ABSTRACT
Whereas Chapter 7 showed how to model the association between two or more categorical variables, this chapter is concerned with predicting a categorical response from a set of continuous predictors. Specifically, in this chapter we discuss the logistic regression model, a generalized linear model appropriate for binary outcomes. The logistic regression model is similar to the more familiar linear regression model in that both models predict an outcome (also known as a response or dependent) variable, from a set of predictor variables (also known as explanatory or independent variables). In the case of linear regression, the response or dependent variable is a continuous variable (typically measured on an interval or ratio scale). In the case of logistic regression, however, the response variable is a binary or dichotomous variable, which means it can only take on one of two possible values. For example, whether a student is proficient in mathematics would be an appropriate dependent variable for logistic regression. Both logistic and linear regression models allow for predictor variables that can be continuous or categorical. Because there are interpretational similarities between the logistic and the more familiar linear regression model with continuous predictors, in this chapter we focus on the logistic model with continuous predictor variables.
