Analysing Household Data
Statistics for Business and Finance
Have you been part of a national census? Privacy issues aside, a census provides lots of data that can inform a government policies and actions - but to be useful, the data needs to be analysed and interpreted.
In this assignment, we will use statistical methods to analyse and interpret real world demographic data.
The goal of this assignment is to
- Test your understanding of statistical methods and approaches
- Improve your ability to use Excel for manipulation of data (see here for some guides on using Excel)
- Understand the real-world applications and implications of statistics
To complete this assignment you must
- Complete a set of statistical analysis tasks on a unique data set (both tasks and data set will be provided to you)
- Submit in a report in word detailing your response to each task (the final answer and reasoning / calculations that led to it)
- Submit an excel document that contains your data set and the calculations you used to complete the tasks
Important:
- This is an individual Assignment worth 30%. Each student will use a unique data set to complete the assignment.
- You should use Excel for all your computational work.
- Please provide detailed calculations for all the tasks. Not explaining how you arrived at your conclusions may result in only partial marks being awarded for those tasks.
- Present your computer results in appropriate tabular form. Also include graphs where required to support your answers.
Don't do:
- Please do not copy your data set or assignment questions on the word files you are submitting since this will trigger plagiarism detection and might affect your submission. Submit the data set you use as a separate excel document
Do:
- Submit the data set on excel files along with the word document
- Demonstrate in the Word report that your manual calculations match the Excel computational work.
Instructions: Analysing Household Data
In this assignment you will be analysing and interpreting household data.
Step 1: Prepare Data Set
- You need to download and modify the generic data set to identify the sample you will be working on. More information on the data set is available below.
- Important: You do not need to analyse the entire data set - only the 100 samples randomly selected for you. This must be done first. The rest of your work will be based on this data set.
- You will need to know how to manipulate data in Excel. You can use any other compatible spreadsheet tool (for example Numbers for Mac OS, OpenOffice, LibreOffice etc) but be aware that some of the functions differ slightly.
- .Use these resources (and Google) to improve your skills as you progress.
Step 2: Analyse Data & Submit Report
- You will be presented with a set of tasks to run on your unique data set (of 100 samples) - the full list of tasks are listed below.
- As much as possible, complete these tasks within the excel document. You will need to submit your excel document at the end of the assignment
- You will also submit a brief report in word addressing all the tasks below. When responding to the tasks, please explain the reasoning behind your answer, and refer to your Excel sheet
Data Set: Household data
The Data Set for this Assignment is already given. This includes information of 1000 households across the following variables.
- Weekly Income (WI),
- Weekly Expenditure on Food (WEF),
- Highest Level of Education (HLE)1,
- Family Size (FS), and
- Gender of the Head of Household (GHH)
Tasks for Analysis of Data Set
Complete the following tasks based on the unique data set you generated. These questions should be answered in a Word document, with brief reasoning to justify your answers. Your answers and reasoning should correlate with the excel sheet (where applicable, make references in this document to your excel sheet)
Task 1
A. Organise your sample data in a spreadsheet as per the instructions in the Excel sheet
B. What sampling method is used to select your sample data?
C. Do you think that is the best method of sampling? Why not? Why yes?
D. What is the best statistic used to compare the volatility in WEF, WI, and FS values? Why?
Task 2
A. Develop the tabular form and graphical bar chart of WI based on the following classification:
- 1st Class = Very Poor
- 2nd Class = Poor
- 3rd Class = Moderate
- 4th Class = Rich
- 5th Class = Very Rich.
Hint: You can divide your data into 5 parts (do not use 20th, 40th, 60th, and 80th percentiles as your bins since it will divide data uniformly into 5 parts) using the range of this data. You can draw Box-Whisker plot to decide the cut points or bins for these classes.
B. What is the most frequent group in your WI sample data? What does that indicate in terms of your data distribution?
C. Do you think your WI of sample data is normally distributed? Provide the "statistical reason" for your answer?
Task 3
A. What is the top 10% value and the bottom 10 % value of WEF of the selected households?
B. What is the probability that your WI values will be less than or equal to $200?
C. What is the probability that FS will be equal to 2?
D. Is there any outlier(s) in your sample data of WEF? Show the graph or prove for that! If yes, what is the best statistic to measure the dispersion of your WEF?
Task 4
A. What is the probability that the head of household is woman and her HLE is Primary?
B. What is the probability that the head of household is man and has the College degree?
C. What is the proportion of having the Secondary as the highest degree from among males?
D. What is the proportion of having the Intermediate as the highest degree from among males?
E. Do you think that the events "gender of household head is male" and "having the College Degree" are independent?