Sample Estimates and Sample Distributions

We select a sample because we want information on a particular feature of the target population; for example, the average walking distance of people with peripheral vascular disease or the effect of screening for abdominal aortic aneurysm in men aged 65-74.

Since we have no means of finding this feature directly, we collect information on a random sample, taken from the target population, and use it as our best estimate of the population value.

To clarify the distinction between the population and the sample values we generally use Greek letters like (mu) or (pi) for the population value, and Roman letters like or p for the sample value. Remember from BS02 that the sample mean of a variable is denoted .