Hypothesis Testing

Once you have data, the question people actually want answered is rarely “what is the parameter”; it is “is this difference real, or is it just sampling noise?”. Hypothesis testing turns that instinct into an executable procedure: state the null and the alternative, choose a test statistic and a significance level, measure how far into the rejection region the observed data falls, and decide. This chapter walks the procedure end to end and applies it separately to a single normal population and to two normal populations.