Complete solutions to all 20 exercise questions of NCERT Statistics for Economics, Chapter 6 “Correlation” (Reprint 2026-27). The questions are printed exactly as in the English edition. Questions 13 and 14 are survey activities, so their answers give the method and how to interpret the result. The numericals (18 to 20) are solved step by step: given, formula, substitution, answer and interpretation.
1 Unit of the correlation coefficient
Q. “The unit of correlation coefficient between height in feet and weight in kgs is (i) kg/feet (ii) percentage (iii) non-existent”
Answer: (iii) non-existent. r has no unit; it is a pure number. Units of measurement (feet, kilograms) are not part of r.
2 Range of the simple correlation coefficient
Q. “The range of simple correlation coefficient is (i) 0 to infinity (ii) minus one to plus one (iii) minus infinity to infinity”
Answer: (ii) minus one to plus one. $-1 \le r \le 1$; if r comes outside this range in any exercise, it indicates an error in calculation.
3 Meaning of a positive r
Q. “If rxy is positive the relation between X and Y is of the type (i) When Y increases X increases (ii) When Y decreases X increases (iii) When Y increases X does not change”
Answer: (i) When Y increases X increases. A positive r means the two variables move in the same direction.
4 Meaning of r = 0
Q. “If rxy = 0 the variable X and Y are (i) linearly related (ii) not linearly related (iii) independent”
Answer: (ii) not linearly related. r = 0 only means there is no linear relation between X and Y; another (non-linear) relation may exist, so “independent” is not correct (see Questions 11 and 19).
5 The measure that shows any type of relationship
Q. “Of the following three measures which can measure any type of relationship (i) Karl Pearson’s coefficient of correlation (ii) Spearman’s rank correlation (iii) Scatter diagram”
Answer: (iii) Scatter diagram. A scatter diagram gives a visual presentation of the relationship and is not confined to linear relations. Both Karl Pearson’s and Spearman’s coefficients measure linear relationship.
6 Simple coefficient with precise data
Q. “If precisely measured data are available the simple correlation coefficient is (i) more accurate than rank correlation coefficient (ii) less accurate than rank correlation coefficient (iii) as accurate as the rank correlation coefficient”
Answer: (i) more accurate than rank correlation coefficient. Rank correlation does not use all the information in the data (only the order, not the gaps between values), so it is generally not as accurate as the ordinary method.
7 Why r is preferred to covariance
Q. Why is r preferred to covariance as a measure of association?
Answer:
- Covariance $\mathrm{Cov}(X,Y) = \sum xy/N$ shows only the direction (sign) of the relation correctly; its size depends on the units and scale of the variables (measure height in inches instead of feet and the covariance changes)
- So covariance has no fixed limits and cannot tell how strong the relation is
- r divides the covariance by both standard deviations: $r = \dfrac{\mathrm{Cov}(X,Y)}{\sigma_x \sigma_y}$
- This makes r unit-free, unaffected by change of origin and scale, and always between −1 and +1
- So r shows both direction and intensity, and lets us compare relations between variables measured in different units
8 Can r lie outside −1 to 1
Q. “Can r lie outside the –1 and 1 range depending on the type of data?”
Answer: No. Whatever the type of data, r always lies in $-1 \le r \le 1$. r = +1 and r = −1 are the limits of perfect correlation. If r comes outside this range in any exercise, it indicates an error in calculation.
9 Correlation and causation
Q. Does correlation imply causation?
Answer:
- No. Correlation measures covariation (moving together), not causation
- Correlation can arise by coincidence: arrival of migratory birds and birth rates in the locality, or size of shoes and money in your pocket
- It can arise from a third variable: ice-cream sales and deaths by drowning are highly correlated, but temperature is behind both
- So a high correlation does not show that one variable causes the other; causation is decided by economic theory and reasoning
10 When rank correlation is more precise
Q. When is rank correlation more precise than simple correlation coefficient?
Answer:
- When the variables cannot be measured precisely, such as honesty, beauty or intelligence, or when there are no measuring instruments; then only the ranks are reliable
- When the data contain extreme values: Pearson’s r is affected by them, rank correlation is not
- When the direction of the relation is clear but it is non-linear (Figs 6.6, 6.7)
- With precisely measured data, the simple coefficient stays more accurate (Question 6)
11 Does zero correlation mean independence
Q. Does zero correlation mean independence?
Answer: No. r = 0 only means there is no linear relation between the variables. They may have a non-linear relation. In Question 19, Y = X² is an exact relation and still r = 0. Independent variables do have r = 0, but r = 0 does not prove the variables are independent.
12 Can the simple coefficient measure any relation
Q. Can simple correlation coefficient measure any type of relationship?
Answer: No. The simple (Karl Pearson’s) correlation coefficient measures only a linear relationship. For a non-linear relation (Figs 6.6, 6.7) it can be misleading. That is why the scatter diagram should be examined before calculating r; only the scatter diagram can show any type of relationship.
13 Correlation of vegetable prices (activity)
Q. Collect the price of five vegetables from your local market every day for a week. Calculate their correlation coefficients. Interpret the result.
Answer (method): this is a survey activity; the data will come from your own market.
- Choose 5 vegetables (for example potato, onion, tomato, lady finger, cauliflower) and note the price per kg every day for 7 days, at the same time of day
- Each pair of vegetables (say potato and onion) gives 7 pairs (X, Y); 5 vegetables make 10 such pairs
- Draw the scatter diagram first; if the points lie around a line, find r by formula (3): $r = \dfrac{\sum XY – \frac{(\sum X)(\sum Y)}{N}}{\sqrt{\sum X^2 – \frac{(\sum X)^2}{N}}\sqrt{\sum Y^2 – \frac{(\sum Y)^2}{N}}}$, N = 7
- Interpretation: a positive r means the two prices rise and fall together (for example, rain or a transport problem hitting both); a negative r means opposite movement; r near 0 means no linear relation. Remember that correlation does not show causation
14 Correlation of classmates’ heights (activity)
Q. Measure the height of your classmates. Ask them the height of their benchmate. Calculate the correlation coefficient of these two variables. Interpret the result.
Answer (method):
- Write each student’s height (X) and the benchmate’s height (Y) in centimetres; each student gives one pair (X, Y)
- Heights are large numbers, so use the step deviation method, for example $U = (X – 150)/5$, $V = (Y – 150)/5$, and find r from U and V; r stays the same
- Interpretation: if benches were given without regard to height, r is likely to be close to zero: a student’s height has no linear link with the benchmate’s height. If taller students were seated together at the back, r may come out positive; that is due to the seating rule (a third factor), not because one height causes the other
15 Variables that are hard to measure accurately
Q. List some variables where accurate measurement is difficult.
Answer: these are attributes, which cannot be measured numerically and can only be ranked:
- Intelligence of people
- Physical appearance and beauty
- Honesty and fairness
- Leadership ability, the taste of a dish, customer satisfaction
- Correlation between such variables is found with Spearman’s rank correlation
16 Interpreting r = 1, −1 and 0
Q. “Interpret the values of r as 1, –1 and 0.”
Answer:
- r = 1: perfect positive correlation. The two variables move in the same direction in an exact linear relation; all points of the scatter diagram lie on one rising straight line (as in Question 20, Y = 2X)
- r = −1: perfect negative correlation. The variables move in opposite directions in an exact linear relation; all points lie on one falling straight line
- r = 0: the variables are uncorrelated, i.e. there is no linear relation between them; a non-linear relation may still exist (Question 19)
17 Why rank correlation differs from Pearson’s
Q. Why does rank correlation coefficient differ from Pearsonian correlation coefficient?
Answer:
- Pearson’s coefficient uses the actual values of the variables; rank correlation uses only their ranks (order)
- Ranking loses the information about the gaps between values: 85 and 75 differ by 10 and 60 and 55 by 5, but as ranks both gaps become 1. So not all the information in the data is used
- The first differences (differences of consecutive values) of a series arranged in order of magnitude are almost never constant; the data usually cluster around the central values with smaller differences in the middle. If the first differences were constant, both coefficients would give identical results
- So in general the rank correlation $r_k$ is less than or equal to r; also, extreme values affect r but not $r_k$
18 Heights of fathers and sons (numerical)
Q. Calculate the correlation coefficient between the heights of fathers in inches (X) and their sons (Y)
X: 65, 66, 57, 67, 68, 69, 70, 72
Y: 67, 56, 65, 68, 72, 72, 69, 71
(Ans. r = 0.603)
Solution: the values are large and the means (66.75, 67.5) are decimals, so take deviations from an assumed mean of 68 (a change of origin does not change r). N = 8
| X | Y | dx = X − 68 | dy = Y − 68 | dx² | dy² | dx·dy |
|---|---|---|---|---|---|---|
| 65 | 67 | −3 | −1 | 9 | 1 | 3 |
| 66 | 56 | −2 | −12 | 4 | 144 | 24 |
| 57 | 65 | −11 | −3 | 121 | 9 | 33 |
| 67 | 68 | −1 | 0 | 1 | 0 | 0 |
| 68 | 72 | 0 | 4 | 0 | 16 | 0 |
| 69 | 72 | 1 | 4 | 1 | 16 | 4 |
| 70 | 69 | 2 | 1 | 4 | 1 | 2 |
| 72 | 71 | 4 | 3 | 16 | 9 | 12 |
| ΣX = 534 | ΣY = 540 | Σdx = −10 | Σdy = −4 | Σdx² = 156 | Σdy² = 196 | Σdxdy = 78 |
Formula: $r = \dfrac{\sum d_x d_y – \frac{(\sum d_x)(\sum d_y)}{N}}{\sqrt{\sum d_x^2 – \frac{(\sum d_x)^2}{N}}\;\sqrt{\sum d_y^2 – \frac{(\sum d_y)^2}{N}}}$
Substituting: numerator $= 78 – \dfrac{(-10)(-4)}{8} = 78 – 5 = 73$; X part $= 156 – \dfrac{100}{8} = 143.5$; Y part $= 196 – \dfrac{16}{8} = 194$
$r = \dfrac{73}{\sqrt{143.5 \times 194}} = \dfrac{73}{\sqrt{27839}} = \dfrac{73}{166.85} = 0.4375$
Answer: r ≈ 0.44. The heights of fathers and sons have a positive but moderate-to-low correlation: taller fathers tend to have somewhat taller sons, but the relation is not strong.
Note: the calculation from the data printed above gives r ≈ 0.44; the answer printed with the question, 0.603, does not follow from these data. In the exam, show your full working from the data given.
19 X and Y = X² (numerical)
Q. Calculate the correlation coefficient between X and Y and comment on their relationship:
X: −3, −2, −1, 1, 2, 3
Y: 9, 4, 1, 1, 4, 9
(Ans. r = 0)
| X | Y | X² | Y² | XY |
|---|---|---|---|---|
| −3 | 9 | 9 | 81 | −27 |
| −2 | 4 | 4 | 16 | −8 |
| −1 | 1 | 1 | 1 | −1 |
| 1 | 1 | 1 | 1 | 1 |
| 2 | 4 | 4 | 16 | 8 |
| 3 | 9 | 9 | 81 | 27 |
| ΣX = 0 | ΣY = 28 | ΣX² = 28 | ΣY² = 196 | ΣXY = 0 |
Substituting in formula (3): numerator $= \sum XY – \dfrac{(\sum X)(\sum Y)}{N} = 0 – \dfrac{0 \times 28}{6} = 0$, so
Answer: r = 0
Comment: r = 0, yet X and Y are completely related: in every pair $Y = X^2$. The relation is not linear but a curve (a parabola): Y falls over the negative values of X and rises over the positive values, and the two halves cancel out. So r = 0 means only “no linear relation”, not independence.
20 X and Y = 2X (numerical)
Q. Calculate the correlation coefficient between X and Y and comment on their relationship
X: 1, 3, 4, 5, 7, 8
Y: 2, 6, 8, 10, 14, 16
(Ans. r = 1)
| X | Y | X² | Y² | XY |
|---|---|---|---|---|
| 1 | 2 | 1 | 4 | 2 |
| 3 | 6 | 9 | 36 | 18 |
| 4 | 8 | 16 | 64 | 32 |
| 5 | 10 | 25 | 100 | 50 |
| 7 | 14 | 49 | 196 | 98 |
| 8 | 16 | 64 | 256 | 128 |
| ΣX = 28 | ΣY = 56 | ΣX² = 164 | ΣY² = 656 | ΣXY = 328 |
Formula (4): $r = \dfrac{N\sum XY – (\sum X)(\sum Y)}{\sqrt{N\sum X^2 – (\sum X)^2}\;\sqrt{N\sum Y^2 – (\sum Y)^2}}$, N = 6
Substituting: numerator $= 6 \times 328 – 28 \times 56 = 1968 – 1568 = 400$; denominator $= \sqrt{984 – 784}\,\sqrt{3936 – 3136} = \sqrt{200}\,\sqrt{800} = \sqrt{160000} = 400$
Answer: r = 400/400 = 1
Comment: perfect positive correlation. In every pair Y = 2X, an exact linear relation; all points of the scatter diagram lie on one rising straight line.