CR
IC
O
S
Pr
ov
id
er
C
od
e
00
09
8G
Week 10 Tutorial
QUESTION 1 35 MARKS
Novaria is a country with 500,000 people located in the southern hemisphere, known for its
picturesque landscapes and thriving tourism industry. With abundant natural wonders, such as
pristine beaches, lush forests, and majestic mountains, Novaria attracts millions of visitors
annually. The tourism industry plays an important role in the country's economy, providing jobs
and generating revenue. The currency of Novaria is the Novarian Dollar (NOD).
The tax office, called the Novarian Revenue Agency (NRA), collects taxes and enforces tax
laws in Novaria. The NRA uses a computerized system called the Taxpayer Risk Assessment
System (TRAS) to select tax returns for audit. The TRAS assigns a score to each tax return
based on various factors, such as the income reported, the number of deductions claimed,
and the taxpayer's compliance history. The scores range from 0-100. A higher score indicates
that the return is more likely to have a discrepancy between the true dollar amount owed to the
NRA and the claimed amount owed to the NRA. A sufficiently high score triggers an audit by
an NRA agent.
Despite the thriving tourism industry in Novaria, there is a large wealth divide between the
owners of hotels, restaurants, casinos, and tour companies and the majority of citizens who
are employed at these establishments. This could potentially impact the tax system, as the
income reported by these establishments could vary greatly, and the TRAS may need to
consider this when selecting tax returns for audit. The following is the % breakdown of the
population by earnings (90% of households are classified as lower-income households and
10% of the population is classified as higher-income households):
There are concerns that lower-income taxpayers are more likely to make errors on their tax
returns due to limited access to resources to help them prepare their returns accurately. Also,
low-income taxpayers may be more likely to claim certain tax credits or deductions, such as
0
10
20
30
40
50
0 - 100 100 -
200
200 -
300
300 -
400
400 -
500
500 -
600
600 -
700
700 -
800
800 -
900
900 -
1000
% of Households in Each Income Bracket
(NOD)
Comm 1190: Data, Insights and Decisions. Module 5: Data Communication
the Earned Income Tax Credit, which are complex to calculate and verify. Thus, there is a
higher likelihood of claiming these credits or deductions could make them more likely targets
for audits.
A recent study by economists at the University of Novaria provided evidence that TRAS was
more likely to audit the poorest households in Novaria. The figures show that the lowest-
income wage earners, defined as those earning under 300 NOD were audited at a rate of 13
per 1,000 returns in 2022. For everyone else, the rate was 2.6 per 1,000 returns.
The head of the NRA, Taylor, has asked you to lead a team to devise a new model - TRAS II -
for determining audits. The key constraint is that a maximum of 32,000 people can be audited
annually. Your team of researchers presented the following two objectives for the TRAS II
model for you to present to Taylor.
• Objective 1: A model that maximizes revenue collection for the government.
• Objective 2: A model that maximizes the likelihood of identifying individuals most likely
to commit tax fraud (regardless of the size of the discrepancy between what is owed
and what is claimed).
Part A [max 500 words] (25 Marks)
Taylor wants you to evaluate the alternative objectives, by comparing the objectives using an
ethical decision framework.
i. Before making an ethical decision. It is important to obtain facts. Identify three pieces of
information that would help you determine which options would produce the best or do
the least harm.
ii. Which option is more likely to better treats people equally or proportionally?
iii. Which option best serves the community, and not just some members?
Part B [max 200 words] (10 Marks)
Taylor has recently read about the philosopher John Taurek who famously argued that in
many types of ethical cases, numbers don’t count, and perhaps the best way to decide
whether to audit any individual is to determine randomly. Taylor proposes that a fair coin is
flipped four times for each person, and, if the results are four heads, the individual is audited.
The probability of four heads is 1/16, so 31,250 people, which is less than the maximum of
32,000 will be audited. Identify the potential benefits of the coin flip approach and the potential
drawbacks of the approach to recommend whether the ethical principles underpinning this
method of selecting audits are appropriate.
Comm 1190: Data, Insights and Decisions. Module 5: Data Communication
QUESTION 2 30 MARKS
Cookie Cats is a popular puzzle game app. As players progress through the levels of the
game, they occasionally encounter gates that force them to wait a non-trivial amount of time or
make an in-app purchase to progress. In addition to driving in-app purchases, these gates
serve the important purpose of giving players an enforced break from playing the game,
hopefully resulting in the player's enjoyment of the game being increased and prolonged. The
question to be explored is whether the placement of the entry-level gate matters. The
hypothesis being investigated is whether moving the entry-level gate from 30 to 40 will
increase player activity (and hopefully in-game purchases).
Data was collected from an A/B test where a total of 90,198 players were randomly assigned
to a treatment group (entered at gate level 40, g_40=1) or a control group (entered at gate
level 30, g_40=0). The output variables of interest included in the data are:
ret_1 = 1 if the player plays within one day of installing the app; and = 0 otherwise.
games = total number of games played in the first 14 days of installing the app.
Note: For this question, the 97.5th percentile of a standard normal distribution is 1.96.
Part A. [max 100 words] (5 Marks)
Before formally investigating the hypothesis being tested within the experiment, good practice
dictates that an analyst determines the key features of the data. See below for selected
summary statistics and visual representations. What are your main conclusions about the key
variables based on this initial analysis?
Table 1: Summary statistics
Variable Mean Std. Deviation Minimum Maximum
g_40 0.504 0.500 0 1
games 51.9 195.1 0 49854
ret_1 0.445 0.497 0 1
Part B. [max 100 words] (5 Marks)
0
.0
2
.0
4
.0
6
D
en
si
ty
0 50 100 150 200
games
Games distribution: Top 5% trimmed
Comm 1190: Data, Insights and Decisions. Module 5: Data Communication
Do you have enough information to test whether random assignment has been successful? If
not, how would you check?
Part C. [max 100 words] (5 Marks)
Now consider the regression of ret_1 on g_40. The results are given below. What do they
indicate? (You can assume random assignment to the two groups has been successful.) Test
the null hypothesis of no treatment effect of using gate 40 as the entry point rather than gate
30.
_1� = 0.448 (4.0)− 0.0059_40(0.0033)
= 90,189, 2 = .000, ( . )
Part D. [max 100 words] (5 Marks)
Given the following regression results test the null hypothesis of no treatment effect using
games as the outcome of interest. Results from a second version of this regression are run but
only for those in the sample who played less than 3000 games. Is this a sensible regression to
run? Why? What do you conclude from these results?
� = 52.45 (0.92)− 1.16_40(1.30)
= 90,189, 2 = .000, ( . )
� = 51.34 (0.49)− 0.046_40(0.68)
= 90,188, 2 = .000, ( . )
Part E. [max 200 words] (10 Marks)
Summarise your results to the management of the game developer Tactile. Given the primary
question posed by the developer, recommend to management whether the placement of the
entry level gate has a causal impact on in-game purchases.
Comm 1190: Data, Insights and Decisions. Module 5: Data Communication
QUESTION 3 40 MARKS
Imagine you work for a large department store, which highly values customer service. The
following chart shows how customers contact the customer service centres.
You begin to discuss the chart with your manager. Immediately, she has the following queries:
“I want to see the overall trends, but it is difficult to see with all the seasonal spikes in the time
series. I’d like a simpler view into the trend.” You decide to create some charts to address your
manager’s queries.
Part A. [max 100 words] (5 Marks)
Using the two-questions four charts frameworks typology, identify the type of chart you would
use to address the query and explain why.
Part B. [max 200 words] (20 Marks)
Sketch two alternative charts for the query. For each chart, provide a brief explanation of your
design choices. To sketch the chart, you can use any tool you want (e.g., you can use a
software tool like info gram, excel, or R). Alternatively, you can sketch the chart using pencils,
pens, or markers on paper, then take a picture of the charts and paste them into your solutions
document. You can access the underlying data “customer_service.xlsx” on Ed.
Part C. [max 200 words] (10 Marks)
Evaluate your two charts and explain which you would select to further develop to present to
your manager.
Part D. [max 100 words] (5 Marks)
Beyond this new chart, your managers also want to present the above chart to a other
stakeholders. Evaluate potential design issues with this graph.
— END OF EXAMINATION PAPER —