Accounting Information Systems, 15e, Global Edition (Romney)

Chapter 5 Introduction to Data Analytics in Accounting

1 Explain what makes a good question and evaluate questions relative to the SMART
Framework.

1) Which of the following is the first step to having an analytics mindset?

A) asking the right questions
B) extracting, transforming, and loading relevant data
C) applying appropriate data analytic techniques
D) interpreting and sharing results with stakeholders
Answer: A
Concept: Asking SMART questions
Objective: Learning Objective 1
Difficulty: Easy
AACSB: Analytical Thinking

2) If a question reflects the objectives of the organization or the situation under consideration, it
fulfills which of the following objectives?
A) Specific
B) Measurable
C) Achievable
D) Relevant
E) Timely
Answer: D
Concept: Asking SMART questions
Objective: Learning Objective 1
Difficulty: Easy
AACSB: Analytical Thinking

3) If a question is answerable and the answer should cause a decision maker to take action, it
fulfills which of the following objectives?
A) Specific
B) Measurable
C) Achievable
D) Relevant
E) Timely
Answer: C
Concept: Asking SMART questions
Objective: Learning Objective 1
Difficulty: Easy
AACSB: Analytical Thinking

1
Copyright © 2021 Pearson Education Ltd.
4) If a question is amenable to data analysis because the inputs to answering the question are able
to be assessed with data, it fulfills which of the following objectives?
A) Specific
B) Measurable
C) Achievable
D) Relevant
E) Timely
Answer: B
Concept: Asking SMART questions
Objective: Learning Objective 1
Difficulty: Easy
AACSB: Analytical Thinking

5) If a question is direct and focused to produce a meaningful answer, it fulfills which of the
following objectives?
A) Specific
B) Measurable
C) Achievable
D) Relevant
E) Timely
Answer: A
Concept: Asking SMART questions
Objective: Learning Objective 1
Difficulty: Easy
AACSB: Analytical Thinking

6) If a question has a defined time horizon for answering, it fulfills which of the following
objectives?
A) Specific
B) Measurable
C) Achievable
D) Relevant
E) Timely
Answer: E
Concept: Asking SMART questions
Objective: Learning Objective 1
Difficulty: Easy
AACSB: Analytical Thinking

2
Copyright © 2021 Pearson Education Ltd.
7) For the following question, identify which SMART objective is most likely violated. An
auditor asks a client "Will you please get me the relevant data for the last year?"
A) Specific
B) Measurable
C) Achievable
D) Relevant
E) Timely
Answer: A
Concept: Asking SMART questions
Objective: Learning Objective 1
Difficulty: Moderate
AACSB: Reflective Thinking

8) A company upgraded production machinery to capture data about the fit tolerance for
assembled parts of each unit produced. Before the upgrade, inspectors created the same data by
sampling 10 percent of units produced. The upgrade represents an increase in
A) data variety.
B) data velocity.
C) data veracity.
D) data volume.
Answer: D
Concept: Asking SMART questions
Objective: Learning Objective 1
Difficulty: Moderate
AACSB: Reflective Thinking

9) A company added a quality assessment station to a production line that re-measured the fit
tolerances of units produced. Fit tolerances collected by the assembly machine are stored in
conjunction with fit tolerances of the assessment station. The new assessment station represents
an increase in
A) data variety.
B) data velocity.
C) data veracity.
D) data volume.
Answer: C
Concept: Asking SMART questions
Objective: Learning Objective 1
Difficulty: Moderate
AACSB: Reflective Thinking

3
Copyright © 2021 Pearson Education Ltd.
10) A company upgraded production machinery to transmit data to the data warehouse about the
fit tolerance for assembled parts of each unit produced as the measures are taken. Before the
upgrade, data was transmitted at the end of a production run. The upgrade represents an increase
in
A) data variety.
B) data velocity.
C) data veracity.
D) data volume.
Answer: B
Concept: Asking SMART questions
Objective: Learning Objective 1
Difficulty: Moderate
AACSB: Reflective Thinking

11) A company upgraded production machinery to collect data about the fit tolerance for
assembled parts of each unit produced along with environmental data such as machine
temperature, parts temperature, ambient air temperature, humidity, and atmospheric pressure.
Before the upgrade, only fit tolerance data was collected. The upgrade represents an increase in
A) data variety.
B) data velocity.
C) data veracity.
D) data validity.
Answer: A
Concept: Asking SMART questions
Objective: Learning Objective 1
Difficulty: Moderate
AACSB: Reflective Thinking

12) While some argue that more data is always better, others might present a competing
hypothesis that there can be too much data. Which arguments/artifacts might support the idea
that the organization is overwhelmed with data?
A) Dark data
B) Data swamp
C) Processing slowdowns
D) All of the above
Answer: D
Concept: Asking SMART questions
Objective: Learning Objective 1
Difficulty: Moderate
AACSB: Reflective Thinking

4
Copyright © 2021 Pearson Education Ltd.
2 Describe the extract, transform, and load (ETL) process and key components of each step of
the process.

1) A real estate investment company keeps a list of the locations of its properties by measuring
the latitude and longitude of each of its properties. This is an example of what type of data?
A) Semi-structured data
B) Structured data
C) Unstructured data
D) None of the above
Answer: B
Concept: ETL process
Objective: Learning Objective 2
Difficulty: Easy
AACSB: Analytical Thinking

2) A bank installs security cameras and keeps a digital copy of all footage for 2 years. The digital
copy of the videos is an example of what type of data?
A) Semi-structured data
B) Structured data
C) Unstructured data
D) None of the above
Answer: C
Concept: ETL process
Objective: Learning Objective 2
Difficulty: Easy
AACSB: Analytical Thinking

3) A company stores a copy of all e-mails sent or received by the organization. The copies of e-
mail are an example of what type of data?
A) Semi-structured data
B) Structured data
C) Unstructured data
D) None of the above
Answer: C
Concept: ETL process
Objective: Learning Objective 2
Difficulty: Easy
AACSB: Analytical Thinking

5
Copyright © 2021 Pearson Education Ltd.
4) For most companies, what data structure typically holds the most to the least amount of data?
A) Data lake, data mart, data warehouse
B) Data lake, data warehouse, data mart
C) Data warehouse, data lake, data mart
D) Data warehouse, data mart, data lake
Answer: D
Concept: ETL process
Objective: Learning Objective 2
Difficulty: Easy
AACSB: Analytical Thinking

5) A company wants to store information about inventory including quantities, prices, cost,
photos, and social media reviews. What type of data repository should be used?
A) Data lake
B) Data warehouse
C) Data mart
D) Data swamp
Answer: A
Concept: ETL process
Objective: Learning Objective 2
Difficulty: Easy
AACSB: Analytical Thinking

6) A company purchases another company. The acquired company has not kept detailed records
of its data. The company from the acquired company is best described as which of the following?
A) Dark data
B) Data mart
C) Unstructured data
D) Data swamp
Answer: D
Concept: ETL process
Objective: Learning Objective 2
Difficulty: Easy
AACSB: Analytical Thinking

7) Which of the following characters would be worst to use as a delimiter (the possible delimiter
is listed between the quotes)?
A) ","
B) "|"
C) "&"
D) "e"
Answer: D
Concept: ETL process
Objective: Learning Objective 2
Difficulty: Easy
AACSB: Analytical Thinking

6
Copyright © 2021 Pearson Education Ltd.
8) Which of the following characters would be best to use as a delimiter (the possible delimiter is
listed between the quotes)?
A) " " (a space)
B) "X"
C) "~"
D) "."
Answer: C
Concept: ETL process
Objective: Learning Objective 2
Difficulty: Easy
AACSB: Analytical Thinking

9) A key element in planning a data transformation is

A) understanding the desired structure of the data.
B) understanding the quality of the question.
C) planning to remove all unstructured data.
D) incorporating dark data.
Answer: A
Concept: ETL process
Objective: Learning Objective 2
Difficulty: Easy
AACSB: Analytical Thinking

10) Which of the following represents a challenge to transforming/combining two sets of data?
A) Different levels of aggregation
B) Different identifiers
C) Differences in field concatenation
D) All of the above
Answer: D
Concept: ETL process
Objective: Learning Objective 2
Difficulty: Easy
AACSB: Analytical Thinking

7
Copyright © 2021 Pearson Education Ltd.
11) When combining data from four data sources, you notice that timestamp data about
transactions are represented differently in each of the data sources. The timestamp for one
transaction is presented in each source respectively as follows: "Fri, 21 Jan 2022 [Link] UTC",
"Friday, January 21, 2022 [Link] AM", "1642748529", "44582.2931597222". The first source
uses UTC, the second uses the local timezone, the third uses the Epoch timestamp, and the fourth
uses the Excel timestamp. All represent that same exact time. How would you determine which
timestamp to output in your transformation?
A) Output all four variants
B) Refer to the data requirements
C) Use the Excel variant as it is the most universal
D) Use the UCT variant as it is not sensitive to time zone
E) Use the Epoch variant as it is common in programming
Answer: B
Concept: ETL process
Objective: Learning Objective 2
Difficulty: Easy
AACSB: Analytical Thinking

12) For the following data, identify what are the text qualifiers and delimiters. If a text qualifier
or delimiter is not used, answer "none."

Flat file, Header Row:

ID-PayeeName-PayeeCompanyCheckNumber-CheckAmnt-ImageLocation
Flat File Data Row:
8335-'Anna-Marie Thomas'-Mountain Vista Corporation-1341-$4,712.10-'c:checkimages\check-
1341'

Answer: The text qualifiers are ' ' and the delimiter is the - character
Concept: ETL process
Objective: Learning Objective 2
Difficulty: Moderate
AACSB: Reflective Thinking

13) For the following data, identify what are the text qualifiers and delimiters. If a text qualifier
or delimiter is not used, answer "none."

Flat file, Header Row:

DocumentNum;BankName BankAddress;DocumentDescription
Flat File Data Row:
331-12;"First Bank" "429 Taylor St. Haines City, FL 33844";"Loan Agreement"

Answer: The text qualifiers are quotes, " " and there are two delimiters—the semicolon ";" and a
space. The space is used between bank name and bank address.
Concept: ETL process
Objective: Learning Objective 2
Difficulty: Challenging
AACSB: Reflective Thinking
8
Copyright © 2021 Pearson Education Ltd.
14) Describe how the data received from one system must be transformed to be input into the
second system. List each of your steps in the order of what you would do to make changes. The
two examples below show the data coming from one system and the data prepared to be loaded
in the second system.

Data coming from first system

Flat file, Header Row:
ID,FName,LName,Address,CreditLimit
Flat File Data Rows:
101,Bobby,Jones,"356 SE. Morris Rd." "Marshfield, WI 54449","$10,000"
102,Sally,Fernandez," 529 Penn Ave." "Derry, NH 03038","$4,000.13"

Data ready to be loaded in second system

Flat file, Header Row:
ID,Name,StreetAddress,City,State,Zip,CreditLimit
Flat File Data Rows:
101,Bobby Jones,356 SE. Morris Rd.,Marshfield,WI,54449,10000
102,Sally Fernandez,529 Penn Ave.,Derry,NH,03038,4000.13

Answer: The steps that must take place are as follows (the ordering is not required)
1. The columns FName and LName need to be concatenated with a space between the FName
and LName fields. The new column should be labeled Name
2. The field Address must be parsed into four columns. The four new columns should be named
StreetAddress, City, State, and Zip. To do the parsing, the following should be done. The
StreetAddress should accept the all text between the text qualifiers (the quotes) up to the space
between the text qualifiers. The second column, City, should receive all text from the text
qualifier to the comma. The third column, State, should accept the next two letters. The fourth
column, Zip, should accept the last five numbers.
3. The field CreditLimit should be formatted so that the $ sign and the comma are removed.
Concept: ETL process
Objective: Learning Objective 2
Difficulty: Challenging
AACSB: Application of Knowledge

9
Copyright © 2021 Pearson Education Ltd.
15) List the four steps the book discussed for transforming data. Define what must be done in
each of the steps.
Answer: The four steps for transforming data and what must be done in each step are
1. Understand the data and the desired outcome. The user must understand the structure and
design of the incoming data and then know what the data needs to look like after the
transformation process.
2. Standardize, structure, and clean the data. The user must perform the necessary steps to
change the data from how it began the process to change it to match how the data must appear to
be loaded into the next application.
3. Validate data quality and verify data meets data requirements. Errors and mistakes can be
introduced into the data transformation process. The user should verify that the data is free of
errors and mistakes and that the data fulfills the objectives for which it was transformed.
4. Document the transformation process. The data dictionary should be updated so that future
users understand the properties of the data.
Concept: ETL process
Objective: Learning Objective 2
Difficulty: Challenging
AACSB: Reflective Thinking

3 Explain the differences between descriptive, diagnostic, predictive, and prescriptive analytics.
Understand the situations for which each type of analytic is appropriate.

1) An analytic that answers the question "what should be done?" is an example of which type of
analytic?
A) Descriptive analytic
B) Diagnostic analytic
C) Predictive analytic
D) Prescriptive analytic
Answer: D
Concept: Types of analytic analyses
Objective: Learning Objective 3
Difficulty: Easy
AACSB: Analytical Thinking

2) An analytic that answers the question "why did this happen?" is an example of which type of
analytic?
A) Descriptive analytic
B) Diagnostic analytic
C) Predictive analytic
D) Prescriptive analytic
Answer: B
Concept: Types of analytic analyses
Objective: Learning Objective 3
Difficulty: Easy
AACSB: Analytical Thinking

10
Copyright © 2021 Pearson Education Ltd.
3) An analytic that answers the question "what happened?" is an example of which type of
analytic?
A) Descriptive analytic
B) Diagnostic analytic
C) Predictive analytic
D) Prescriptive analytic
Answer: A
Concept: Types of analytic analyses
Objective: Learning Objective 3
Difficulty: Easy
AACSB: Analytical Thinking

4) An analytic that answers the question "what might happen in the future?" is an example of
which type of analytic?
A) Descriptive analytic
B) Diagnostic analytic
C) Predictive analytic
D) Prescriptive analytic
Answer: C
Concept: Types of analytic analyses
Objective: Learning Objective 3
Difficulty: Easy
AACSB: Analytical Thinking

5) If the controller asks an accounts payable clerk to determine if the weather causes delays in
collecting payments, what type of analytic would be computed?
A) Descriptive analytic
B) Diagnostic analytic
C) Predictive analytic
D) Prescriptive analytic
Answer: B
Concept: Types of analytic analyses
Objective: Learning Objective 3
Difficulty: Easy
AACSB: Reflective Thinking

6) If the CIO asks for a report of how many website visits come from IP addresses that are
known to be associated with website hacking, what type of analytic would be computed?
A) Descriptive analytic
B) Diagnostic analytic
C) Predictive analytic
D) Prescriptive analytic
Answer: A
Concept: Types of analytic analyses
Objective: Learning Objective 3
Difficulty: Easy
AACSB: Reflective Thinking
11
Copyright © 2021 Pearson Education Ltd.
7) If a sales clerk creates an RPA bot that automatically computes the size of an order and then
evaluates if this amount is acceptable as part of a credit check for a customer, this is an example
of what type of analytic?
A) Descriptive analytic
B) Diagnostic analytic
C) Predictive analytic
D) Prescriptive analytic
Answer: D
Concept: Types of analytic analyses
Objective: Learning Objective 3
Difficulty: Easy
AACSB: Reflective Thinking

8) If a data analyst uses machine learning to create an estimation of which suppliers are likely to
be late in providing goods for a manufacturing process, what type of analytic did the data analyst
create?
A) Descriptive analytic
B) Diagnostic analytic
C) Predictive analytic
D) Prescriptive analytic
Answer: C
Concept: Types of analytic analyses
Objective: Learning Objective 3
Difficulty: Easy
AACSB: Reflective Thinking

12
Copyright © 2021 Pearson Education Ltd.
9) For the following scenario, identify what type (i.e., descriptive, diagnostic, predictive, or
prescriptive) of data analytics are produced. A team of employees are working together to create
a presentation about the positives and negatives of purchasing another company. Each member
of the team was tasked with analyzing a different aspect of the potential acquisition. The first
team member has analyzed the financial performance of the target company. This team member
identified how well the company did in the past and identified key reasons why the target
company had performed so well. The second team member analyzed the key customers of the
target company and developed an analysis predicting how many of the customers would likely
continue if the target company was purchased. The third team member analyzed the accounting
information system of the target company to understand internal controls, processes, and how
long it took the company to compute important accounting information. Finally, the fourth team
member prepared the report and discussed how the different findings integrated into the final
recommendation of the team.
Answer:
• The first team member likely computed descriptive and diagnostic analytics. That is, the first
team member looked at what happened in the past and why it happened.
• The second team member likely computed predictive analytics because s/he was using past
data to predict the future.
• The third team member likely computed descriptive analytics because the analytics described
what was currently happening.
• The fourth team member did not compute any analytics, but compiled the other data into a
report. This is not a data analytic technique.
Concept: Types of analytic analyses
Objective: Learning Objective 3
Difficulty: Moderate
AACSB: Application of Knowledge

10) An analyst uses Excel to perform a regression on historic data. The resulting regression
equation is then used to estimate the number of clients who would default on their accounts.
Which type of analytic is represented in this estimate?
A) Descriptive analytic
B) Diagnostic analytic
C) Predictive analytic
D) Prescriptive analytic
Answer: C
Concept: Types of analytic analyses
Objective: Learning Objective 3
Difficulty: Easy
AACSB: Reflective Thinking

13
Copyright © 2021 Pearson Education Ltd.
11) An analyst uses Excel to perform a regression on historic data. The coefficient of the
regression is compared with the prior period's regression coefficient to determine if it has
changed. Which type of analytic is represented in this comparison?
A) Descriptive analytic
B) Diagnostic analytic
C) Predictive analytic
D) Prescriptive analytic
Answer: A
Concept: Types of analytic analyses
Objective: Learning Objective 3
Difficulty: Easy
AACSB: Reflective Thinking

12) An analyst uses Excel to perform a regression on historic data. The coefficient of the
regression is compared with the prior period's regression coefficient to determine if customer
credit score thresholds should be adjusted. Which type of analytic is represented in this
comparison?
A) Descriptive analytic
B) Diagnostic analytic
C) Predictive analytic
D) Prescriptive analytic
Answer: B
Concept: Types of analytic analyses
Objective: Learning Objective 3
Difficulty: Easy
AACSB: Reflective Thinking

4 List the principles that lead to high-quality data visualizations.

1) Which of the following is not an important principle for data visualizations to be effective?
A) Simplifying the presentation of data
B) Emphasizing what is important
C) Choosing the right type of visualization
D) Using a data dashboard
Answer: D
Concept: High-quality data visualizations
Objective: Learning Objective 4
Difficulty: Easy
AACSB: Analytical Thinking

14
Copyright © 2021 Pearson Education Ltd.
2) The process of translating complex data analyses into more easy to understand terms to enable
better decision making is an example of what?
A) Data storytelling
B) Data visualization
C) Data ethicality
D) Data analytics.
Answer: A
Concept: High-quality data visualizations
Objective: Learning Objective 4
Difficulty: Easy
AACSB: Analytical Thinking

3) Which of the following reasons is not a reason why interpreting data analysis results is
difficult?
A) Human have biases that might cause them to misinterpret data.
B) Humans use automation that increases the speed data is analyzed.
C) Humans often mistake correlation for causation.
D) All of the above are reasons why interpreting data analysis results is difficult.
Answer: B
Concept: High-quality data visualizations
Objective: Learning Objective 4
Difficulty: Easy
AACSB: Reflective Thinking

4) Which of the options below are important components of effective data storytelling?
A) Remembering the question that initiated the analytics process.
B) Consider the audience.
C) Use data visualizations.
D) All of the above.
Answer: D
Concept: High-quality data visualizations
Objective: Learning Objective 4
Difficulty: Easy
AACSB: Analytical Thinking

15
Copyright © 2021 Pearson Education Ltd.
5) Discuss how you could improve the following viz based on the four principles taught in the
chapter. If the viz does one of the principles well, indicate so.
• Choosing the right visualization.
• Simplifying the presentation of the viz.
• Emphasizing what is important.
• Representing the data ethically.

16
Copyright © 2021 Pearson Education Ltd.
Answer:
• Choosing the right visualization. The viz uses an appropriate chart type.
• Simplifying the presentation of the viz. The viz can be simplified in several different ways.
There is not a single answer for this question, but rather principles that can be used in different
ways. Examples include:
o Splitting the viz into two vizs.
o Using different colors so it is clear what values are the dollar amount and
what values are the number of transactions.
o Removing the x-axis label "Days Overdue" or removing the words "Days"
from each of the labels.
o Removing the y-axis label of Dollar Amount"
o Simplifying the y-axis labels to values remove the two decimal points.
o Removing the legend if two charts are used.
o Removing the axes and instead using data labels.
• Emphasizing what is important. There are various ways to emphasize what is important.

Depending on what is important, you can use these different techniques.

• Highlighting the 90+ Days category in a different color.
• Separating the graphs
• Using bold for different titles that are more important, or only labeling the data
that is most important.
• Representing the data ethically. The viz is not misleading, so it is ok on this dimension.

Here is an example of an improved visualization:

Concept: High-quality data visualizations

Objective: Learning Objective 4
Difficulty: Challenging
AACSB: Application of Knowledge

17
Copyright © 2021 Pearson Education Ltd.
6) An employee responsible for decreasing the number of defects in the parts the company
manufactures used the following table to produce a visualization showing her performance.
Identifying how the employee created a misleading graphic.

Data
Week Defective Parts
1 191
2 142
3 190
4 126
5 194
6 104
7 207
8 220
9 275
10 95

18
Copyright © 2021 Pearson Education Ltd.
Answer: The employee did several things to be deceptive including:

• Only plotting the weeks that make the graph show a decline over time.
• Cutting the y-axis at 90 rather than showing it go to zero. At week 10 the viz looks like it is
almost zero.
• Only labeling the first point rather than each point. Week 10 should have 95 defects, but it
looks like it is almost zero since most people assume the x-axis is zero unless told otherwise.

A more complete visualization that includes a trend line revealing the actual relation of time and
defect shows defects are increasing over time.

Concept: High-quality data visualizations

Objective: Learning Objective 4
Difficulty: Moderate
AACSB: Ethical Understanding and Reasoning

5 Describe how automation interacts with the analytics mindset and when data analytics is not
the right tool for making a decision.

1) Tasks that have which of the following attributes are the best candidates for automating except
A) Repetitive.
B) Rules-based.
C) Creative.
D) Time-consuming.
Answer: C
Concept: Automation
Objective: Learning Objective 5
Difficulty: Easy
AACSB: Analytical Thinking

19
Copyright © 2021 Pearson Education Ltd.
2) Select all that apply. What parts of the analytics mindset can automation help perform?
A) Ask the right questions.
B) Extract, transform, and load relevant data (ETL).
C) Apply appropriate data analytic techniques.
D) Interpret and share the results with stakeholders.
Answer: A, B, C, D
Concept: Automation
Objective: Learning Objective 5
Difficulty: Easy
AACSB: Reflective Thinking

3) The application of machines to automatically perform a task once performed by humans is

called what?
A) A bot
B) Automation
C) Data visualization
D) Descriptive analytic
Answer: B
Concept: Automation
Objective: Learning Objective 5
Difficulty: Easy
AACSB: Analytical Thinking

4) Select all that apply. When might data analytics not be appropriate for making a decision?
A) When data does not exist in a decision context
B) When data is dirty and requires effort to clean
C) When data suggest making an unethical decision
D) When data cannot be reliability measured
Answer: A, C, D
Concept: Automation
Objective: Learning Objective 5
Difficulty: Moderate
AACSB: Reflective Thinking

5) Software that can be automatically programmed to perform tasks across applications just like
humans do, and is often able to be programmed with click-and-drop interfaces is called what?
A) Robotic process automation (RPA)
B) Big data
C) Diagnostic analytics
D) Prescriptive analytics
Answer: A
Concept: Automation
Objective: Learning Objective 5
Difficulty: Easy
AACSB: Analytical Thinking

20
Copyright © 2021 Pearson Education Ltd.
6) What does RPA stand for?
A) Real-time process automation
B) Real-time professional automation
C) Robotic process automation
D) Robotic professional automation
Answer: C
Concept: Automation
Objective: Learning Objective 5
Difficulty: Easy
AACSB: Analytical Thinking

7) Select all that apply, which of the following represent risks to automating business processes?
A) Incorrect automation can change data to be incorrect.
B) Automation can automatically change to adapt to new circumstances.
C) Bots did not deviate from what they are programmed to do.
D) Automation can have a computer perform mundane tasks rather than a human.
Answer: A
Concept: Automation
Objective: Learning Objective 5
Difficulty: Easy
AACSB: Reflective Thinking

21
Copyright © 2021 Pearson Education Ltd.