ABSTRACT
So far you have learned one test statistic and a confidence interval. The z test statistic was a great way to introduce you to hypothesis testing because we could use it in an uncomplicated example, such as our story of the rats that seemed different from the usual healthy rats (in the version of the story with no direction predicted). Unfortunately, we doubt you ever will find this z test statistic in a scientific journal article. It is an unrealistic statistic for two reasons. First, we had to know the numeric values of both the population mean and population standard deviation, which in reality we rarely can know. Second, we are limited to a simple hypothesis. For example, the rat shipment example had a null hypothesis that said the sample came from a population with a mean equal to 33 seconds. Research hypotheses usually are more nuanced, leading to more complex ways of setting up studies and therefore more complex statistical analyses.
