October 5

BSB123 Research Report

Background
Recent statistics 1 indicate that there are over 13 million credit cards in Australia as of April 2021. Consumers need
the cards for a variety of reasons and banks and other institutions are keen to provide these services because of the
revenues they generate. However the Royal Commission on Banking was scathing of the practices surrounding
credit cards and led ultimately to a number of changes around credit card rules – particularly in terms of assessing
eligibility and credit limits.
You have recently been hired as part of the data science team for a regional bank and have been tasked with
investigating your credit card system. The bank has provided you with information on 246 randomly selected
customers for which they have data on a range of relevant variables from the previous year.
The Royal Commission was particularly interested in the extent to which personal indebtedness had increased
through credit cards and other loans, and the social and economic disruption that this caused. From the banks
perspective, the ability to repay credit card debt is crucial for their own bottom line, but more importantly, they
recognise the social, health and welfare issues associated with credit card abuses. As a consequence you have been
asked to investigate those factors which may lead to excessive debt or inability to manage card repayments.
In the first instance you have been given information on 9 variables for each of the customers. In addition to their
customer number you have been provided:
1. Balance – the average monthly unpaid credit card balance ($)
2. Income – the average monthly household income from all sources ($)
3. Spend – average amount spent using the credit card per month ($)
4. Loans – the average amount paid servicing other loans per month ($)
5. Card – level of card issued (Blue for lowest level, Gold for mid tier and Platinum highest)
6. Gender – gender of the card holder restricted to Male or Female
7. Status – household income arrangements defined as couple or single income earner
8. Children – Number of children in home
9. Education – highest level of education of card owner (High School, Diploma, Bachelor or Postgraduate)
Your Task
Your role is to analyse the data as thoroughly as possible and provide any recommendations to the bank on issues
including what factors seem to have the biggest impact on credit card debt; to what extent is overall debt an issue
among your customers and are the correct type of cards being issued to customers. In considering these issues
please answer the following questions:
Question One
Your first task is to consider Average Monthly Credit Card Debt to get an understanding of the problems associated
with card debt. To do so:
(i) Calculate the full range of appropriate descriptive statistics and create appropriate graphs to describe
this data.

(ii) Identify the monthly unpaid debt associated with the top 10% of customers
(iii) Construct a 95% confidence interval for the average amount of unpaid credit card debt.
(iv) Using the information created in parts (i) to (iii) describe what you have learned about Average Monthly
Credit Card debt in no more than 120 words.

(10 Marks)

Question Two
In addition to credit card indebtedness, the coronavirus pandemic and recent inflationary pressures have meant that
overall indebtedness is becoming an issue and the bank is concerned that it may be a problem for them.
(i) Prior to this last year, 15% of all customers had been fully paying off their credit cards each month (i.e.
average credit card debt was zero(0)). The bank is concerned this has now fallen. Conduct a test to see
if the proportion of people fully paying off their cards has decreased.
(ii) The bank defines a customer in Debt Stress as a household whose total monthly debt – defined as
average unpaid credit card debt + other monthly loan payments – is more than one third of their average
monthly household income. Normally we would expect that those households with a lower average
monthly income would be more likely to be in stress. Conduct a test at the 1% level of significance to see
if the average of the income for customers in distress is less than that of customers not in distress. (Hint
– you will need to define which customers are in Stress and which are not and then work on the
averages of those two groups).
(iii) There is a view that Females are better at managing budgets and money than males. Conduct a test at a
5% level of significance to determine if the proportion of females under debt stress is less than that of
males.
(iv) Finally, the bank is also concerned that while the overall level of indebtedness has increased along with
inflation, household incomes have not been increasing to keep pace. Twelve months ago the average
monthly household income was $9350. Conduct a test at the 5% level of significance to see if average
monthly household income has increased.

(20 Marks)

Question Three
The bank is hoping to identify those factors which impact Credit Card non payment.
(i) Carry out a stepwise regression analysis to determine which factors impact the overall average monthly
unpaid credit card balance. In conducting the analysis, and as you build the model, make sure to explain
why you are taking each step until you have identified the best model.
(ii) Once you have identified the final model please interpret all necessary statistics, and show fully in full
detail all of the tests you would normally conduct as part of a regression analysis including testing any
assumptions.
(iii) Forecast the average monthly debt balance for a male customer with a postgraduate degree who has a
Gold Card, lives in a household with a single income of $10,000, spends on average $8000 per month on
their card, and services other loans of $1500 per month and has one child.
(iv) Comment on any concerns you have about your final model and / or next steps you would take in
building a better model.

(25 Marks)

Question Four
In no more than 200 words write a report in non technical language outlining what you have learned about
indebtedness among customers at this bank. Highlight any issues you believe the bank should consider in particular
and those that do not seem to be an issue for them.

============================================================================

25624 Financial Metrics for Decision Making
Friday 14th October 2022, 11.59pm

Netflix’s Latest Pivot

Netflix success is largely attributed to the company’s ability to pivot their business model. The first
pivot was when the company shifted their DVD home delivery service to becoming the leading online
streaming platform. More recently, the second pivot was in response to the inherent risk of being a
content distribution platform: content producers could refuse to deliver their content via Netflix
(Disney, Paramount, etc.). Netflix response was to become a content creation business themselves1
.
Both pivots were successfully landed, providing Netflix with a sustained increase in subscriptions. The
company’s reflexes are now being put to the test once again, this time as a more mature business.
In the first quarter of 2022, Netflix lost subscribers for the first time in 10 years. The drop came after
the growth experienced throughout the pandemic, gaining millions of users per quarter through 2021.

The decline in number of subscribers adds to the trouble the company faces because of the large
number of households sharing their accounts with families and friends. In response, Netflix began
experimenting with a feature that monitors whether someone accesses an account from outside the
home and offers the ability to raise the price by adding 'sub-accounts' for other people.
During the Q1 2022 earnings call, Netflix’s management unveiled that they were considering rolling
out an ad-supported tier to re-ignite subscription growth - especially in UCAN (USA and Canada). The
market feels much more penetrated given the COVID pulled-forward bump2

. Instead of paying the full
subscription, Netflix will offer the possibility of paying less in exchange for watching ads, thus

1 Misra, A. 2020. Netflix – Constantly Pivoting its Business Model to Success. Sep 13, 2020.
https://thestrategystory.com/2020/09/13/netflix-pivoting-business-model/
2 Wells Fargo, 2022. Equity research. NFLX: The AVOD Devil Is in the Detail. Jun 3, 2022.
https://wellsfargo.bluematrix.com/docs/pdf/d5ef4344-9429-4eb6-bdda-382f2a44ad57.pdf

eliminating one of the great advantages of the streaming service over traditional television. The
timeline to introduce this new subscription tier has been brought forward a couple of times.

Netflix is the most expensive streaming service, compared to rivals from giants like Disney and
Amazon, especially if we want the full experience with the best video and sound quality. So-called
'sub-accounts' and new ad-supported subscriptions will become the cheapest way to use Netflix3
.
In addition to these measures, Netflix executives want to increase their presence in markets outside
the United States, where growth is stronger, producing more local content and improving the quality
of its series.
The company faces several potential ways forward:
• Remaining a pure Subscription video-on-demand (SVOD) player
• Introducing a second subscription tier with the Advertising video-on-demand (AVOD) model
• Opening sub-accounts to monetise the large segment of password-sharing users
In this context, it is worth evaluating the potential impacts each combination of products will have on
the company’s revenue stream. Please use the Excel workbook provided to do a quarterly revenue

3 Translated by Content Engine LLC. (2022). Netflix loses users for the first time in a decade and plans to
introduce advertising. CE Noticias Financieras.
https://www.proquest.com/docview/2653220902?parentSessionId=jwgbxZhsHdcKlLloJSTvkPJOu7bpZkXvzVVl
WN5YZPk%3D&pq-origsite=primo&accountid=17095

forecast for Netflix from September 2022 to June 2025 under each scenario described in the
questions.
In each scenario, you can search for data to support your assumptions. This Wells Fargo Equity
Research report could be of great use.

Read the report

Please make you to reference your sources and to form your own views and not to simply “copy and
paste” from reports.
QUESTION 1. DISCUSSION ON FORECASTING TECHNICS [4 marks]
Before you build your own Revenue forecasts, please watch these two videos created by a tutor who
used Netflix to teach forecasting using:
• Holt’s Method https://youtu.be/55_vmSIba3A, and
• Regression model https://youtu.be/089eZCuBhtg
In the written report, use bullet points to:
• highlight the main limitations in using these technics
• propose amendments to the methods to make them more appropriate to forecast Netflix’s
revenue
• This question does not have a spreadsheet component.

QUESTION 2. FORECAST REVENUE | PURE SVOD PLAYER [6 marks]
• Assuming Netflix decides to add maintain their current business model as is, please use the
“Question 2” section in the Revenue worksheet to do a quarterly revenue forecast for
Netflix from September 2022 to June 2025.
• Include any inputs and assumptions in the Inputs worksheet
• This question does not have a written report component.

QUESTION 3. FORECAST REVENUE | SVOD and AVOD [8 marks]

• Assuming Netflix decides to add a second subscription tier with the Advertising video-on-
demand (AVOD) model, please use the “Question 3” section in the Revenue worksheet to do

a quarterly revenue forecast for Netflix from September 2022 to June 2025.
• Include any inputs and assumptions in the Inputs worksheet.
• Make sure you at least account for:
o cannibalisation across products,
o new potential customers and
o different attrition rates by product
o advertising revenue
• This question does not have a written report component.
QUESTION 4. FORECAST REVENUE | SVOD, AVOD AND SUB-ACCOUNTS [8 marks]
• Assuming Netflix decides to:
o add a second subscription tier with the Advertising video-on-demand (AVOD) model,
o add sub-accounts for “sharing-password” customers
please use the “Question 4” section in the Revenue worksheet to do a quarterly
revenue forecast for Netflix from September 2022 to June 2025.

• Include any inputs and assumptions in the Inputs worksheet.
• Make sure you at least account for:
o cannibalisation across products,
o new potential customers and
o different attrition rates by product
o advertising revenue
• This question does not have a written report component.

QUESTION 5. OUTPUTS [6 marks]
• In the Outputs worksheet, please use tables and charts to report the key results of your
analysis.
• Please include any sensitivity analysis you judge relevant
• This question does not have a written report component.
QUESTION 6. DISCUSS YOUR FINDINGS [6 marks]
In the written report, use bullet points to:
• Discuss your main findings
• Propose a way forward to the Netflix management

QUESTION 7. FINANCIAL MODELLING BEST PRACTICES [2 marks]
The use of financial modelling best practices in the revenue model, inputs and outputs will be
marked in this question.

==========================================================================

PSYC3020

1. Define/ operationalise what skill/ ability/ trait the proposed test measures (e.g., post-
concussion cognitive ability; clinical skills; driving ability). Identify up to 3-4 tests/

underlying variables that are used to measure this skill/ ability/ trait. Word count limits the
number of tests/ underlying variables you include in the proposal.
2. State how the proposed test is novel. Does it address problems of existing tests? Does it
apply an existing test to a new population? Is it brand new? Why should the proposed
test be funded?
3. Provide a rough description of the form of the proposed test.
4. How will two reliability and two validity strategies be used to evaluate the proposed test?
Depending on your proposal idea, your tutor may also follow up with additional questions that
are related to the points provided below [refer to Section 3 and the Appendix of the briefing for
more details]:
5. Why is it important to measure your chosen variable(s)? What is the rationale?
Remember, it needs to be of real benefit to society.
6. Who is your population of interest? Industry? Age?
7. How will your variable(s) be measured? Self-report? Behavioural observation? How will
the test be administered? How is it scored? HOW does WHO do WHAT, WHEN and
WHY? Refer to the Assignment Primer 2 – Concussion Activity tutorial for inspiration.
8. What practical and/ or ethical considerations of your proposed test are important? How
are they addressed?
9. Is a dependent/ outcome variable required? What about a contrast group?

TOPIC 2 (hardest of the prescribed topics; least support): Design and validate a new test
or new battery of tests to select the best applicants for entry into a postgraduate training
program for clinical psychology.

Background information references:
Fauber, R. L. (2006). Graduate admissions in clinical psychology: Observations on the present
and thoughts on the future. Clinical Psychology: Science and Practice, 13(3), 227-234.
https://doi.org/10.1111/j.1468-2850.2006.00029.x
Johnson, W. B., & Campbell, C. D. (2002). Character and fitness requirements for professional
psychologists: Are there any? Professional Psychology: Research and Practice, 33(1),
46-53. https://doi.org/10.1037/0735-7028.33.1.46
Johnson, W. B., & Campbell, C. D. (2004). Character and fitness requirements for professional
psychologists: Training directors' perspectives. Professional Psychology: Research and
Practice, 35(4), 405-411. https://doi.org/10.1037/0735-7028.35.4.405

Other potentially interesting references:
Carr, S. E. (2009). Emotional intelligence in medical students: Does it correlate with selection
measures? Medical Education, 43(11), 1069-1077. https://doi.org/10.1111/j.1365-
2923.2009.03496.x
Kelly, E. L., Goldberg, L. R., Fiske, D. W., & Kilkowski, J. M. (1978). Twenty-five years later: A
follow-up of the graduate students in clinical psychology assessed in the VA Selection
Research Project. American Psychologist, 33(8), 745-755. https://doi.org/10.1037/0003-
066X.33.8.746
Pope-Davis, D. B., Reynolds, A. L., Dings, J. G., & Nielson, D. (1995). Examining multicultural
counseling competencies of graduate students in psychology. Professional Psychology:
Research and Practice, 26(3), 322-329. https://doi.org/10.1037/0735-7028.26.3.322
Rem, R. J., Oren, E. M., & Childrey, G. (1987). Selection of graduate students in clinical
psychology: Use of cutoff scores and interviews. Professional Psychology: Research and
Practice, 18(5), 485-488. https://doi.org/10.1037/0735-7028.18.5.485

TOPIC 3: Design and validate a new test or new battery of tests to assess the
competence of radiologists when screening (i.e., interpreting) mammograms for
tumours.
Background information references:
Barlow, W. E., Chi, C., Carney, P. A., Taplin, S. H., D'Orsi, C., Cutter, G., Hendrick, R. E., &
Elmore, J. G. (2004). Accuracy of screening mammography interpretation by
characteristics of radiologists. Journal of the National Cancer Institute, 96(24), 1840-
1850. https://doi.org/10.1093/jnci/djh333
Beam, C. A., Layde, P. M., & Sullivan, D. C. (1996). Variability in the interpretation of screening
mammograms by US radiologists: Findings from a national sample. Archives of Internal
Medicine, 156(2), 209-213. https://doi.org/10.1001/archinte.1996.00440020119016
Elmore, J. G., Jackson, S. L., Abraham, L., Miglioretti, D. L., Carney, P. A., Geller, B. M.,
Yankaskas, B. C., Kerlikowske, K., Onega, T., Rosenberg, R. D., Sickles, E. A., & Buist,
D. S. M. (2009). Variability in interpretive performance at screening mammography and
radiologists' characteristics associated with accuracy. Radiology, 253(3), 641-651.
https://doi.org/10.1148/radiol.2533082308
Nodine, C. F., Kundel, H. L., Lauver, S. C., & Toto, L. C. (1996). Nature of expertise in
searching mammograms for breast masses. Academic Radiology, 3(12), 1000-1006.
https://doi.org/10.1016/S1076-6332(96)80032-8
Other potentially interesting references:
Carney, P. A., Sickles, E. A., Monsees, B. S., Bassett, L. W., Brenner, R. J., Feig, S. A., Smith,
R. A., Rosenberg, R. D., Bogart, T. A., Browning, S., Barry, J. W., Kelly, M. M., Tran, K.
A., & Miglioretti, D. L. (2010). Identifying minimally acceptable interpretive performance
criteria for screening mammography. Radiology, 255(2), 354-361.
https://doi.org/10.1148/radiol.10091636
Elmore, J. G., Wells, C. K., Lee, C. H., Howard, D. H., & Feinstein, A. R. (1994). Variability in
radiologists interpretations of mammograms. New England Journal of Medicine, 331(22),
1493-1499. https://doi.org/10.1056/NEJM199412013312206
Goddard, C. C., Gilbert, R. J., Needham, G., & Deans, H. E. (1998). Routine receiver operating
characteristic analysis in mammography as a measure of radiologists' performance.
British Journal of Radiology, 71(850), 1012-1017.
https://doi.org/10.1259/bjr.71.850.10211059
Miglioretti, D. L., Gard, C. C., Carney, P. A., Onega, T. L., Buist, D. S. M., Sickles, E. A.,
Kerlikowske, K., Rosenberg, R. D., Yankaskas, B. C., Geller, B. M., & Elmore, J. G.
(2009). When radiologists perform best: The learning curve in screening mammogram
interpretation. Radiology, 253(3), 632-640. https://doi.org/10.1148/radiol.2533090070

TOPIC 4 (easiest; most support): Design and validate a new test or new battery of tests to
assess whether older drivers are safe to continue to drive. (You may want to narrow the
scope of this topic to a specific context e.g., a new test that is suitable for a busy
doctor’s surgery or for a Queensland Transport office.)

Background information references:
Hill, A., Horswill, M. S., Whiting, J., & Watson, M. O. (2019). Computer-based hazard perception
test scores are associated with the frequency of heavy braking in everyday driving.
Accident Analysis & Prevention, 122, 207-214. https://doi.org/10.1016/j.aap.2018.08.030
Horswill, M. S. (2016a). Hazard perception in driving. Current Directions in Psychological
Science, 25(6), 425-430. https://doi.org/10.1177/0963721416663186
Horswill, M. S. (2017). Hazard perception tests. In D. L. Fisher, J. K. Caird, W. J. Horrey & L. M.
Trick (Eds.), Handbook of Teen and Novice Drivers: Research, Practice, Policy, and
Directions (pp. 439-450). CRC Press. https://doi.org/10.1201/9781315374123
Horswill, M. S., Anstey, K. J., Hatherly, C. G., & Wood, J. (2010). The crash involvement of
older drivers is associated with their hazard perception latencies. Journal of the
International Neuropsychological Society, 16(5), 939-944.
https://doi.org/10.1017/S135561771000055X
Horswill, M. S., Hill, A., & Wetton, M. (2015). Can a video-based hazard perception test used for
driver licensing predict crash involvement? Accident Analysis & Prevention, 82, 213-219.
https://doi.org/10.1016/j.aap.2015.05.019
Morgan, R., & King, D. (1995). The older driver - A review. Postgraduate Medical Journal,
71(839), 525-528. http://dx.doi.org/10.1136/pgmj.71.839.525
Wetton, M. A., Hill, A., & Horswill, M. S. (2011). The development and validation of a hazard
perception test for use in driver licensing. Accident Analysis and Prevention, 43(5), 1759-
1770. https://doi.org/10.1016/j.aap.2011.04.007
Wetton, M. A., Horswill, M. S., Hatherly, C., Wood, J. M., Pachana, N. A., & Anstey, K. J.
(2010). The development and validation of two complementary measures of drivers'
hazard perception ability. Accident Analysis and Prevention, 42(4), 1232-1239.
https://doi.org/10.1016/j.aap.2010.01.017

Other potentially interesting references:
George, S., Clark, M., & Crotty, M. (2008). Validation of the Visual Recognition Slide Test with
stroke: A component of the New South Wales occupational therapy off-road driver
rehabilitation program. Australian Occupational Therapy Journal, 55(3), 172-179.
https://doi.org/10.1111/j.1440-1630.2007.00699.x
Horswill, M. S., Anstey, K. J., Hatherly, C. G., & Wood, J. M. (2010). The crash involvement of
older drivers is associated with their hazard perception latencies. Journal of the
International Neuropsychological Society, 16(5), 939-944.
https://doi.org/10.1017/S135561771000055X
Mallon, K., & Wood, J. M. (2004). Occupational therapy assessment of open-road driving
performance: Validity of directed and self-directed navigational instructional components.
American Journal of Occupational Therapy, 58(3), 279-286.
https://doi.org/10.5014/ajot.58.3.279

O'Connor, M. G., Kapust, L. R., & Hollis, A. M. (2008). DriveWise: An interdisciplinary hospital-
based driving assessment program. Gerontology and Geriatrics Education, 29(4), 351-

362. https://doi.org/10.1080/02701960802497894
Unsworth, C. A., Pallant, J. F., Russell, K. J., Germano, C., & Odell, M. (2010). Validation of a
test of road law and road craft knowledge with older or functionally impaired drivers.
American Journal of Occupational Therapy, 64(2), 306-315.
https://doi.org/10.5014/ajot.64.2.306