Chapter 9 - Confidence Intervals: Admitting That Estimates Are Not Exact

Abstract

This chapter explains the concept of confidence intervals. The process of generalizing from sample data to make probability-based statements about the population is called statistical inference. A confidence interval is an interval computed from the data in such a way that there is a known probability of including the (unknown) population parameter of interest, where this probability is interpreted with respect to a random experiment, which begins with the selection of a random sample. The probability that the population parameter is included within the confidence interval is called the confidence level, which is set by tradition at 95%, although levels of 90%, 99%, and 99.9% are also commonly used. The higher the confidence level, the larger is (and usually less useful) the confidence interval. This chapter explains concepts of the confidence interval for a population mean or a population percentage along with assumptions needed for validity,interpretation of confidence interval, and details of one-sided confidence intervals.