5. (i) 0.60 (ii) 0.30 (iii) 0.95

 Reasoning: correlation (iii) must be nearly 1; correlations (i) and (ii) are moderate, with (i) being stronger.

 6. This is false (p.126).

 8. (a) 42 inches

2. 2.5 inches
3. 0.80
4. solid

 9. (a) -0.80

2. 0.3
3. 1.00

 Comment. In (c), all the points lie on the line y = 2x, so there is no need to do any arithmetic.

 10. Smaller when the score on form L is 75. If you take narrow vertical chimneys over 75 and 125 in the scatter diagram, there is less vertical spread--therefore, less uncertainty in the predictions--in the chimney over 75.

 2. (a) False. The scatter diagram slopes down: if x is below average, y is generally above average.

(b) False. For example, height at age 9 is usually less than height at age 18, but the correlation is positive.

 4. Somewhat higher; see exercise set B, pp. 145ff.

 7. No: section 4.

 8. False. The data are cross-sectional, not longitudinal. Younger people were born later and educational levels have been going up over time.

 12. (a) Education is measured in whole years-years of schooling completed.

(b) Some dots correspond to many couples. For example, in many families, the husband and the wife both have 12 years of education (high school degree). All such families are plotted on the (12, 12) dot.

(c) A (iv) B (iii) C (i)