STAT500 Assignment Questions

This assignment will follow the statistical enquiry cycle discussed in week 2. Your analysis should bewritten as a report (2-4 pages of A4 is about right) and should follow the structure outlined in questions1 – 6 below.

Question 1. Problem 

(Total for Question 1: 10 marks)

Consult the list of variables provided in the dataset and choose a topic to explore (if your dataset iscalled a, typing names(a) on the command line tells you the variable names you have available; inRstudio, you can see the details of a dataset which you have read in the ‘environment’ pane).You should choose a broad overall topic, and then write two specific questions to investigate. You should also provide a brief explanation (about 100-150 words) about why you think these questionsare interesting.Two examples of topics and questions are provided below (do not choose these examples!). The questions must be able to be answered using data in the dataset

Example 1:

Topic: What factors impact literacy?

Two specific questions:

• Do countries with a higher GDP also have a higher literacy rate?
• Does this relationship differ by geographic region?

Example 2:

Topic: How do CO2emissions vary globally?

Two specific questions:

• Do countries with higher CO2emissions also have large agricultural sectors?
• Do CO2emissions vary by geographic region?

Question 2. Plan 

(Total for Question 2: 5 marks)

What variables will be required to answer your questions? Provide a brief explanation (in your ownwords) about each of your chosen variables. You are encouraged to refer to the World Bank websitefor details about variables.

If you need to compute an additional variable in order to answer one ofyour questions, the code used to do this should be provided.

Question 3. Data 

(Total for Question 3: 5 marks)

Read the dataset into R and extract the data that you will need for this analysis. You can extract thecolumns by referring to them by name or by number. Provide the code that you use to do this.

Example:

#Read CSV file
allglobal <- read.csv(file.choose())

#Extract chosen columns using names
global <- allglobal[, c(“Country”, “Region”, “GDP”, “TaxRevenue”)]

#Extract chosen columns using column numbers
global <- allglobal[, c(1, 3, 12, 30)]

#Inspect data
head(global)

Question 4. Analysis 

(Total for Question 4: 45 marks)

Explore your two questions using appropriate summary statistics and graphs (15 marks per question).

The following structure is recommended for each question:

• Clearly state the question being analysed (e.g. with a heading)
• Summary statistics and graphs. In the example above, you might use R idiom such as
  –mean(global$GDP) or maybe median(global$GDP) (why the difference)?
  –hist(global$GDP)
  –boxplot(GDP~Region, data=global) . . . although in this case you might be better offusing logarithmic axes, boxplot(GDP~Region, data=global, log=’y’)
• 50-150 words explaining the key messages of your summary statistics and graphs.

Across the analysis of the two questions, you should include at least 3 different types of graphsand 4 different summary statistics. Marks will only be awarded if the graphs, summary statisticsand explanations are correct and appropriate.

Question 5. Conclusion. 

(Total for Question 5: 15 marks)

Write one or two paragraphs (approximately 100-200 words) to summarise your findings. You shouldanswer your question by referring to the appropriate analysis conducted in part 4. You should alsomention further analysis that could be conducted.

Question 6. Critique 

(Total for Question 6: 10 marks)

Choose two of the graphs that you used in part 4. For each graph, explain why that type of graph wasappropriate for the analysis that you conducted (approximately 100 – 150 words total).

Question 7. Presentation and Grammar 

(Total for Question 7: 10 marks)

Your report should be professionally presented and should be free of spelling and grammar errors,and all sources should be clearly referenced. Therefore, 10 marks will be allocated based on theoverall presentation of your assignment.