In this Assignment, you will be working with a large dataset containing the details of pa-
tients screened for a thyroid condition. The dataset is saved as a file under this Assignment.
For each patient the dataset includes age, sex, the system by which they were referred for
testing, the results of four blood tests and the result of an indicator test. You can assume
that multivariate normality holds for the quantitative observations in this dataset.
The main aim of the Assignment is to use the ‘R’ skills you have learnt in this module
to analyse this dataset in detail and answer the questions below, with written explanations
where necessary.
Questions
1. Give an introduction describing the dataset being analysed, with brief explanations
of the different columns of the data.
[4 marks
2. i) Create a new column separating the data into four Age Groups (’35 and under’,
’36 to 507, *51 to 65, “66 and over’). What are the proportions of Indicator
Status “P* and ‘N” in each age group?
[6 marks
ii) Determine the overall sample mean vector containing four variables correspond-
ing to the four quantitative variables (i.e. the four blood test results).
[4 marks
iii) Which of the four quantitative variables exhibits the greatest level of variation
around the mean?
[4 marks
iv) Which two quantitative variables, out of the four, show the highest correlation
and what type of correlation do they have?
[6 marks
3. i) Test if the population mean vector of the four quantitative variables is equal to
ji = (5,2,110,110) ”
[6 marks
