Basic Statistics

236 ANSWERS TO SELECTED PROBLEMS

Chapter 2

2.1 No answer.

2.2 No answer.

2.3 It may not appear to be random since there are more odd numbers than even

numbers and there were six odd numbers in a row. But the numbers in this table

are in a random order and this table has been carefully tested. What counts is

how the numbers were generated. No, you will not have equal numbers in the

two groups if 50 is the total sample size. Often when random assignment is

used, the sample size in the two treatment groups does not come out the same

unless a method of forcing the sample size to be equal is used.

2.4 The easiest to explain is a systematic sample. We have k = 180/25 = 7.2, so

pick a random number between 1 and 7, say 5. The instructions are starting

with the fifth student in the list; take every seventh student from there on. The

student numbers will be 5, 12, 19,26, 33, and so on. Note that 12 = 5 + 7 and

19 = 12 + 7. In other words, you keep adding 7 to the previous number.

2.5 No answer.

2.6 Would use a stratified sample. This would ensure having an adequate sample

size in each class.

2.7 First, sort the records of the cases by age and obtain the records of the women

who were < 50 years old at the time of diagnosis. One possibility is to chose

matched neighborhood controls. Each control would be within, say, 5 years of

age of the case and live in the same neighborhood. Hospital or friend controls

could also be used.

2.8 No answer.

Chapter 3

3.1 No answer.

3.2 No answer.

3.3 Have two professionals write out their results independently and then get to-

gether, compare results, and agree on an answer,

Chapter 4

4.1 The stem and leaf graph can be obtained from the following compilations:

There was no excess of 0’s or 5’s in the leaf columns.

4.2-4.7 No answers,

4.8 Expect the vertical bars to be of similar height.