The Essentials of Biostatistics for Physicians, Nurses, and Clinicians

32 CHAPTER 2 Sampling from Populations

suggested, 10 is not a large number, and you can see that since the
original sample mean is 33.17, the estimate is off by 1.29 years. A value
of B = 100 or 500 should make the estimate much closer.
Properties of the bootstrap samples to note are the repetitions.
In bootstrap sample 1, E occurs three times and C and D are both
left out. In bootstrap sample 2, E and D each repeat once, and B
and F are left out. Bootstrap sample 9 has only one repetition, and
only B is left out. I that sense it is closest to the original sample,
but its mean is 42.0 compared with the mean of 33.17 for the
original sample. The large difference is due to the fact that the
oldest patient E is repeated and the second youngest is the one
left out.

2.6 EXERCISES

1. Why do we need to collect samples when we want to determine population
   characteristics?


2. Provide a defi nition in your own words for the following terms:

(a) Sample

(b) Census

(c) Parameter

(d) Statistic

3. Describe and contrast the following types of sampling designs. Also, state
   when if ever it is appropriate to use the particular designs.
   (a) Simple random sample
   (b) Stratifi ed random sample
   (c) Convenience sample
   (d) Systematic sample
   (e) Cluster sample
   (f) Bootstrap sample


4. What is meant by parameter estimation?


5. For sample designs (a), (b), (c), and (d) in exercise 3, explain under what
   circumstances bias can enter?


6. How does bootstrap sampling differ from simple random sampling?


7. What is the rejection sampling method and when is it used?