‹ Back to the paper
Question 7(a)
5 marks
MathematicsISC 2017 Mathematics Paper 1RegressionLinear Regression
Answer the following on regression and correlation.
Calculate the Spearman’s rank correlation coefficient for the following data and interpret the…
Calculate the Spearman’s rank correlation coefficient for the following data and interpret the result:
| $X$ | 35 | 54 | 80 | 95 | 73 | 73 | 35 | 91 | 83 | 81 |
|---|---|---|---|---|---|---|---|---|---|---|
| $Y$ | 40 | 60 | 75 | 90 | 70 | 75 | 38 | 95 | 75 | 70 |
Answer
Answer
Rank correlation: 0.876
Ranks (highest value = rank 1; equal values share the mean rank):
$R_x$: 35 → 9.5, 54 → 8, 80 → 5, 95 → 1, 73 → 6.5, 73 → 6.5, 35 → 9.5, 91 → 2, 83 → 3, 81 → 4
$R_y$: 40 → 9, 60 → 8, 75 → 4, 90 → 2, 70 → 6.5, 75 → 4, 38 → 10, 95 → 1, 75 → 4, 70 → 6.5
$d = R_x - R_y$: 0.5, 0, 1, -1, 0, 2.5, -0.5, 1, -1, -2.5, so $\sum d^2 = 17$.
Tie corrections $\frac{m(m^2-1)}{12}$: $X$ has two pairs, $0.5 + 0.5$; $Y$ has one triple and one pair, $2 + 0.5$. Total $= 3.5$.
$r_s = 1 - \frac{6\left(\sum d^2 + 3.5\right)}{n(n^2-1)} = 1 - \frac{6(17+3.5)}{10\times 99} = 1 - \frac{123}{990} \approx 0.876$
Interpretation: $r_s \approx 0.88$ is close to $+1$, so there is a high positive correlation between the ranks of $X$ and $Y$.