CHAPTER 18 A Yes-or-No Proposition: Logistic Regression 249

Chapter 18

A Yes-or-No Proposition:

Logistic Regression

Y

ou can use logistic regression to analyze the relationship between one or

more predictor variables (the X variables) and a categorical outcome vari-

able (the Y variable). Typical categorical outcomes include the following

two-level variables (which are also called binary or dichotomous):»

» ^{Lived or died by a certain date»}

» ^{Did or didn’t get diagnosed with Type II diabetes»}

» ^{Responded or didn’t respond to a treatment»}

» ^{Did or did not choose a particular health insurance plan}

In this chapter, we explain logistic regression. We describe the circumstances

under which to use it, the important related concepts, how to execute it with soft-

ware, and how to interpret the output. We also point out the pitfalls with logistic

regression and show you how to determine the sample sizes you need to execute

such a model.

IN THIS CHAPTER»

» ^{Figuring out when to use logistic}

regression»

» ^{Getting a grip on the basics of logistic}

regression»

» ^{Running a logistic regression model}

and making sense of the output»

» ^{Watching for common issues with}

logistic regression»

» ^{Estimating the sample size you need}

for logistic regression