Assignment:
All exercises in this and subsequent chapters may be solved by computer, unless the question states specifically that they should be done by hand.
1. Extract the data on INCOME for Cambridgeshire (County 5) by parish from the Boyer relief data set.
(i) By hand, use these data to construct:
• An array
• A frequency distribution using 5 class intervals of equal width of £5, beginning at <£20
• A frequency distribution using 10 class intervals of equal width of £2, beginning at <£20
• A frequency distribution using 16 class intervals of equal width of £1, beginning at <£20.
(ii) Which frequency distribution provides the information on Cambridgeshire incomes in the most useful fashion?
(iii) What difference does it make if you round the income figures to 1 decimal place before constructing the frequency distributions?
(iv) What difference does it make if you round the income figures to the nearest whole number before constructing the frequency distribu-tions?
(v) Instruct the computer to construct a histogram with 10 intervals. Are the resulting class intervals the same as for the relative frequency distribution in (i) above? Explain your finding.
2. Take the WEALTH variable for every parish in the Boyer relief data set.
(i) Calculate the following statistics:
• Upper, lower, and middle quartiles
• 10th and 90th percentiles
• Range
• Mean, median, and mode
• Variance
• Standard deviation
• Coefficient of variation.
(ii) Plot a histogram of WEALTH with: (a) 31 intervals and (b) 11 intervals.
(iii) Plot a bar chart of mean WEALTH in each of 21 counties.
Comment on the relationship between the different measures of central tendency and dispersion, and discuss what they tell you about wealth in
these parishes.
3. Using Boyer's birth rate data set, calculate the means and standard deviations of INCOME, BRTH RATE, and DENSITY, counting each parish as a single observation. Now calculate the means and standard deviations of the same variables, with each parish weighted by its population (POP). How would you explain the difference between the two measures? Is there any reason to prefer one set of measures to the other?
Attachment:- WEALTH variable dataset.rar