Confidence Intervals

Point estimators and their measures of reliability are combined to create data-based intervals of plausible estimates of an unknown parameter. Interval estimators are simply rules that tell a researcher how to use the observed sample data to generate an interval of reasonable estimates of the parameter of interest. A confidence interval is an interval estimator that is designed to produce an interval of estimates that captures the unknown value of the parameter being estimated with a prespecified probability. This chapter discusses confidence intervals for the parameters of a single population. It also discusses the confidence interval procedure for estimating population proportions, the population mean, and the bootstrap confidence interval procedure. The chapter considers confidence intervals for the difference of two proportions and confidence intervals for the relative risk.