Chapter 6 Correlation
Practice questions
Case: three made-up judges rank five dishes. rs is 0.7 for judges A and B, 0.2 for A and C, and −0.4 for B and C. Which pair agrees most?
- A and C
- B and C
- All pairs equally
- A and B
Answer and explanation
Answer: D
The highest positive rank correlation, 0.7, is between A and B, so their judgements are closest. B and C, at −0.4, tend to disagree.
Illustrative: Y values 30, 25, 25, 20 are ranked from highest. The ranks are:
- 1, 2, 2, 4
- 1, 2.5, 2.5, 4
- 1, 2, 3, 4
- 1, 2.5, 2.5, 3
Answer and explanation
Answer: B
The two 25s would take ranks 2 and 3, so each gets (2 + 3) ÷ 2 = 2.5; the next value, 20, gets rank 4, not 3.
Illustrative: for 4 pairs of ranks, ΣD² = 2. Spearman's coefficient is:
- 0.9
- 0.6
- 0.8
- 0.2
Answer and explanation
Answer: C
rs = 1 − 6 × 2 ÷ (64 − 4) = 1 − 12 ÷ 60 = 1 − 0.2 = 0.8.
Assertion (A): For precisely measured data, Karl Pearson's r is generally more accurate than rank correlation. Reason (R): Rank correlation uses only the order of values, so some information in the data is not used.
- Both A and R are true, and R is the correct explanation of A
- Both A and R are true, but R is not the correct explanation of A
- A is true, but R is false
- A is false, but R is true
Answer and explanation
Answer: A
Both are true, and R explains A: ranking discards the sizes of the gaps between values, so for accurately measured data r, which uses the actual values, is more accurate.
Which statement is NOT a property of Karl Pearson's r?
- It has no unit
- It lies between −1 and +1
- r = 0 means no linear relation
- r = 0 proves the variables are independent
Answer and explanation
Answer: D
r = 0 means no linear relationship; a non-linear relationship may exist, so it does not prove independence. The other three are properties of r.
If every value of X is converted from rupees to thousands of rupees and 5 is subtracted, the correlation between X and Y:
- is divided by 1,000
- becomes zero
- remains unchanged
- changes sign
Answer and explanation
Answer: C
r is unaffected by change of origin (subtracting 5) and change of scale (dividing by 1,000), so it stays the same.
The covariance of X and Y is −12, σx = 4 and σy = 5. r is:
- −0.6
- −2.4
- 0.6
- −0.48
Answer and explanation
Answer: A
r = Cov ÷ (σx σy) = −12 ÷ (4 × 5) = −12 ÷ 20 = −0.6.
Illustrative: N = 4, ΣX = 20, ΣY = 40, ΣXY = 210, ΣX² = 110, ΣY² = 420. Using the actual-value formula, r is:
- 0.5
- 1
- 0.71
- 0
Answer and explanation
Answer: C
Numerator = 4 × 210 − 20 × 40 = 840 − 800 = 40. Denominator = √[(4 × 110 − 400) × (4 × 420 − 1,600)] = √(40 × 80) = √3,200 = 56.57. r = 40 ÷ 56.57 = 0.71. (These sums come from X = 3, 4, 6, 7 and Y = 7, 11, 9, 13.)
Illustrative: Σxy = −45, Σx² = 25 and Σy² = 100 (deviations from means). r is:
- −0.9
- −0.45
- 0.9
- −1.8
Answer and explanation
Answer: A
r = −45 ÷ √(25 × 100) = −45 ÷ √2,500 = −45 ÷ 50 = −0.9.
Illustrative: points (1, 7), (2, 9), (3, 11), (4, 13) lie on one line. The correlation is:
- −1
- 0
- 0.5
- +1
Answer and explanation
Answer: D
Each rise of 1 in X raises Y by exactly 2, so all points lie on one upward straight line: perfect positive correlation, r = +1.
Which measure can show any type of relationship, linear or non-linear?
- Scatter diagram
- Karl Pearson's coefficient
- Spearman's rank correlation
- Covariance
Answer and explanation
Answer: A
The scatter diagram shows the pattern visually and is not confined to linear relations; the two coefficients measure linear association.
A scatter diagram whose points form a U-shape indicates:
- perfect positive correlation
- a high linear correlation
- a non-linear relationship for which Karl Pearson's r would mislead
- no relationship at all
Answer and explanation
Answer: C
A U-shape is a curve, so the relationship is non-linear; r could be near zero even though Y depends on X, which is why the scatter is checked first.
Assertion (A): Correlation between two variables may arise by coincidence. Reason (R): Correlation measures covariation, not causation.
- Both A and R are true, and R is the correct explanation of A
- Both A and R are true, but R is not the correct explanation of A
- A is true, but R is false
- A is false, but R is true
Answer and explanation
Answer: B
Both are true. A is true (for example, migratory birds and a local birth rate), and R is true. But R does not explain why coincidences happen; it explains why a correlation cannot be read as causation. So R is not the explanation of A.
Which of the following shows positive correlation?
- Price of tea and demand for tea
- Price of coffee and demand for tea
- Rate of interest and demand for loans
- Supply of tomatoes and price of tomatoes
Answer and explanation
Answer: B
Coffee is a substitute for tea: when coffee becomes dearer, people buy more tea, so the two move in the same direction. The other pairs move in opposite directions.
Case: in a made-up city, the number of fans sold and the number of glasses of sugarcane juice sold both rise each summer. The most sensible explanation is:
- fans cause juice sales
- juice sales cause fan sales
- pure coincidence with no link
- a third variable, hot weather, raises both
Answer and explanation
Answer: D
Neither good causes the other; rising temperature increases demand for both, so temperature is the third variable behind the correlation.
Assertion (A): Spearman's rank correlation is preferred to Karl Pearson's r when the data contain extreme values. Reason (R): Ranks are not affected by how extreme a value is, only by its order.
- Both A and R are true, and R is the correct explanation of A
- Both A and R are true, but R is not the correct explanation of A
- A is true, but R is false
- A is false, but R is true
Answer and explanation
Answer: A
Both are true, and R explains A: an extreme value simply gets the top or bottom rank, so it cannot distort the coefficient as it can distort r.
Illustrative: a group of 3 tied items adds a correction factor of:
- 2
- 0.5
- 3
- 24
Answer and explanation
Answer: A
Correction = (m³ − m) ÷ 12 = (27 − 3) ÷ 12 = 24 ÷ 12 = 2. 24 is the numerator before dividing by 12.
Illustrative: the values 60, 55, 55, 55, 40 are ranked from highest. The rank given to each 55 is:
- 2
- 3
- 4
- 2.5
Answer and explanation
Answer: B
The three 55s would occupy ranks 2, 3 and 4; each gets the average, (2 + 3 + 4) ÷ 3 = 9 ÷ 3 = 3. The value 40 then gets rank 5.
Rank correlation is the appropriate measure when studying the relation between:
- income and expenditure in rupees
- height in cm and weight in kg measured accurately
- price and quantity in kg
- honesty and punctuality of employees, as judged by a manager
Answer and explanation
Answer: D
Honesty and punctuality are attributes that can be ranked but not measured, which is when Spearman's rank correlation is used.
Illustrative: for 5 pairs of ranks, ΣD² = 10. Spearman's rank correlation coefficient is:
- 0.9
- 0.5
- −0.5
- 0.1
Answer and explanation
Answer: B
rs = 1 − 6ΣD² ÷ (n³ − n) = 1 − 60 ÷ (125 − 5) = 1 − 60 ÷ 120 = 1 − 0.5 = 0.5.