis and opportunities for
improvement. This is a chance for you to exercise your analytical skills and gain experience in an audit area that the
PCAOB has highlighted as critical for future audits.

1.2.1. Phase I requirements

1. Create a new project in IDEA and import the following data files provided by your manager (instructor):
a. ChartofAccounts.xls
b. Control Totals.txt
c. Summary of Automated JEs.xls
d. TrialBalance.xls
e. Manual JEs.txt
R. Fay, E.M. Negangard / Journal of Accounting Education 38 (2017) 3749 41

Observations from PCAOB inspections

The PCAOB (2015a) recently issued a report summarizing their observations from inspections of
audits conducted in 2012 and 2013. Within this report, the PCAOB identified instances where one or
more firms failed to complete the following required audit procedures as it relates to the testing of
journal entries:
- Obtain a sufficient understanding of how fraud could be perpetrated or concealed by the
issuer in order to appropriately tailor their tests for the issuer
- Tailor risks, resulting in deficiencies in identifying risk factors related to manual journal
entries
- Consider the risk of management override of controls
- Perform procedures to test journal entries and other adjustments for possible material
misstatement due to fraud
- Obtain an understanding of the company and its environment
- Understand the companys controls over journal entries
- Evaluate whether the controls were properly designed
- Perform sufficient procedures to identify and assess risks of fraud related to journal entries
when selecting entries for testing
- Tailor the sample for fraud risk
- Evaluate the results of CAAT of journal entries
- Assess which journal entry categories should be included in testing
- Perform a critical assessment of specific risks of fraud presented by the journal entries
These deficiencies are not limited to small firms. In their most recent annual inspections, each of
the Big Four had at least one finding related to the validation of data. In most cases it is because
they failed to test their clients controls over the completeness and accuracy of source data for a
variety of line items revenues, costs, investments, derivatives, insurance related assets,
estimated tax credits, fair value of leases, actuarial reports for loss reserves, variance reports,
internal controls over financial reporting, and fraud risk (PCAOB 2015b, c, d, e). Further, two of
the Big Four were specifically cited for failing to test client-provided data for completeness
before conducting tests on the data; one for journal entry testing (PCAOB 2015e) and the other
for revenue and inventory (2015c).

Fig. 2. Observations from PCAOB inspections. (See above-mentioned reference for further information).

2. Test for the completeness of the manual journal entry file to determine whether or not the file represents all the manual
journal entries for the period. Note, this process takes considerable thought and you should familiarize yourself with each
of the files prior to attempting the completeness check. It may help you to create a diagram of the inputs and outputs of
the journal entry process.

1.2.2. Phase I items for submission


Upon completing the assignment, submit the following deliverables:

1. Your project history file for the final database used in your test of completeness. Expand all in the history settings to show
all detailed activity in the final database and save in PDF format.
2. A screen print of your IDEA Project Overview which can be pasted at the conclusion of the memo created for #5 below.
3. An archive of your project.
4. A presentable MS Excel file that shows all reconciled and unreconciled accounts. This analysis should be conducted in
IDEA, exported to MS Excel, and formatted as a method of sharing your results with members of management who
are not familiar with IDEA.
5. A written memo describing your reconciliation efforts. It may help you to think of this memo as a workpaper describing
your procedures performed. This memo should discuss:
a. the purpose of performing the reconciliation process,
b. the source files used and what they represent,
c. the steps used to complete the process, and
d. the related implications of your findings.
42 R. Fay, E.M. Negangard / Journal of Accounting Education 38 (2017) 3749

Your memo should be written such that another person with equal expertise could re-perform the same procedures and
get a similar result. It should clearly demonstrate your conceptual understanding of the task you have completed.

1.3. Phase II evaluating journal entries for evidence of fraud

Congratulations! It is obvious by the expression on Gregs face as he approaches your desk that he was impressed with
your work on validating the journal entry files for Transport Titania.
Greg: Since you did an excellent job confirming completeness of the manual journal entry file for Transport Titania, I
have journal entries from another company for you to analyze. After validating the data, you can extend your work to
test the manual journal entries for anomalies and a heightened risk of fraud. The company, Epicenter, is a Manhattan-
based company that provides both products and services in the telecommunications sector. Epicenter utilizes the per-
centage of completion method to account for most of its revenue.
You will focus on risk factors identified in the audit standards such as the employee who created the journal entries, the
day and time entries were created, the accounts that were affected, the description provided (or lack thereof), back-
posting of entries into prior periods, round numbers, patterns of identical small entries, out-of-balance entries, and Ben-
fords Law (Benford, 1938). If you are unfamiliar, with the concepts of Benfords Law or fuzzy matching, I have prepared
more detailed descriptions for you (see Fig. 3).
You skim the document that Greg provides and open the client provided data files. While you appreciate that the docu-
ment describes the more complex IDEA functions, you quickly realize you will need clarification on the actual data files.
You: Greg, Im not sure that I understand all of the date fields that are included in the clients journal entry file. When we
prepared journal entries for school there was only one date except when we discussed dates related to stock dividends
and other transactions with multiple dates. Do you have time to explain these to me?
Greg: Thats an excellent question. I think I have a document that can get you started.
Greg hands you an overview of journal entry dates (see Fig. 4). You begin skimming the document as you return to your
computer, hoping that it will help explain why there are so many dates in the clients system!
Greg: Oh, one more thing, Greg calls after you take a few steps. The clients controller mentioned that she might have
re-used a journal entry number or two. See if you can find a way to identify or develop a unique journal entry identifier
that can distinguish between these duplicate values in the Journal Entry ID field.

Fraud detection functions in IDEA

Benfords Law In the early 1920s, Frank Benford, a physicist at GE realized that numbers in the
real world were more likely to have a low number as the first digit (e.g., 1 or 2) than a high number
(e.g., 8 or 9). Benford used this realization to predict the likelihood of digit frequencies across a wide
variety of data, including financial data. Auditors can compare the frequency of digits predicted by
Benfords Law to the actual frequency in a data file. Any category of numbers outside of
expectations (with a margin for error) represents a heightened risk of fraud. In IDEA, auditors can
conduct this sophisticated data analysis technique with the click of a button. Additional information
on Benfords Law may be found in Nigrini 1999, available at
http://www.journalofaccountancy.com/issues/1999/may/nigrini.html.
Fuzzy Matching Fuzzy matching allows database users to select any records that sound like or
are similar to a search term of interest. This function is often used by auditors to identify
heightened risk of fraud. For instance, a common audit procedure is to compare the addresses of
company employees to the addresses of company vendors in order to identify any shell companies
created by employees to fraudulently redirect corporate funds to their personal accounts. Traditional
Boolean comparisons would compare the addresses, but would be deceived with minor changes in
the address (e.g., Drive vs. Dr. or 432 Meadows Apt A vs. 432-A Meadows Apartments). Fuzzy
matching allows the auditor to match any items similar in content or similar phonetically. See
IDEAs description of fuzzy matching available at
https://www.audimation.com/Resources/Articles/applying-fuzzy-logic-to-acquire-clear-results-1310.
For an excellent discussion of why Fuzzy Matching is beneficial, see
www.ideasillustrated.com/blog/tag/fuzzy-matching

Fig. 3. Fraud detection functions in IDEA. (See above-mentioned reference for further information).
R. Fay, E.M. Negangard / Journal of Accounting Education 38 (2017) 3749 43

An overview of journal entry dates

- Created Date The created date reflects the date when a user (typically an accountant) actually begins to
generate a complete or partial journal entry in the accounting information system (AIS). Employees can
create journal entries in the AIS throughout the day as events occur, information becomes available, or the
preparer has time. Some AISs, such as the one used by the client in this case, record the creation date for
each line item (e.g., individual debits and credits) rather than for the journal entry overall. As a result,
different line items (e.g., debits and credits) can show as being created on different days and/or at different
times.
- Approved Date As an internal control over journal entries, companies often require one employee to
record the journal entry in the system while a second, usually more senior or authoritative employee,
reviews the entry along with supporting documentation and provides his/her approval. The approved date
reflects the date at which this control was completed. Review of the journal entry file for this case suggests
that the company either does not implement this particular internal control or does not take advantage of the
AISs capability to record and document the approval process.
- Batched Date Historically, complex AISs were designed with independent subsidiary ledgers and a
general ledger. Transactions would be recorded in the special journals of subledgers (e.g., sales, cash
receipts, and cash disbursements) throughout the day but would effectively serve as pending transactions,
with no impact on the companys financial statements, until they were batched for processing. This allowed
an opportunity for employees to conduct a final review as they selected entries to be batched and prepared
for the final posting to the general ledger.
Batch processing often includes certain automated checks on the data performed by the AIS (e.g.,
verifying account numbers entered represented valid accounts on the trial balance) before the entries get
posted and became final. Thus, batch processing serves as both a control and a practical solution since 1)
processing large volumes of transactions during normal business hours could tax the system and limit other
users access until the processing was completed, and 2) many legacy systems lacked functionality to edit
transactions after they were posted as final. As a result, batch processing can eliminate the need to generate
new entries whose sole purpose is to reverse or adjust transactions that were recorded in error.
Oftentimes, companies have scheduled times at which they batch process transactions to ensure that
they are processed in an efficient fashion. This process may either be automated or require an employee to
initiate the batch. Review of batch times included in this case suggests that the company automatically
batches journal entries daily at midnight. Many current AISs have transitioned to real-time processing, for
some or all of their ledgers, in order to have accurate and reliable information available real-time. These
systems utilize a Changed Date to retain an audit trail for entries that have been modified.
- Posted Date The posted date is the point in time when transactions are considered fully processed,
reviewed, and approved and were released to the general ledger. As mentioned previously, many legacy
systems do not allow users to edit a transaction after it has been posted to preserve the audit trail.
- Effective Date The effective date reflects the date when a transaction impacts the financials. For instance,
if an employee receives an invoice on January 5 for goods received (and title transferred) on December 22,
the Effective Date of the transaction should be December 22 while the Created Date, Batch Date, and Posted
Date will be in January when the accountant received the invoice and began entering the transaction.
- Reversed Date Many journal entries recorded in the general journal are intended to adjust for accruals and
prepaid amounts. As a result, some of these transactions must be reversed in a subsequent period. For
instance, if a company pays $300 on December 1 for insurance coverage December 15 January 15, at the
end of December the company will need to make an adjustment for prepaid insurance. The company would
prepare a journal entry with an effective date of December 30 that debits Prepaid Insurance and credits
Insurance expense for $150. On January 1, that entry would need to be reversed in order to reduce the
prepaid asset and record the portion of Januarys insurance expense that was paid in December. For
transactions like this one that will be completely reversed in a subsequent period, many AISs allow the user
to simply indicate the date that the transaction should be reversed rather than creating a new entry in the
subsequent period. Review of the journal entries for this case suggests that the company does not take
advantage of this time-saving function in the AIS.

Fig. 4. An overview of journal entry dates.

1.3.1. Phase II requirements


To assist you in completing the assignment, Greg has provided you with a specific list of procedures to be performed. He
has asked you to create an MS Word document to record the results of each test.

Procedure #1

Create a new project in IDEA. Import the following data files provided by your manager (instructor) and determine
whether the file of manual journal entries is complete. Report the steps you performed as well as your results. (Note, while
there is no account name field included in the files, the Line_Description field appears to reflect the account name).

a. AccountTotals.txt
b. June2017ManualJEs.MDB
44 R. Fay, E.M. Negangard / Journal of Accounting Education 38 (2017) 3749

Procedure #2

Check the June Manual Journal Entry file for data anomalies. What date range does the file represent for each of the fol-
lowing dates?

a. Effective dates
b. Posted dates
c. Created dates

Procedure #3

Create a database that shows the total number of manual journal entries for the month of June (hint: You need to do a
summarization by unique journal entry, not line item.)

a. Considering Gregs warning about the reuse of journal numbers, what field or combination of fields could be used to
develop a unique identifier?
b. Do you have any recommendations for how the client can strengthen their internal controls over journal entries?
c. How many total journal entries were effective in June 2017?
d. Are there any out of balance journal entries? If so, how many?
e. What is a possible explanation for the out of balance journal entries?

Procedure #4

Export the following transactions to an MS Excel document to share with members of management and other members of
the audit team:

a. The top 10 accounts with the largest total credit entries


b. The top 10 accounts with the largest total debit entries
c. Which of the accounts would you recommend for further investigation? Be sure to provide your rationale for your
decision.

Procedure #5

Identify any journal entries prepared at unusual times:

a. Late nights
b. Weekends
c. Holidays
d. Please discuss which, if any, journal entries you recommend investigating further and the rationale supporting your
decision.

Procedure #6

Identify any journal entries that are missing descriptions (either journal entry or journal line descriptions).

Procedure #7

Determine whether any journal entries are back-posted more than five days into a prior period.

Procedure #8

IDEA has several functions that utilize fuzzy matching to determine whether words or phrases sound similar or represent
a similar concept. Use the @SimilarPhrase to determine if any records reflect the same in formation in the JournalDescription
and the LineDescription.

Procedure #9

Run Benfords Law on the journal entry amount field. Does anything look anomalous for Benfords first digit, first two dig-
its, or first three digits?
R. Fay, E.M. Negangard / Journal of Accounting Education 38 (2017) 3749 45

Procedure #10

Identify any journal entries that have the word restate in either the journal entry description or the line item
description.

Procedure #11

a. Create a graph showing total debit amount by user ID, Is there any user you would want to investigate or test further?
b. What information would you request to help you make a more informed decision for 11a?
c. Are there any journal entries that are missing a user ID?

1.3.2. Phase II items for submission


Upon completing the assignment, submit the following deliverables:

1. A print screen of your File Explorer window in IDEA showing all your expanded databases (which can be pasted into #3
below).
2. A detail history file from IDEA showing at least one of the procedures you performed.
3. An archive of your project.
4. A MS Word document containing your answers to the questions above. Be sure to provide thorough answers to demon-
strate the level of thought you have applied to each question.
5. A memo written to Epicenters audit workpaper file, summarizing the conclusions from your analysis. In drafting this
memo, you should review your work and identify which results warrant the highest priority for further analysis. The
memo should, at a minimum:
a. Document the results of your procedures and highlight which of your findings you deem to be of greatest importance
and priortity
b. Discuss any areas of heightened risk identified by your analysis,
c. Discuss the nature of the tests that were completed, including support images when appropriate, and
d. Recommend specific follow-up tests to be completed.

2. Teaching notes

2.1. Case overview and learning objectives

The Center for Audit Quality (2008) recommends the use of CAATs such as IDEA for the testing of journal entries both to
test client-provided data for completeness and to perform analytical procedures on data such as journal entries. This data
analysis performed on a population of journal entries is the primary form of data analytics used in auditing today. This case
provides students with the opportunity for hands-on experience performing these procedures in IDEA, one of the top two
GAS packages currently utilized by both internal and external auditors.
Phase I of the case focuses on the essential step of validating data to ensure a file is complete before performing any
audit tests. This key procedure has been highlighted by the PCAOB as a step that audit firms of all sizes, from Big Four down
to the smaller triennially audited firms, have overlooked and omitted on their audit engagements. This case begins to address
the PCAOBs (2015a) concern that firms may not place the appropriate level of importance on, or may not provide adequate
training with respect to, testing journal entries. Specifically, this task involves importing five client files into IDEA and using
key functions (join, summarization, append) to determine whether the client-provided manual journal entry file is complete.
Phase II of the case repeats the test of completeness on data obtained from a second client, then proceeds to analyzing journal
entries for red flags of fraud an audit procedure required for every audit of financial statements. After completing the valida-
tion test, students utilize a variety of IDEA functions to look for common risk factors such as entries that are back-posted, lacking
descriptions, or created on unusual dates or at unusual hours. The specific learning objectives of the case are as follows:

- Increase knowledge of data analytics, including the ability to work with large data files and the ability to identify a unique
journal entry identifier
- Understand the assessment of risk related to journal entries and learn to identify classes of journal entries and specific
entries with heightened levels of risk
- Demonstrate the ability to perform tests for large and unusual journal entries

These learning objectives reflect learning across multiple levels of Blooms (1956) taxonomy, from Understanding to Eval-
uation. Students begin with an Understanding of the methods of analyzing large data sets (including the validation test pre-
sented in Phase I) and factors that heighten fraud risk. In Phase II, students are asked to Apply this knowledge to a second
client. After completing the required procedures for Phase II, students must Analyze the data to identify any heightened fraud
46 R. Fay, E.M. Negangard / Journal of Accounting Education 38 (2017) 3749

risk. At the conclusion of the audit, each student must Evaluate their results to identify the findings with the highest priority.
Each student will document the results of his/her evaluation, providing support for his/her conclusion.

2.2. Implementation

The case has been utilized and refined over several years with students from three large public universities. In the most
recent semester of use, the case was implemented in a senior level audit class at a large public university, and a Masters level
(MSA) Forensic Accounting course at a second large public university. This case has been designed for use as either an out-of-
class assignment, with the instructor taking time to introduce and conclude the case, or as an in-class lab in which the
instructor can circulate throughout the classroom answering questions and helping students troubleshoot syntax and
parameters specific to IDEA functions. The case is typically assigned at or after the halfway point of the semester. If the case
is used earlier in the semester, the instructor may wish to provide additional instruction on fraud and internal controls.
The case is designed in two parts, to allow students to focus on the essential task of validating data files by testing for
completeness before conducting further procedures. In Phase II, students apply the validation test to journal entry files from
a second company before analyzing the content of the journal entries for indications of fraud as required by auditing stan-
dards (AU-C 240 for private companies and AS 2401 for public companies).
For the MSA course, students completed IDEAs introductory workbook (available through Casewares academic program)
outside of class as a prerequisite for the assignment. The instructor then spent a portion of the class period introducing the
nature of the assignment and tasked students with developing a method of testing the completeness of the journal entry files
without specific instructions. After Phase I was completed and debriefed, the instructor assigned Phase II as a homework
assignment. In order to maximize sharing and learning, the students were told they could work together on the assignment,
however, each individual had to complete the case and turn in his/her own work product. Graduate students have, on aver-
age, spent three to four hours completing the data analysis for Phase I, and five to ten hours completing the analysis for Phase
II. The instructor has found it beneficial to provide a check figure midway through the time allowed for Phase II (e.g., the
number of journal entries from Procedure #3c) to allow students the opportunity to assess their performance, re-evaluate
their approach, and seek guidance if necessary.
The undergraduate course designated one day to introduce the class to IDEA and three days to complete the case. The
instructor spent the first day guiding students through basic functionality of IDEA. During the second class period, students
broke into small groups to consider the challenge of testing completeness of the data files for Phase I. After discussing the results
together as a class, the instructor led the class through the necessary steps to summarize, join, and append the files to complete
the first phase. The class then worked as individuals or in pairs to complete the second phase of the assignment over the next
two class periods. The instructor provided a handout with abbreviated instructions for the necessary IDEA procedures (avail-
able with supplemental files), and circulated throughout the room to answer questions and assist with syntax errors.
IDEA software is available through multiple formats. It comes packaged with many audit textbooks. It is available via an
academic license for installation on university servers, and an academic version can be downloaded by students directly
from IDEA at no cost after the instructor has completed the initial registration process. For this particular case, it is essential
that the instructor utilizes the full Education Version rather than the demonstration version. The full Education Version
allows the user to analyze up to 10,000 records, while the demonstration version is limited to 1000 records. This case incor-
porates over 9000 items of ledger activity. It is also important to note that IDEA requires a Windows operating system. Stu-
dents with Mac computers can use the software by installing Windows on their machine via an emulator such as Virtual Box,
Parallels, or Bootcamp or, if the university installs IDEA on a server, students may access the server using a Mac or PC using
remote desktop. In addition to the use of IDEA, this case requires students to export their results to MS Word and MS Excel, to
practice the steps necessary to share the results of their analysis with non-IDEA users.
Student work was graded with 25% allocated to the memo for Phase I, 20% allocated to other deliverables for Phase I, 25%
allocated to the memo from Phase II, 25% allocated to the MS Word document from Phase II, and 5% allocated to other deliv-
erables for Phase II. All student files were graded using a rubric. A detailed rubric is included in the solution files available to
instructors. The MS Word document provided by students was graded for accuracy by comparing it to the solution; all ques-
tions were equally weighted.

2.3. Data files

The data files used in both Phase I and Phase II were derived from the accounting records of a real company with sub-
stantial modifications. The files were made available for educational purposes. All files have been digitally scrubbed and san-
itized in order to remove any identifiable information that could link them back to the original company. Red flags of fraud
were seeded into the data for teaching purposes, and the dates were converted to 2017 to promote realism for present day
students. As such, the data files include red flags of fraud that students should identify while completing the case, but no
determination of fraud could be made without additional testwork.3 The analysis conducted in this case can help identify

3
Students are asked to recommend follow-up procedures based on the results of their analysis. Completion of the recommended testwork is outside the
scope of the current manuscript.
R. Fay, E.M. Negangard / Journal of Accounting Education 38 (2017) 3749 47

Table 2
Evaluation of student learning via application of rubric to student memos.

Learning objectives Level of achievement as percentage of class


% Not % Minimally % % Highly
competent competent competent competent
Be able to verify completeness of a large data file (or series of data files) 0% 0% 36% 64%
Gain exposure to electronic testing of manual journal entries (MJEs). Be able to identify 2% 4% 31% 63%
large and MJEs. Tests include but are not limited to, MJEs booked late at night, on
weekends, holidays; round dollar amounts, Benfords Law, key word searches, reserve
trending, reversals, and backposting
Students should be able to articulate the process by which they would procure ledger data 0% 4% 24% 73%
and perform electronic testing of MJEs

Note: The feedback above was received from graduate students only.

specific journal entries that represent a heightened risk of fraud. Additional tests, such as asking for supporting documentation
and interviewing client personnel, are required in order to determine whether the red flags represent actual fraud.

2.4. Student assessment

Student learning was assessed by a variety of methods including (1) analysis of the final memo by an independent pro-
fessor using a grading rubric, (2) student performance on a related question incorporated into the midterm exam, and (3)
student perceptions of learning.4

2.4.1. Student memo


To evaluate learning of students in the graduate course, the professor evaluated the final deliverable for the project using
a rubric for assessment. Each student memo was evaluated on three dimensions with four possible levels ranging from 1-not
competent to 4-highly competent. Review of the results, presented in Table 2, reveals that 94100% of students were com-
petent or highly competent in all three dimensions measured.

2.4.2. Exam question


Ninety-two percent of graduate students correctly answered a data analysis question drawn from concepts in this case
that was incorporated into the midterm exam for the course.5 A copy of the exam question is available as part of the accom-
panying solution.

2.4.3. Student feedback


In the most recent semester of use, feedback was gathered from students at two large state universities. Upon completion,
47 students from a senior level audit class, and 55 students from a Masters level (MSA) Forensic Accounting course provided
anonymous feedback. Students overwhelmingly recommended use of the case at other universities with 100% of the Masters
level students and 90% of the undergraduate students recommending its use, respectively. Using a 5-point Likert scale (with
1-Strongly Disagree and 5-Strongly Agree), students expressed their agreement that the case achieved its learning objectives
by increasing their knowledge of data analytics (MSA mean 4.49, undergrad 3.65, combined 4.05), improved their ability to
test ledger files for large and unusual entries (MSA mean 4.61, undergrad 3.43, combined 3.99), and helped them understand
how to assess risks related to fraudulent journal entries (MSA mean 4.56, undergrad 3.37, combined 3.93) (see Table 3).6
Students appreciated the realistic nature of the assignment, the chance to see how accountants utilize Big Data, and the
problem-solving opportunities it presented. One student expressed his/her appreciation of the realistic nature of the case
saying, This assignment was a straightforward and interesting way to show how risks associated with journal entries might
be measured in real life. Another agreed that it was one of the first tangible assignments that I have done that I know I will
have to re-perform on the job, and another reported I think that this assignment was the most realistic case I had to do
within accounting. . .. It gave me something to talk about in my interviews.
Others reported their favorite aspect was the opportunity to see how accountants conduct data-driven investigations on a
daily basis. The case taught me how [to] break down large data and use analytical tools to analyze it the data and draw
conclusions from it. Another student agrees, I would highly recommend this assignment as it pushes students to think

4
The final memo and midterm exam questions were only utilized in the MSA course.
5
The exam question was not used in the undergraduate course.
6
Scores from the MSA students are significantly higher than the undergraduate students across the board. The undergraduate class was accessing IDEA
software through a virtual computer lab being beta tested during the semester. The students experienced frequent and recurring technology issues generated
by the virtual lab that were unrelated to IDEA or the case but nonetheless prevented access to IDEA. The disruption generated significant frustration, impacting
students satisfaction with the case as evidenced by 47% of students mentioning their experience suffering through the technical difficulties with comments
such as This assignment could have been improved by having a more fluid feel due to technology issues, by no fault of the instructor. Or more pointedly, This
is a great project but due to the technology malfunctions it frustrated the students. . .. In my opinion, that took over the whole scope of the project because I was
more focused on the technology than the actual project because something was always messing up.
48 R. Fay, E.M. Negangard / Journal of Accounting Education 38 (2017) 3749

Table 3
Students case evaluations.

N = 87 Students Forensic Accounting Auditing Combined


(MSA) (undergrad)
n = 41 n = 46 n = 87
The case materials . . .
Increased my knowledge of data analytics 4.49 3.65 4.39
Improved my ability to test ledger files for large and unusual entries 4.61 3.43b 4.19
Helped me understand how to assess risks related to fraudulent journal entries 4.56 3.37 4.28
Will help me in my future job tasks 4.39 3.43 3.87
Were interesting 4.50 3.43 4.52
Were realistic 4.49 3.83 4.43
Included instructions that were clear 4.37 2.52b 4.23
Percent of students who recommended using the assignment at other universitiesa 100% 90% 95%
41/41 37/41(c) 78/82
a
When calculating the percentage of students who recommended using the case at other universities, we only included those students who responded to
the question. There were five students from the Auditing class that did not complete that question. If we consider the nonresponse students as a
recommendation against using the case, the recommendation rates for the Auditing and combined groups are 80% and 90%, respectively.
b
Scores from the MSA students are significantly higher than the undergraduate students across the board. The undergraduate class was accessing IDEA
software through a virtual computer lab being beta tested during the semester. The students experienced frequent and recurring technology issues
unrelated to IDEA or the case that nonetheless prevented access to IDEA. The disruption generated significant frustration, impacting students satisfaction
with the case as evidenced by 47% of students mentioning their experience suffering through the technical difficulties.

about how and why certain procedures are carried out. Learning how to use IDEA is just a plus. Others appreciated the ana-
lytic power the software provided. One student wrote, [my favorite part was] being able to understand the simplicity of the
database once it was learned. I love that you can perform a simple search and receive feedback instantly without manually
looking at each journal entry. Another agreed, the best part was being able to crunch out massive quantities of data.
Many students appreciated the opportunity for problem-solving and the sense of accomplishment after the case was
completed. One student wrote, I like accounting because of the problem-solving element, and this assignment was in line
with that thinking. Being given a broad set of information with a few directions and being able to dive in was my favorite
part. Another reported, It was definitely challenging, but it made me feel accomplished when I truly understood and com-
pleted the process.

2.5. Suggested solutions

Suggested solutions, including a step-by-step guide for completing the case and accompanying data files, are available by
contacting Dr. Eric Negangard emn2j@comm.virginia.edu.

Acknowledgements

Many thanks to John Reisch, anonymous reviewers, and participants of the 2016 Forensic Accounting Research Confer-
ence, the 2015 AAA Annual Meeting, and the 2016 ACFE Global Fraud Conference for their valuable feedback.

Appendix A. Supplementary material

Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jaccedu.
2016.12.004.

References

American Accounting Association (AAA) (2016). Accounting is Big Data webinar. April 11. Recording available at <http://commons.aaahq.org/posts/
df9f9f5426>.
AICPA (2003). Journal entries and other adjustments. AICPA Practice Alert 200302. New York, NY: AICPA.
American Institute of Certified Public Accountants (AICPA) (2016). Consideration of fraud in a financial statement audit.AU-C Section 240. New York, NY:
AICPA.
Applebaum, D. & Nehmer, R. (2016). Using drones in internal and external audits: an exploratory framework. In Presentation at the 38th world continuous
auditing & reporting symposium. Newark, NJ. November 5. Available at <http://raw.rutgers.edu/38wcars/appelbaum_nehmer.html>.
Bauguess, S. (2016). Has Big Data made us lazy? Speech to the Midwest Region Meeting American Accounting Association (AAA), Chicago, Illinois. October
21. Available at <https://www.sec.gov/news/speech/bauguess-american-accounting-association-102116.html>.
Benford, F. (1938). The law of anomalous numbers. Proceedings of the American Philosophical Society.
Beresford, D. R., Katzenbach, N. D., & Rogers, C. B. Jr., (2003). Report of investigation by the special investigative committee of the board of directors of WorldCom,
Inc. Washington, D.C.: U. S. Securities and Exchange Commission. Available at <https://www.sec.gov/Archives/edgar/data/723527/
000093176303001862/dex991.htm>.
Best, J. (2016). IBM Watson: The inside story of how the Jeopardy-winning supercomputer was born, and what it wants to do next. Tech Republic. Available
at <http://www.techrepublic.com/article/ibm-watson-the-inside-story-of-how-the-jeopardy-winning-supercomputer-was-born-and-what-it-wants-
to-do-next/>.
R. Fay, E.M. Negangard / Journal of Accounting Education 38 (2017) 3749 49

Bloom, B. S. (1956). Taxonomy of educational objectives, handbook I: the cognitive domain. New York: David McKay Co Inc.
Brown-Liburd, H., Issa, H., & Lombardi, D. (2015). Behavioral implications of Big Data on audit judgment and decision making and future research directions.
Accounting Horizons, 29(2), 451468.
Cao, M., Chychyla, R., & Stewart, T. (2015). Big Data analytics in financial statement audits. Accounting Horizons, 29(2), 423429.
Carmichael, D. R. (2010). Double-entry, nonstandard entries, and fraud. The CPA Journal, 80(1), 6265.
Center for Audit Quality (CAQ) (2008). Practice aid for testing journal entries and other adjustments pursuant to AU Section 316. Washington, DC. Available
at <http://www.thecaq.org/docs/practice-aid/caqjetestingpracticeaid12082008.pdf?sfvrsn=2>.
Chartered Global Management Accountant (CGMA) (2014). CGMA briefing Big Data: Readying business for the Big Data revolution Available at <http://
www.cgma.org/Resources/Reports/DownloadableDocuments/CGMA-briefing-big-data.pdf>.
Claburn, T. (2016). IBM Watson: Machine-of-all-trades. Information Week. May 16. Available at <http://www.informationweek.com/big-data/big-data-
analytics/ibm-watson-machine-of-all-trades/d/d-id/1325483>.
Cokins, G. (2014). Top 7 Trends in management accounting, Part 2. Strategic Finance, 95(7), 4147.
Data, data everywhere (2010). The Economist. February 25. Available at <http://www.economist.com/node/15557443>.
DeVries, D. D., & Kiger, J. E. (2004). Journal entries and adjustments - your biggest fraud danger. Journal of Corporate Accounting & Finance, 15(4), 5762.
Fay, R. & Negangard, E. (2016). Searching Enrons emails for clues: Performing electronic document review and analysis. Working paper.
Gray, G. L., & Debreceny, R. S. (2010). Data mining journal entries for fraud detection: An exploratory study. International Journal of Accounting Information
Systems, 11(3), 157181.
IBM (2016). Language translation Available at <http://www.ibm.com/smarterplanet/us/en/ibmwatson/developercloud/language-translation.html>.
Lee, D. (2016). KPMG recruits IBM Watson for cognitive tech audits, insights. Accounting Today. March 8. Available at <http://www.accountingtoday.com/
accounting-technology/news/kpmg-partners-with-ibm-watson-for-cognitive-tech-audits-insights-77431-1.html>.
Loraas, T. M., & Searcy, D. L. (2010). Using queries to automate journal tests: Agile Machinery Group, Inc., Issues. Accounting Education, 25(1), 155174.
Maxwell, S., Steiner, D., & Stein, S. (2013). Leveraging Big Data in the modern tax function. August 14. Available at <http://www.accountingweb.com/
technology/accounting-software/leveraging-big-data-in-the-modern-tax-function>.
McKinsey Global Institute (2011). Big Data: The next frontier for innovation, competition, and productivity Available at <http://www.mckinsey.com/
business-functions/business-technology/our-insights/big-data-the-next-frontier-for-innovation>.
Murphy. M. L. (2014). How to prepare for auditing in a digital world of Big Data. Journal of Accountancy. October 16. Available at <http://www.
journalofaccountancy.com/news/2014/oct/201411104.html>.
Murphy, M., & Tysiac, K. (2015). Data analytics helps auditors gain deep insight. Journal of Accountancy, 219(4), 5458.
Nigrini, M. J. (1999). Ive got your number. Journal of Accountancy, 187(5), 79.
Novack, J. (2013). How SECs new RoboCop profiles companies for accounting fraud. Forbes. August 9. Available at <http://www.forbes.com/
sites/janetnovack/2013/08/09/how-secs-new-robocop-profiles-companies-for-accounting-fraud/#7090f45a3226>.
Public Company Accounting Oversight Board (PCAOB) (2016). Consideration of fraud in a financial statement audit: Auditing standard 2401. Washington, D.C.:
PCAOB.
PCAOB (2015a). Inspection observations related to PCAOB risk assessment Auditing Standards (No. 8 through No.15): PCAOB Release No. 2015-007.
Washington, D.C.: PCAOB.
PCAOB (2015b). Report on 2014 inspection of Deloitte & Touche LLP: PCAOB Release No. 104-2015-095. Washington, D.C.: PCAOB.
PCAOB (2015c). Report on 2014 inspection of Ernst & Young LLP: PCAOB Release No. 104-2015-121. Washington, D.C.: PCAOB.
PCAOB (2015d). Report on 2014 inspection of PricewaterhouseCoopers LLP: PCAOB Release No. 104-2015-122. Washington, D.C.: PCAOB.
PCAOB (2015e). Report on 2014 inspection of KPMG LLP: PCAOB Release No. 104-2015-189. Washington, D.C.: PCAOB.
Ramlukan, R. (2015). How Big Data and analytics are transforming the audit. Ernst & Young Reporting 9: 10-13. Available at <http://www.ey.com/
Publication/vwLUAssets/ey-reporting-issue-9-interactive/$FILE/ey-reporting-issue-9-interactive.pdf>.
Smith, S. (2015). 3 ways accountants can use Big Data to fight fraud. Accounting Web. June. Available at <http://www.accountingweb.com/practice/
practice-excellence/3-ways-accountants-can-use-big-data-to-fight-fraud>.
Wallace, W. (2000). Reporting practices: Potential lessons from Cendant Corporation. European Management Journal, 18(3), 328333.
Weld, L. G., Bergevin, P. M., & Magrath, L. (2004). Anatomy of a financial fraud: A forensic examination of HealthSouth. The CPA Journal, 74, 4449.