Comparing Means and Proportions in Two Samples
In chapter 10, we used parametric significance tests to compare the mean or proportion of a single sample with a population goal or parameter. In this chapter, we turn to a more commonly used application of parametric tests of statistical significance: comparisons between samples. Let's say, for example, that you are interested in whether there is a difference in the mean salaries of male and female police officers or in the proportions of African Americans and others arrested last year. Your question in either of these cases is not whether the population parameters have particular values, but whether the parameters for the groups examined in each case are different. This involves comparing means and proportions for two populations. If you take samples from these populations, you can make inferences regarding the differences between them by building on the normal distribution tests covered in Chapter 10.