Workshop Statistics: Discovery with Data, Fourth Edition

TOPIC 21

Comparing Two Proportions

Can swimming with dolphins help depression patients recover from their symptoms? How can a randomized experiment help us decide? Did the proportion of babies who slept on their stomachs decrease between 1992 and 1996? How would collecting data for such a study be different from collecting data for the dolphin study? How do we analyze the results? How do both of these studies differ from those in Unit 4? In this topic, you will learn how to apply the inference principles that you learned in the previous unit to address questions of comparisons of two groups.

Overview

In the previous unit, you discovered and explored techniques of statistical inference for drawing conclusions about population parameters on the basis of sample statistics. All of those procedures applied to a single parameter (proportion or mean) from a single population. In the next two topics, you will investigate and apply inference procedures for comparing parameters between two populations or treatment groups. These techniques are especially important due to the crucial role of comparison in experimental design. The randomization involved in experiments enables you to make inferences about the effects of the explanatory variable on the response variable. These procedures are similarly applicable for comparing two groups that have been randomly selected from two populations. In this topic, you will investigate inference procedures for comparing proportions between two experimental ...