ETF1100 Business Statistics Group Assignment – 2021S1

A. How it works

• This project will be undertaken in small groups and involve analysing a data set using the approaches to quantitative problems you have learnt in this unit.

• The final product will be a set of presentation slides which you will submit via Moodle by Friday 11pm on 14 May 2021 (the end of Week 10). A penalty of 10% per day applies for late submissions. The assignment should contain no more than 15 slides.

• The slides should show, in an interesting way, the main things you have done and learned in addressing the main question of the assignment. Usually, you would present these slides to your workshop members and tutor. However, given there is some uncertainty around COVID- 19, and we have students in Melbourne and overseas, you will not actually present the slides this semester.

• The assignment is structured such that it gives groups some freedom to explore the problem in ways they see fit. There is no single correct answer for the assignment. It is a research project and different groups will approach things differently. This is encouraged.

• Students have been put in groups based on their preferences. All group members will generally be enrolled in the same workshop. There will be around 5 members in each group (though some groups may have more or less members than this). You should get in touch with your group members, organise to meet regularly and share ideas and the workload. It can be challenging, but also rewarding, to work in groups. The assignment is aimed at building your groupwork experience as well as fostering contact with your peers in this unit. If you find you are not in a group, then contact us as soon as possible. Also, it is possible that some of your allocated group members may have dropped the unit, or you may find that your group members may not be entirely cooperative. If you email someone and do not get a response, then proceed with the remaining group members.

• All group members must contribute to the assignment. There will be an opportunity to give feedback on the contributions of other members of your group, and this feedback will be used when allocating an individual’s final mark for this assessment. If you do not participate fully in the assignment, and cooperate with your group members, then you should expect your grade to be adjusted downwards as a result.

• The project is worth 18% of your final mark and your grade will depend on the quality and content of your presentation slides as well as your participation in the assignment. See the marking guide at the end of this document.

• You will receive written feedback on the assignment. In addition, we will run a workshop where the tutor will talk through each of the assignments submitted by groups in the workshop. This will provide useful feedback to you. It will also to be helpful, I think, to see how others approached the exercise.

• The exam for this semester will mostly use the same dataset and explore similar and related issues.

B. The Topic

(i) The Context: Climate Change and C02 Emissions

In this project you will analyse some up-to-date data tracking CO2 (carbon dioxide) emissions of countries of the world over the last few centuries. CO2 emissions are an important part of the story of the human contribution to climate change. The idea is to use techniques that we have been learning about in this unit to study some patterns in CO2 emissions across time and across countries. The main

focus is on how economic and population growth affect CO2 emissions, and on trends across time and patterns across countries. Below we describe the data and the kinds of analysis we would like you to do. Once you complete the analysis you will prepare a presentation where you show what you have done. The presentation should include some background of the issues, why you are studying this topic, and what implications we can draw from your analysis.

(ii) The Data Your raw data can be downloaded into a spreadsheet file here: https://github.com/owid/co2-data (use the XLSX file) As well as the data file, there is important information at this link about how the different variables are defined. You will need to study this carefully, because it affects how you interpret the analysis you undertake below. Some technical notes:

• The data file is big file (~6MB), and we will not ask you to analyse all the data, as it covers more than 200 countries over hundreds of years, so has many many thousands of rows. Instead, each section of the analysis requires you to extract a part of the dataset. Take that subset of the data to a new file and do your analysis on that file. In the end, you should have an Excel file for each Part below.

• Often you do not have all the data you need for a particular analysis – perhaps not all the years are available, or all the countries. You need to just do your analysis for the dataset you have available (e.g., just a subset of the countries, or not using all the years). Sometimes this involves rearranging rows of the data, or just selecting specific rows or columns. Be very careful when you do these kinds of operations.

• Some of the “countries” are not actually Countries, like “Africa”, “Asia”, “World”, etc. These are totals across a group of countries (e.g., a continent). When you do the analysis in Part C, you will need to make sure you only include real countries, so check the list carefully.

C. The Tasks

Please read this part of the document carefully. Here we outline various tasks we would like you to do. They are in separate sections, so your group can share them among yourselves. In each case we give you some guidance about what to do, but there is also some freedom for you to choose what data to focus on, or how you will analyse it. The idea is to do these tasks and analysis, then work on a presentation that explains what you have done and discusses what it all means. Please note that the assignment should contain no more than 15 slides. This is a strict maximum. Also, make sure that the font is a reasonable size, so it is easily readable in presentation mode. A common minimum font size for presentation slides is 20pt. You can use a slightly smaller font for tables of data or figures. But it must be readily readable. What these constraints will mean is that you need to think very carefully about what you do and do not include in your slides. My advice is to divide your presentation up in 5 parts as follows: Part 1: Introduction (2 slides) Create a slide with the title of your project and the full names and student numbers of your group members. On your second slide outline the contents of the presentation. You may also want to outline the issues and broadly how you are going to address them. You could also (very briefly) summarize your conclusions.

Part 2: The World Situation Over Time (~4 slides) Take the data for the “country” called World. This is the total for the variables across all countries. You can get the World data by filtering the data and selecting just World in the Country column. Once you have the relevant rows, copy these to a new Excel file and analyse this data. Here are some suggestions for what you might do:

• Take a look at the most recent year of data, and look at total CO2 emissions, then at emissions from coal, gas, oil and other sources. Find a way to present the information about the different sources of CO2 emissions.

• Plot a time series chart of CO2 emissions across time and describe its features.

• Run a regression with CO2 emissions as the dependent variable, and a linear time trend as the independent variable. Can you think of a better way to capture the trend, as the linear trend over the whole time period does not look appropriate? Perhaps you can include both time and time squared in the model to capture the nonlinear relationship with time.

• Once you have a sensible time series model for the trend in CO2 emissions, use that model to predict what emissions will be in 2050 and 2100, and then compare these with the experts are saying needs to happen to CO2 emissions over the next few decades. Is there any sign of slowing of the trend in recent years, based on your data here?

• Run a regression with CO2 emissions for the World as the dependent variable, and a set of independent variables: World population, World GDP per capita (you will need to calculate this variable and use a subset of the data because these variables are missing in the early years). Think carefully about what you learn from this regression and include your analysis in your slides.

Part 3: Study a Particular Country (~4 slides) Select a country to study. Make sure it is one with plenty of data over several years on the variables we are interested in. Here are some suggestions for what you might do:

• Tell us a little about this country compared to the rest of the world: size, location, GDP, population, GDP per capita, CO2 emissions per capita, and its share of total CO2 emissions.

• Take a look at the most recent year of data, and look at total CO2 emissions, then at emissions from coal, gas, oil and other sources. Find a way to present the information about the different sources of CO2 emissions. How does this mix compare to what you found for the World (Part 2)?

• Plot a time series chart of CO2 emissions across time and describe its features.

• Run a regression with CO2 emissions as the dependent variable, and a linear time trend as the independent variable. Compare the model and time series graph for this country to the overall World situation (Part A). You may also want to include ‘time squared’ in your time series model if the series is nonlinear. Maybe it is worth looking at growth rates to help with this comparison: are emissions for your country growing faster or slower than the World growth? Are things different in more recent years?

• Run a regression with CO2 emissions as the dependent variable, and a set of independent variables: population, GDP per capita. Compare your model with the World model from Part 2.

Part 4: Cross Country Models as of 2016 (~4 slides) Create a dataset that has one row for each country, corresponding to the 2016 values for that country. [Why 2016? Because this is the most recent year for which a good number of countries have GDP data.] (Using filtering on the Year column is the easiest way to do this but make sure you drop non-countries like “Africa”. Note that it generally seems that if the variable “iso_code” is missing then it is not a country). Now you can perform some “cross country analysis”, as of 2016. Here are some suggestions for what you might do:

• Have a look at some of the characteristics of 2-3 interesting variables and compare these across countries. Be aware of appropriate standardisation in making such comparisons. You might consider things like the share of coal or oil in CO2 emissions (variations in how much countries depend on these resources), CO2 per unit of energy (comparing the types of energy countries use), etc.

• Run a series of regressions with the same set of independent variables: population, and GDP per capita (which you will need to calculate). But consider 3 different dependent variables:

o CO2 emissions o CO2 emissions per capita (you will need to calculate and perhaps rescale this variable) o CO2 emissions per GDP (you will need to calculate and perhaps rescale this variable)

Think carefully about what you learn from these models, and from the similarities and differences. Part 5: Summary and conclusions (~1 slide) Provide a summary and some brief conclusions regarding what you have done. This could also address some of the weaknesses of your analysis and/or the data and some uncertainties about your results. You may also want to discuss the implications of your findings for government and society.

D. Assessment Criteria There are two primary components as to how your presentation will be assessed: (1) content, (2) presentation. Both are worth 9 marks out of 18 marks. In addition, you will give feedback to your group members—this is the group participation aspect of the assessment criteria. We will adjust your grade based on the feedback of your groupmates. These various aspects of the assessment process are listed below:

(1). Content (9 marks)

• The statistical analysis was correctly implemented, a range of techniques were used, it was clearly explained, with valid interpretations.

• The statistical analysis undertaken is well linked to the overall conclusion/theme/message of the presentation.

(2). Presentation (9 marks)

• The presentation follows a logical flow, and clear and logical conclusions/themes/main points are evident.

• The results are interpreted correctly, and conclusions are well justified

(3). Group participation (used for mark adjustment if it appears a group member has not contributed based on the responses of other group members)

• You will receive an email using the TEAMMATES software system after you have completed the assignment, i.e. a couple of days after the due date. You will be asked if each member of your group contributed sufficiently to the group project. If a large portion of the group indicates that a person did not make a reasonable contribution, then that person’s mark may be adjusted downward.

• If your group experiences major difficulties working together on the project, please contact a member of teaching staff to discuss the situation.

Each of the two main criteria will be graded on a scale from 0 to 9. The following table includes the sorts of comments that we would associate with the respective grades.

Mark Description

0 to 4 Fail: The work has fallen below the minimum required standard. i. The work exhibits fundamental misunderstandings, errors and/or omissions.

ii. A major revision of the work is required and there is extensive room for improvement.

iii. The work shows an extremely limited appreciation and understanding for the issues examined and the methods which should be appropriately used.

iv. The work incorporates no or limited creativity and/or advanced/mature thinking.

5 Pass: The work has narrowly met the minimum required standard. i. The work exhibits a significant level of misunderstandings, errors and/or

omissions. ii. There are several areas in which there is room for improvement.

iii. The work shows a limited appreciation and understanding of the issues examined and the methods which should be appropriately used.

iv. The work incorporates a limited degree of creativity and/or advanced/mature thinking.

6 Credit: The work is of a satisfactory standard. i. The work exhibits minor misunderstandings, errors and/or omissions.

ii. There is some room for improvement. iii. The work shows a reasonable appreciation and understanding for the issues

examined and the methods which should be appropriately used. iv. The work incorporates some level of creativity and/or advanced/mature

thinking.

7 Distinction: The work is of a high standard. i. The work exhibits no meaningful misunderstandings, errors and/or omissions.

ii. There is some minor room for improvement. iii. The work shows a clear appreciation and understanding for the issues examined

and the methods which should be appropriately used. iv. The work incorporates creativity and/or advanced/mature thinking.

8 to 9 High Distinction: The work is of outstanding quality. i. The work exhibits no misunderstandings, errors and/or omissions.

ii. There is little or no room for improvement. iii. The work shows an insightful appreciation and understanding for the issues

examined and the methods which should be appropriately used. iv. The work incorporates a high degree of creativity and/or advanced/mature

thinking.

For any issues regarding your assignment, contact us via the following email address: [email protected].