On Determining Sample Size

Another frequent question researchers pose to themselves is, "How large must my sample be?" In order to answer this question, the researcher often must estimate the very quantity that the study is being designed to obtain.

Sample size depends on the standard error of the estimate, the desired precision, and the desired degree of confidence. The tools available here for determining sample size are for descriptive studies; studies involving inferential tests such as ANOVA designs are more complicated and take into account TypeI and TypeII error levels.

Like many areas of research, there is a trade-off inherent between sample size and accuracy, since the two are integrally related. This idea is best illustrated through an example:

Form to determine sample size for proportion data

Form to determine sample size for continuous data