ECON1203 经济与统计

UNSW Business School
ECON1203 Business Economics and Statistics

Project Overview
The project aims to enhance your career-focused learning experience by bringing real-world
scenarios and a real business problem into the classroom, creating a safe space for you to
explore, collaborate and make changes.
The assignment is intended to promote problem-based learning (PBL), in which you learn about
a subject by working in teams to solve real-life problems. It is also intended to develop your skills
in research, critical thinking and problem-solving, your data analysis and Excel skills, and your
ability to present your ideas and solutions concisely and coherently.
Solving real-life problems is an inherently complex and messy process, but such a process also
offers plenty of learning opportunities. You will learn about working through problems
persistently, seeking creative solutions, and being comfortable changing solution paths where
necessary.
In this sandboxed assignment (see Sandbox Education Program), you will have an opportunity
to solve a real-world problem and receive feedback from the problem owner (i.e. our project
partner). Your experience in this project will be helpful in your transition into the professional
environment – you will be prepared to leverage your existing knowledge and skills while at the
same time identifying and acting on knowledge and skill gaps, responding to new challenges
and seizing emerging opportunities coming your way.
Project Brief
Business Problem:
How can the Career Accelerator team increase Business School student
participation rates in the Microsoft Excel Certification?

Industry Partner/Problem Owner: Career Accelerator @ UNSW Business School

1. Background: Career Accelerator @ UNSW Business School
The Career Accelerator @ UNSW Business School is a specialised team that provides a suite
of opportunities and experiences designed to help UNSW Business School students build

their professional skills and improve student employability and career readiness across all
undergraduate, postgraduate and MBA programs.
Career Accelerator provides students with a diverse range of curricular, co-curricular and
extra-curricular offerings ranging from internships, global opportunities, mentoring programs,
industry events and networking opportunities, PASS classes, learning consults, and a suite of
technology-driven tools and resources.
2. What is the Excel Certification Program?
The Microsoft Office Excel Certification Program is a free and exclusive offering to UNSW
Business School students. Since launching in 2018 as a co-curricular opportunity, the
program has expanded. It is now embedded in key UNSW Business School programs (e.g.
Bachelor of Commerce and its combined degrees), providing students with the opportunity
to build their technical skills for data analysis, interpretation and presentation and gain an
industry-recognised digital credential that can be shared on their LinkedIn profile.
There are two parts to the program:
1. The Excel Training Program (ETP): This provides students access to online skills
modules, resources and practice exams that are self-paced and allow students to develop
their skills. For COMM1110 and ECON1203 courses, the training program is mandatory
(e.g. in COMM1110, the Excel Training Program is your Assessment 1), and students are
required to complete the practice exams at two levels - Associate level and Expert level.
Students are graded on their Practice Exam results for their course assessment.
2. The Excel Certification Test (ECT): on completion of the course assessment, students
are encouraged (but not required) to take the optional next step to complete the official
certification test, to receive their digital credential and be considered fully qualified.
By completing the course assignment, students have undertaken the majority of the work
required. The official certification step involves booking into a 50-minute online
invigilated test and achieving a pass rate of 70% to obtain the certification.
The key difference between the official ECT and the Practice Exams (i.e. Assessment 1a
and 1b students completed in their ETP in the course) is that the ECT is invigilated by a
test proctor. Apart from this, the ECT has the same exam length (50 minutes), question
types and difficulty levels as the Practice Exams students completed in the ETP as part
of their assessment. This means that students who completed their Assessment 1a and
1b and achieved a score over 700 (out of 1,000) would pass the ECT and receive the
official industry-recognised Excel Certificate if they choose to sit in the ECT. However, the
student participation rates in the ECT have been very low (see section 4 below).
Students must take the ECT within the calendar year that they start the ETP in the course
as their access code (i.e. the one received in Week 1 to access GMetrix for your
Assessment 1) expires at the end of that year.

The ECT is available to any enrolled student who signs up for the ETP as part of their
course (e.g. all students in ECON1203) or as a co-curricular activity. The ECTs are
scheduled regularly throughout each term as well as during term breaks, and students
are able to select their preferred timeslot through a dedicated Excel Certification Moodle
site: https://moodle.telt.unsw.edu.au/course/view.php?id=58401 (sign in using self-
enrolment key: excel_student).
3. Benefits of the Excel Certification Program
Increasingly, employers are looking for graduates with strong technical skills, and the
Microsoft Excel Certification Program provides students with the opportunity to develop
these essential skills and demonstrate their competencies to future employers. On
successful completion (i.e. completion of the ECT), it is a verified skill that students can add
to their CV. They also receive an industry-recognised Microsoft digital credential that can be
shared on their LinkedIn profile to demonstrate their Excel competency to potential
employers.
An added benefit is the Excel Certification test earns students experience points towards the
COMM1999/COMM3999 requirements of their program. Points are converted to BCoins,
which can be redeemed for UNSW Business School merchandise.
The Microsoft Excel Certification Program is offered to UNSW Business School students free
of charge, saving them the current fee of approx. $140 if they were to enrol in the program
independently.
4. The Business Problem
There is a significant drop off rate between students completing the Excel Training Program
(ETP) as part of their course assessment or co-curricular activity and those who take the
optional next step to complete the Excel Certification Test (ECT) to receive the official
certification. In 2021, almost 3,000 students completed the training component of the
program (i.e. the ETP), but only 20% took the next step to complete the ECT to get officially
certified.
UNSW Business School recognises the value in the certification that not only allows students
to develop key technical skills, but also enables them to demonstrate proven competency at
an industry level, ultimately enhancing their employability skills and employer demand for
their qualifications. In support of this initiative, approximately $50,000 is allocated to the
program every year, but with current participation rates, this is a low return on investment.
The licences that are issued to students are valid only for one calendar year, so students must
complete both the training (i.e. ETP, which students already completed in their courses, like
ECON1203, as part of their assessment) and the certification test (i.e. ECT, which they need
to self enrol via the Excel Moodle site1) within that year. The low participation rate results in
unused licences that cannot be carried over to the following year. It also means that if

students fail to certify within the calendar year but choose to do so later, they will be liable for
the cost to undertake certification independently.
5. The Task
We would like you to help the Career Accelerator team understand student behaviours with
the Excel Program, identify and evaluate the potential barriers to students taking the optional
next step of official certification (i.e. take the ECT - Excel Certification Test) and provide any
recommendations on how the Career Accelerator team can improve the ECT participation
rates – ideally increasing from 20% to 80%.

Note about the data
These data are real-world data provided by Career Accelerator. The data has been anonymised
to remove any personal identifiers. Please do NOT share this data with anyone without written
approval from the course authority.
Access Data:
Your Assessment data can be accessed via the R-Shiny App at the following site:
>>> https://unswteaching.shinyapps.io/ECON1203_ShinyApplets/ <<<
Click the link above and follow the steps below to obtain and download your personalised
Assessment dataset:

(1) Click on “Assessment data”.
(2) Enter your student ID without the "z" to load your assessment data. Click “Load
Assessment Data” to access your assessment data.
(3) To download your data, click "Download Data".
*You can use the R-Shiny App to perform preliminary analysis to explore the key
features of the data. However, you are required to use Excel to analyse the full data
(see Note below).
Video Guide: R-Shiny App Overview Video This is a general video guide introducing you to the
R-Shiny App (NOT the assessment data or your Assessment this term). The dataset used in
this example video is different from the one that you are required to use for your Assessment
(so please ignore any references to data or assessment requirements in the video, and focus
on the use of the R-Shiny App itself). Some of the topics discussed in the video (e.g.
hypothesis test) will be covered in the course in later weeks.
Note: While the R-Shiny App allows you to perform some quick analysis to understand the
data, it is restricted to 50 observations. For Assessment 2, your dataset contains 100
observations, which means that you are required to download the data from the R-Shiny App
and perform your final analysis in Excel.
IMPORTANT: the R-Shiny app selects a unique sample of 100 observations for each student.
This means that the results of identical analyses will vary across students depending on their
zID numbers. While this provides markers with a plagiarism check, this is not the primary reason

for providing the data in this form. Instead, it is an opportunity for different students to discuss
common modelling issues without necessarily coming to the same conclusion.
The data set contains 100 observations that were collected over a one-year period (over three
teaching terms) during 2021. Each observation refers to a different student in 2021. The
variables that have been selected for your use are:
Variable Name Description
Term The teaching term that the student
undertook the course (2021_T1 = 2021 Term
1, 2021_T2 = 2021 Term 2, 2021_T3 = 2021
Term 3)
Term_code The teaching term that the student undertook
the course (2021_T1 = 1, 2021_T2 = 2,
2021_T3 = 3)
Num_of_Attempts_Associate_Test Excel Training Program (ETP) - Associate
level - Number of Practice Exam 2 attempts a
student completed before the Assessment 1a
deadline
Score_1st_Attempt_Associate_Test

Excel Training Program (ETP) - Associate
level - Practice Exam 2 test score of the first
attempt (out of 1,000)
Score_Best_Attempt_Associate_Test

Excel Training Program (ETP) - Associate
level - Practice Exam 2 test score of the best
attempt (out of 1,000), i.e. the highest test
score obtained
1st_Attempt_Before_Deadline_Associate_Test Excel Training Program (ETP) - Associate
level - Practice Exam 2 first attempt
completion time before the assessment
deadline, measured in hours (e.g. if this
variable = 48, it means that the student
completed the first Associate Level Practice
Exam 2 attempt 48 hours, i.e. 2 days, before
the Assessment 1a deadline)*
last_Attempt_Before_Deadline_Associate_Test

Excel Training Program (ETP) - Associate
level - Practice Exam 2 last attempt
completion time before the assessment
deadline, measured in hours (e.g. if this
variable = 2, it means that the student
completed the last Associate Level Practice
Exam 2 attempt 2 hours before the
Assessment 1a deadline)*
Num_of_Attempts_Expert_Test

Excel Training Program (ETP) - Expert level -
Number of Practice Exam 2 attempts a

student completed before the Assessment 1b
deadline
Score_1st_Attempt_Expert_Test

Excel Training Program (ETP) - Expert level -
Practice Exam 2 test score of the first
attempt (out of 1,000)
Score_Best_Attempt_Expert_Test Excel Training Program (ETP) - Assessment
1b (Expert level) - Practice Exam 2 test score
of the best attempt (out of 1,000) – i.e. the
highest test score obtained
1st_Attempt_Before_Deadline_Expert_Test

Excel Training Program (ETP) - Expert level -
Practice Exam 2 first attempt completion
time before the assessment deadline,
measured in hours (e.g. if this variable = 24,
it means that the student completed the first
Expert Level Practice Exam 2 attempt 24
hours, i.e. 1 day, before the Assessment 1b
deadline)*
last_Attempt_Before_Deadline_Expert_Test

Excel Training Program (ETP) - Assessment
1b (Expert level) - Practice Exam 2 last
attempt completion time before the
assessment deadline, measured in hours
(e.g. if the variable = 1, it means that the
student completed the last Expert Level
Practice Exam 2 attempt 1 hour before the
Assessment 1b deadline)*
Male_Dummy Student gender (0=Female, 1=Male)
Age Student age (in 2021)
Local_Student_Dummy Domestic (=1) or international student(=0)
Participate_Cert_Exam A binary variable (i.e. a variable that takes
only the value 1 or 0) that indicates whether
a student participated in the Excel
Certification Test (ECT) (1=participated,
0=not participated)
*If this variable = 0, it means that the student completed his/her attempt just before (i.e., less
than 30 mins) the assessment deadline.
If a student only had one attempt, the first and last attempt data would be the same: e.g., if
Num_of_Attempts_Associate_Test = 1, Score_1st_Attempt_Associate_Test will have the same
value as Score_Best_Attempt_Associate_Test, and
1st_Attempt_Before_Deadline_Associate_Test will have the same value as
last_Attempt_Before_Deadline_Associate_Test

UNSW Business School
ECON1203 Business Economics and Statistics
Final report
The purpose of the final report is for you to synthesize your findings from the short answer
question and to produce report which attempts to address the needs of the client. In this
case this is to improve participation rates for the excel certification test (i.e. variable
Participate_Cert_Exam).
To break this analysis into digestible parts first they are interested identifying behvioural
traits which could affect these probabilities. As a starting point, these include (for both
tests):
Number of attempts (A1a_Num_of_Attempts, A1b_Num_of_Attempts);
Score of the attempts(A1a_Score_1st_Attempt, A1a_Score_Best_Attempt,
A1b_Score_1st_Attempt, A1b_Score_Best_Attempt); and
Completion of attempts hours before deadline (A1a_1st_Attempt_Before_Deadline,
A1a_last_Attempt_Before_Deadline, A1b_1st_Attempt_Before_Deadline,
A1b_last_Attempt_Before_Deadline);
Further details of these variables can be found in “Note about the data”.
To understand whether these factors play a important role, broadly speaking, you will need
to (i) formulate a multiple linear regression model (refer to week explaining your choice of
variables) and (ii) explain whether the relationship is statistically and/or economically
significant. Aforementioned, the use of a multiple linear regression, confidence intervals and
hypothesis testing would help you address this and thus provide evidence for your
arguments.
More specifically, you need to consider relevant factors to understand the relationship
between Participate_Cert_Exam and the behavioural variables incorporating any control
variables which are relevant.
In addition to interpreting the results of your analysis you will also need to draw to attention
issues of causality and confoundment which can impact the conclusions from the analysis.
As a part of this assessment, assumptions and limitations need to be explicitly identified.
e.g., What variable would you want to have in an ideal situation to measure different
variables in this analysis? Do you have this variable in the dataset? If not (which is often the
case in practice: we often don’t have all the ideal data/variables that we need to perform an
analysis, and have to rely on the data available to us), what variable in the dataset do you
have to use as an performance/ability measure? What are the assumptions and limitations
of using this variable?]

Once, you have considered the above issues and analysis, you will need to understand its
implications and consider the appropriate recommendations for the career accelerator team
as they try to improve participation in the certification exam. (as above) As this is the purpose
of this report.
Report structure
This report will consist of three parts:
Introduction – Which highlights the purpose of the report and how the report is
structured.
Body – Which describes the data and the analysis which is undertaken. As part of
this you will need to state the assumptions you are making around the sample and
the model. You will need to justify you model and acknowledge the limitations of your
model. Naturally, you are expected to run your model, report and interpret the results
of your model.
Conclusion – Here you summarise your results and its implications. What are the
limitations of your analysis and how they may be overcome? What additional data
you would like? What is your recommendation?
Below we will highlight some of the technical requirements in relation of the descriptive
statistics and modelling aspects in the body of your report.
Descriptive statistics
Like with your few short answer questions the first part of any analysis is to describe the
data you have. This could include a mix of graphs and summary statistics. Note that there
are a few variables which you need to consider, details of these variables can be found in
the data notes section above. Remember that the dependent variable of interest here is
“Participate_Cert_Exam”. Thus, in preparation of the modeling section you may want
to read about linear probability models in chapter 7 and chapter 8 as this will be
extremely useful for the modelling section.
Remember that you will be assessed on the presentation of the summary statistics as well
as any charts that your produce. As part of the reporting requirements, you are expected
to summarise the key features of the data including any interesting relationships.
Modelling participation
Next, you will need to run a multiple linear regression. In many ways, this will complement
the descriptive statistics that you have found above and to identify whether there are any
behavioural traits (see above) and other variables that can influence participation. As part
of this exercise, you need to:
Explain why a multiple linear regression is beneficial i.e., justify the need for multiple
linear regression and the issues associated with running a simple linear regression.
Contextualise it in the context of the current problems are there confounding factors
which motivate you to do this?
Associated with this think about the what type of model you’d like to use e.g. level-
level, log-level, log-log or level-log model for each of the considered variables. You
need to justify this. Given that this is a longitudinal dataset you may want to also
consider interaction variables and other non-linear aspects.
Choose your independent variables and justify. Remember that the dependent
variable is already chosen for you which is Participate_Cert_Exam.
Run the model and interpret.
As part of the reporting requirements for the multiple linear regression:
Interpret the coefficients.
Define and comment whether each of the coefficients are statistically significant.
Remember to state your assumptions.
Define and comment whether each of the coefficients are economically significant.
Remember to state your assumptions. You’ll also need to define what is economically
significant and use a benchmark to determine this.
Limitations and issues with your model. If you decide to use technical terms e.g.,
multicollinearity, homoskedasticity, bias, consistency etc. you need to explain what
these terminologies are and place them in the context of your problem and how
it will affect your results.
We will be paying attention to the presentation of the data and excel output as
well as whether you have used the full sample to conduct the analysis.
Recommendations
The regression above will allow you to better control the impact of an independent variable
on the dependent variable and provide correlations on these relationships. However, it
may not allow you to understand the mechanism which will improve participation which is
the core of the client’s problem.
As a first step you will need to summarise your results, its implications, and the limitations
of the analysis. Following this, you will need to provide some recommendations using
the results above and any additional literature which can improve participations.
Some exploration of the academic literature on how to improve participation may be worth
exploring.
Common reporting mistakes made by students (avoid these ?)
Regarding the descriptive statistics component it is important that you follow best reporting
practice. Many students tend to neglect things like decimal places, labels and font size at an
elementary level. For graphs, think about what is appropriate should you use a bar chart or
pie chart. How should I construct my histogram? What is the best practice here? We will be
paying attention to the presentation of the data, excel output as well as whether you
have used the full sample to conduct the analysis.

Regarding multiple linear regression. The common mistakes of made by your peers in the
past is the lack of justification of the functional form and the reason why a multiple linear
regression model is needed. Remember the ability to control for other factors is one thing
but it is very important for you to contextualise it in the context of your problem.
In the same vein, many of your peers in the past really did not justify why they used a pure
level-level model. Think about what a linear model means, do you expect the relationship
to be linear. You should construct an interaction variable; you will need to justify it (often
missed by the student) and interpret it (often misinterpreted). In fact, many students
tended to also mis-interpret the marginal impacts given the presence of the interaction
variables.
Interpretation is also a common issue especially when it comes to log models. Your peers
in the past have got this confused. So please pay attention to this. Finally, a very common
mistake was the discussion of economic significance. Often your peers say it without much
justification, you’ll need a benchmark to make your case more compelling e.g., using the
average of that variable as a benchmark may be a good starting point but we also know
what the issues are with averages. Another common mistake is that students often forget
what statistical significance means, it means that it is not zero, and often do not state this
nor do they recognize its implications. Remember that you can also perform other tests
beyond testing it as zero.
I urge you to please take note of all of this when you are writing up your report. If this is still
unfamiliar you should immediately see your demonstrator or lecture

WX：codehelp mailto: [email protected]