ABSTRACT

This chapter considers the issue of accuracy before looking at how one should proceed with choosing an appropriate sample design. Usually, one of the aims of a survey is to estimate what are referred to as population parameters, such as the mean average value for a particular variable. In other words, one may want to use the mean from the sample as an estimate of the mean for the population. For example, in the library survey, one may wish to measure the mean number of loans per book per year from the sample and use this as an estimate for the whole library. People can never say exactly, because they will never know the true mean value for the population. However, if they use one of the sample designs that involves random selection then they can use statistical theory to calculate various statistical measures such as standard errors and confidence intervals.