15 extra multiple-choice questions for Correlation Analysis (12th Standard Statistics, Samacheer Kalvi), beyond the ones printed in the textbook — each with the correct option highlighted and a clear, worked explanation. Free to read in English and Tamil.
Q1
Two judges rank 5 candidates and the sum of squares of rank differences is \( \sum D_i^2 = 10 \). Spearman's rank correlation coefficient is
- A. 0.52
- B. -0.5
- C. 0.5Correct
- D. 0.917
Explanation. \( \rho = 1 - \dfrac{6\sum D_i^2}{n(n^2-1)} = 1 - \dfrac{6 \times 10}{5 \times 24} = 1 - \dfrac{60}{120} = 0.5 \). Using \( n^3 = 125 \) gives 0.52; leaving out the 6 gives 0.917.
Q2
If \( \text{cov}(x, y) = 12 \), \( \sigma_x = 4 \) and \( \sigma_y = 5 \), Karl Pearson's coefficient of correlation is
- A. 0.48
- B. 0.6Correct
- C. 0.75
- D. 2.4
Explanation. \( r = \dfrac{\text{cov}(x,y)}{\sigma_x \sigma_y} = \dfrac{12}{4 \times 5} = \dfrac{12}{20} = 0.6 \). Dividing by \( 5^2 \), \( 4^2 \) or by one standard deviation gives the other values.
Q3
Asked in board exam:
March 2026
If \( r = 0.5 \), \( \sigma_x = 3 \) and \( \sigma_y = 4 \), then \( \text{cov}(x, y) \) is
- A. 6Correct
- B. 3.5
- C. 0.125
- D. 24
Explanation. From \( r = \dfrac{\text{cov}(x,y)}{\sigma_x \sigma_y} \), \( \text{cov}(x,y) = r\,\sigma_x \sigma_y = 0.5 \times 3 \times 4 = 6 \).
Q4
For \( n = 5 \) pairs, \( \sum x = 10 \), \( \sum y = 20 \), \( \sum x^2 = 30 \), \( \sum y^2 = 90 \) and \( \sum xy = 46 \). Karl Pearson's coefficient of correlation is
- A. 0.36
- B. 0.92
- C. 0.3
- D. 0.6Correct
Explanation. Numerator \( 5 \times 46 - 10 \times 20 = 30 \). \( 5 \times 30 - 10^2 = 50 \) and \( 5 \times 90 - 20^2 = 50 \). So \( r = \dfrac{30}{\sqrt{50 \times 50}} = \dfrac{30}{50} = 0.6 \).
Q5
If Spearman's rank correlation coefficient is 0.8 and \( \sum D^2 = 4 \), the number of pairs n is
- A. 4
- B. 5Correct
- C. 6
- D. 10
Explanation. \( 0.8 = 1 - \dfrac{6 \times 4}{n^3 - n} \) gives \( \dfrac{24}{n^3 - n} = 0.2 \), so \( n^3 - n = 120 = 5 \times 4 \times 6 \). Hence \( n = 5 \); n = 4 gives 60 and n = 6 gives 210.
Q6
If three items are ranked equal at the 6th place, the rank given to each of them and the next rank are
- A. 6 each; next rank 7
- B. 7 each; next rank 9Correct
- C. 7 each; next rank 8
- D. 6.5 each; next rank 9
Explanation. Tied items get the average of the ranks they would have received: \( \dfrac{6 + 7 + 8}{3} = 7 \) each. Ranks 6, 7 and 8 are used up, so the next rank is 9.
Q7
In computing rank correlation with repeated ranks, one value occurs 3 times. The correction factor \( \dfrac{m^3 - m}{12} \) added to \( \sum D_i^2 \) for it is
- A. 0.5
- B. 2.25
- C. 2Correct
- D. 24
Explanation. With \( m = 3 \), \( \dfrac{m^3 - m}{12} = \dfrac{27 - 3}{12} = \dfrac{24}{12} = 2 \), as in Example 4.6. A value repeated twice adds \( \dfrac{8 - 2}{12} = 0.5 \).
Q8
The correlation coefficient between x and y is 0.6. If \( u = \dfrac{x - 10}{2} \) and \( v = \dfrac{y - 5}{3} \), the correlation coefficient between u and v is
- A. 0.4
- B. 0.9
- C. 0.1
- D. 0.6Correct
Explanation. The correlation coefficient is unaffected by change of origin and scale, so \( r_{uv} = r_{xy} = 0.6 \). This is why the short-cut method of Example 4.1 gives the same 0.603 as the direct method.
Q9
In a \( 2 \times 2 \) table, \( (AB) = 20 \), \( (A\beta) = 10 \), \( (\alpha B) = 10 \) and \( (\alpha\beta) = 20 \). Yule's coefficient of association Q is
- A. 0.6Correct
- B. -0.6
- C. 0.25
- D. 0.8
Explanation. \( Q = \dfrac{(AB)(\alpha\beta) - (A\beta)(\alpha B)}{(AB)(\alpha\beta) + (A\beta)(\alpha B)} = \dfrac{400 - 100}{400 + 100} = \dfrac{300}{500} = 0.6 \), a positive (direct) association.
Q10
Given \( N = 200 \), \( (A) = 120 \), \( (B) = 90 \) and \( (AB) = 5 \), the cell frequency \( (\alpha\beta) \) and the conclusion about the data are
- A. \( (\alpha\beta) = 5 \); the data are consistent
- B. \( (\alpha\beta) = 75 \); the data are consistent
- C. \( (\alpha\beta) = -5 \); the data are consistent
- D. \( (\alpha\beta) = -5 \); the data are inconsistentCorrect
Explanation. \( (\alpha) = 200 - 120 = 80 \), \( (\alpha B) = 90 - 5 = 85 \), so \( (\alpha\beta) = 80 - 85 = -5 \). A negative cell frequency means the data are inconsistent.
Q11
The correlation between two variables while removing the effect of other variables is called
- A. simple correlation
- B. partial correlationCorrect
- C. multiple correlation
- D. rank correlation
Explanation. Partial correlation, denoted \( r_{xy.z} \), measures the correlation between two variables after removing the effect of the others. Multiple correlation relates a variable to a group of variables.
Q12
Consider the following statements about correlation:
1. Outliers strongly influence the correlation coefficient.
2. A high correlation proves that a change in one variable causes a change in the other.
3. Spurious correlation is an association that may not exist in reality.
Which of these are correct?
- A. 1 and 2 only
- B. 2 and 3 only
- C. 1 and 3 onlyCorrect
- D. 1, 2 and 3
Explanation. The book lists outliers as a limitation and says correlation does not imply a causal relationship, so statement 2 is wrong. 'Spurious' means false: an association drawn from r that may not exist in reality.
Q13
Assertion (A): For data on age and health care, the linear correlation coefficient may be zero.
Reason (R): \( r = 0 \) means only that there is no linear relationship; a non-linear relationship may exist.
- A. Both A and R are true and R is the correct explanation of ACorrect
- B. Both A and R are true but R is not the correct explanation of A
- C. A is true but R is false
- D. A is false but R is true
Explanation. Children and elderly people need more health care than adults, a curvilinear pattern. Computed r may be zero because uncorrelated means no linear relationship, which is exactly what R says.
Q14
Match the measure with the type of data it suits:
(i) Karl Pearson's coefficient - (P) ordinal data
(ii) Spearman's rank correlation - (Q) two dichotomous attributes
(iii) Yule's coefficient - (R) ratio scale data
- A. (i)-R, (ii)-P, (iii)-QCorrect
- B. (i)-P, (ii)-R, (iii)-Q
- C. (i)-R, (ii)-Q, (iii)-P
- D. (i)-Q, (ii)-P, (iii)-R
Explanation. Pearson's r is for ratio-scale data, Spearman's \( \rho \) for two ordinal variables, and Yule's Q for the association between two dichotomous attributes such as smoking or blindness.
Q15
Using the book's scale for interpreting Spearman's coefficient, a value \( \rho = 0.45 \) indicates
- A. weak agreement
- B. moderate agreementCorrect
- C. strong agreement
- D. very strong agreement
Explanation. On the positive range, 0.40 to 0.59 is 'moderate agreement'; 0.20 to 0.39 is weak, 0.60 to 0.79 strong and 0.80 to 1.0 very strong agreement.