- kg/feet
- percentage
- non-existent
(iii) non-existent. The correlation coefficient has no unit; it is a pure number, so the units of measurement (feet, kg) do not form part of r.
(iii) non-existent
NCERT solutions for Class 11 Economics Statistics for Economics Chapter 6: Correlation (2026-27). Answers to all exercises, with full working for Karl Pearson's correlation coefficient, plus the in-text activities. All 25 questions are answered, with the key answer highlighted.
NCERT textbook Chapter 6 (PDF)
Exam revision? Chapter 6 notes with key points, diagrams and practice questions: see the free sample · US$2.99 for all 13 chapters
(iii) non-existent. The correlation coefficient has no unit; it is a pure number, so the units of measurement (feet, kg) do not form part of r.
(iii) non-existent
(ii) minus one to plus one. The value of r always lies between −1 and +1; a value outside this range indicates an error in calculation.
(ii) minus one to plus one
(i) When Y increases X increases. A positive r means the two variables move in the same direction.
(i) When Y increases X increases
(ii) not linearly related. Zero correlation means there is no linear relation between X and Y; some other (non-linear) relation may still exist, so it does not prove independence.
(ii) not linearly related
(iii) Scatter diagram. Karl Pearson's coefficient and Spearman's rank correlation measure only linear association numerically, whereas a scatter diagram shows the nature of any relationship, linear or non-linear.
(iii) Scatter diagram
(i) more accurate than rank correlation coefficient. Rank correlation uses only the ranks, so part of the information in the data is not used; with precisely measured data the simple (Karl Pearson) coefficient is the more accurate.
(i) more accurate than rank correlation coefficient
Covariance, Σ(X − X̄)(Y − Ȳ)/N, is expressed in the units of X and Y multiplied together, so its value changes when the units or scale change and it cannot be compared across different pairs of variables. The correlation coefficient r = Cov(X, Y)/(σxσy) divides covariance by the two standard deviations, so it is a pure number free of units and of change of origin and scale. It always lies between −1 and +1, so its size shows the strength of the relation and its sign gives the direction, which makes r easy to interpret and compare.
No. Whatever the type of data, r lies between −1 and +1. If a calculation gives a value outside this range, it indicates an error in the calculation.
No
No. Correlation only means covariation: two variables move together. It does not show that one causes the other. The relation may be due to a third factor or may be a simple coincidence. For example, the number of doctors sent to epidemic-hit villages and the number of deaths may be positively correlated, but doctors do not cause the deaths; the epidemic raises both.
No
Rank correlation is more suitable when the variables cannot be measured precisely, such as beauty, honesty or fairness, or when the data are only ranks. It is also preferable when the data contain extreme values, since ranks are not affected by them, and when the relation is clearly directional but non-linear. In these situations the simple coefficient can be misleading while the rank coefficient is better. When data are precisely measured and the relation is linear, the simple coefficient is more accurate.
No. Zero correlation only means there is no linear relationship between the variables. They may still be related in a non-linear way. For example, for X = −3, −2, −1, 1, 2, 3 and Y = X² (Exercise 19) r = 0, yet Y is completely determined by X.
No
No. The simple (Karl Pearson) correlation coefficient measures only the strength and direction of a linear relationship. If the true relation is non-linear, as in the scatter diagrams of Figs. 6.6 and 6.7, r can be misleading, so the scatter diagram should be examined first.
No, only linear relationship
Guidance (data must be collected by you): make a table of the daily prices of the five vegetables for seven days. For each pair of vegetables take the prices as X and Y and use r = [ΣXY − (ΣX)(ΣY)/N] / √{[ΣX² − (ΣX)²/N][ΣY² − (ΣY)²/N]} with N = 7, which gives 10 coefficients for the 10 pairs. Interpret each: a value near +1 means the prices rise and fall together (for example because of the same season or transport cost); a value near −1 means they move in opposite directions; a value near 0 means no linear relation. Remember that correlation does not prove that one price causes the other.
Guidance (data must be collected by you): record for each student the measured height X and the height Y stated for the benchmate. Calculate r with formula (3) or by taking deviations from the means. Since the stated heights are estimates of the benchmate's height (and a benchmate is often of similar height), you should expect a positive r. Interpretation: a value close to +1 means the students estimate heights well and tall students sit with tall benchmates; a low value means weak linear association. It shows association only, not cause.
Beauty, honesty, fairness, intelligence, leadership quality, taste of a dish and skill are such variables. They cannot be measured exactly like income, height or weight; at most they can be ranked, so rank correlation is used for them.
Spearman's rank coefficient is worked out from the ranks assigned to the observations, not from their actual values, so it does not use all the information in the data. Pearson's coefficient uses the actual values. The two give identical results only if the first differences of the values (the differences between consecutive values in order of magnitude) are constant. Usually the values cluster around the centre with smaller differences in the middle, so the rank coefficient is generally less than or equal to r and is not as accurate; it is, however, not affected by extreme values.
| X | 65 | 66 | 57 | 67 | 68 | 69 | 70 | 72 |
|---|---|---|---|---|---|---|---|---|
| Y | 67 | 56 | 65 | 68 | 72 | 72 | 69 | 71 |
N = 8. Use r = [ΣXY − (ΣX)(ΣY)/N] / √{[ΣX² − (ΣX)²/N][ΣY² − (ΣY)²/N]}.
| X | Y | X² | Y² | XY |
|---|---|---|---|---|
| 65 | 67 | 4,225 | 4,489 | 4,355 |
| 66 | 56 | 4,356 | 3,136 | 3,696 |
| 57 | 65 | 3,249 | 4,225 | 3,705 |
| 67 | 68 | 4,489 | 4,624 | 4,556 |
| 68 | 72 | 4,624 | 5,184 | 4,896 |
| 69 | 72 | 4,761 | 5,184 | 4,968 |
| 70 | 69 | 4,900 | 4,761 | 4,830 |
| 72 | 71 | 5,184 | 5,041 | 5,112 |
| ΣX = 534 | ΣY = 540 | ΣX² = 35,788 | ΣY² = 36,644 | ΣXY = 36,118 |
ΣXY − (ΣX)(ΣY)/N = 36,118 − (534 × 540)/8 = 36,118 − 36,045 = 73.
ΣX² − (ΣX)²/N = 35,788 − 534²/8 = 35,788 − 35,644.5 = 143.5.
ΣY² − (ΣY)²/N = 36,644 − 540²/8 = 36,644 − 36,450 = 194.
r = 73/√(143.5 × 194) = 73/√27,839 = 73/166.85 = 0.4375.
Check by deviations from the means (X̄ = 66.75, Ȳ = 67.5): Σxy = 73, Σx² = 143.5, Σy² = 194, which give the same r.
Comment: r is positive but moderate, so taller fathers tend to have taller sons, though the relation is not strong. (The answer printed in the book, 0.603, does not match these data; the working above is from the table as printed.)
r = 0.44 (approximately)
| X | −3 | −2 | −1 | 1 | 2 | 3 |
|---|---|---|---|---|---|---|
| Y | 9 | 4 | 1 | 1 | 4 | 9 |
N = 6.
| 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 |
ΣXY − (ΣX)(ΣY)/N = 0 − (0 × 28)/6 = 0.
The denominator is not zero: ΣX² − (ΣX)²/N = 28 and ΣY² − (ΣY)²/N = 196 − 784/6 = 65.33, so r = 0/√(28 × 65.33) = 0.
Comment: there is no linear correlation, yet Y = X² exactly. The relation is perfect but non-linear (U-shaped), so zero correlation does not mean the variables are independent.
r = 0
| X | 1 | 3 | 4 | 5 | 7 | 8 |
|---|---|---|---|---|---|---|
| Y | 2 | 6 | 8 | 10 | 14 | 16 |
N = 6.
| 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 |
ΣXY − (ΣX)(ΣY)/N = 328 − (28 × 56)/6 = 328 − 261.33 = 66.67.
ΣX² − (ΣX)²/N = 164 − 784/6 = 164 − 130.67 = 33.33.
ΣY² − (ΣY)²/N = 656 − 3,136/6 = 656 − 522.67 = 133.33.
r = 66.67/√(33.33 × 133.33) = 66.67/√4,444.4 = 66.67/66.67 = 1.
Comment: perfect positive linear correlation. Every Y is exactly twice the corresponding X (Y = 2X), so all the points lie on one rising straight line.
r = +1 (perfect positive correlation)
Guidance (data must be collected by you): plot one variable on the X-axis and the other on the Y-axis for each student, one point per student, for pairs such as height-weight, marks in subject 1-marks in subject 2, height-marks. Typically height and weight show a positive relation, with the points rising from left to right; marks in two subjects often also move together positively; height and marks usually show a scatter with no upward or downward direction, that is, no correlation. Classify each diagram as in Figs. 6.1 to 6.7 (positive, negative, none, perfect, or non-linear).
| Year | Annual growth of National Income (X) | Gross Domestic Saving as percentage of GDP (Y) |
|---|---|---|
| 1992-93 | 14 | 24 |
| 1993-94 | 17 | 23 |
| 1994-95 | 18 | 26 |
| 1995-96 | 17 | 27 |
| 1996-97 | 16 | 25 |
| 1997-98 | 12 | 25 |
| 1998-99 | 16 | 23 |
| 1999-00 | 11 | 25 |
| 2000-01 | 8 | 24 |
| 2001-02 | 10 | 23 |
N = 10. Use r = [ΣXY − (ΣX)(ΣY)/N] / √{[ΣX² − (ΣX)²/N][ΣY² − (ΣY)²/N]}.
| X | Y | X² | Y² | XY |
|---|---|---|---|---|
| 14 | 24 | 196 | 576 | 336 |
| 17 | 23 | 289 | 529 | 391 |
| 18 | 26 | 324 | 676 | 468 |
| 17 | 27 | 289 | 729 | 459 |
| 16 | 25 | 256 | 625 | 400 |
| 12 | 25 | 144 | 625 | 300 |
| 16 | 23 | 256 | 529 | 368 |
| 11 | 25 | 121 | 625 | 275 |
| 8 | 24 | 64 | 576 | 192 |
| 10 | 23 | 100 | 529 | 230 |
| ΣX = 139 | ΣY = 245 | ΣX² = 2,039 | ΣY² = 6,019 | ΣXY = 3,419 |
ΣXY − (ΣX)(ΣY)/N = 3,419 − (139 × 245)/10 = 3,419 − 3,405.5 = 13.5.
ΣX² − (ΣX)²/N = 2,039 − 139²/10 = 2,039 − 1,932.1 = 106.9.
ΣY² − (ΣY)²/N = 6,019 − 245²/10 = 6,019 − 6,002.5 = 16.5.
r = 13.5/√(106.9 × 16.5) = 13.5/√1,763.85 = 13.5/42.00 = 0.32.
Comment: a weak positive correlation, so higher growth of national income is only slightly associated with a higher saving rate.
r = 0.32 (weak positive correlation)
Guidance (data must be taken from a source such as the Economic Survey): take at least ten years of population (X) and national income (Y). Choose assumed means A and C and common factors B and D, and form U = (X − A)/B and V = (Y − C)/D. Then calculate ΣU, ΣV, ΣU², ΣV² and ΣUV and apply formula (3) to U and V: rxy = ruv, because r is independent of change of origin and scale. Both series usually rise over time, so r is expected to be strongly positive; remember that this is covariation, not causation.
Guidance for the calculation: rank the 10 students by Class IX marks and by Class X marks (rank 1 for the highest), find D = difference of ranks, and use rs = 1 − 6ΣD²/(n³ − n) with n = 10. If marks repeat, give the tied items the average of the ranks they would have taken and add the correction factor (m³ − m)/12 for each repeated group to ΣD² before applying the formula.
Guidance (data must be taken from a source such as the Economic Survey): list the national income (X) and exports (Y) for at least ten consecutive years in a table.
All the methods must give the same value of r. Both series generally rise over time, so a strong positive r is expected; it shows that they move together and not that one causes the other.

Found a mistake or need help with a question? Message us on WhatsApp. Free study tips and new solutions: join our WhatsApp Channel.