HOMEWORK #2 ANSWERS

Chapter 4

1. (a) Average = 50. Deviations = -9, -2, 0, 0, 4, 7. The SD is

(b) 48, 50, 50 are within 0.5 SDs of average, i.e., in the range 47.5-52.5.

48, 50, 50, 54, 57 are within 1.5 SDs of average, i.e., in the range 42.5-57.5.

3. (a) 5: only three of the numbers are smaller than 1, and none are bigger than 10.

(b) 3: if the SD is 1, the entries 0.6 and 9.9 are much too far from average; the SD can't be 6, because none of the numbers are more than 6 away from the average.

7. (a) Average weight of men = 66 * 2.2 . 145 pounds, SD . 20 pounds. Average weight of women . 121 pounds, SD . 20 pounds.

(b) 68%: the range is average plus or minus SD

(c) bigger than 9 kg: if you take the men and women together, the spread in weights goes up.

10. (a) 163, the average.

(b) $8, because the SD is 8; see p.68.

Chapter 5

4. (a) 20%

(b) 12%

6. No. For example, the normal curve says that about 16% of the scores should be more than 1 SD above average, and none are.

8. (a) True. See pp.92ff.

(b) False: all the deviations from average stay the same.

(c) True.

(d) True: all the deviations from average are doubled.

(e) True.

(f) False: all the deviations from average have their signs changed, but that goes away in the squaring. The SD has to be positive (or, exceptionally, zero).

Chapter 6 Special Review Exercises

6. (a) (600 + 650)/2 = 625

More than 125. When you put the men and women together, the spread goes up, because the two distributions are different.

15. The experimental comparison is treatment (groups B + C) versus controls (group A). The investigator is making observational comparisons, because treatment cases select themselves into Group B or C. Cases that opt for a pretrial conference are probably different from cases that don’t so there will be lots of confounding. Similar issues came up when comparing the consent and no-consent groups in the Salk vaccine field trial in Chapter 1, the adherers and non-adherers in the clofibrate trial in chapter 2, or the examined and refused groups in the HIP trial (pp. 22-23). Furthermore, the investigator does not seem to be presenting all the data. For example, there were 2954 cases in all; but only 2780 are reported in the first table. Also, 22% of 701 or approximately 154 of the group B cases reached trial, according to that first table; but the second table on reports on 63 cases in group B. There is a similar issue for groups A and C.

DISC (Lab manual)

4. 6