In real-life data, we will have to deal with various types of categorical variables. In this chapter, we discussed different types of categorical outcome variables. Here, we discuss the models for binary and count outcome variables. In case of binary (dichotomous) dependent variable, a binary logistic regression model is used to investigate the relation of the outcome variable with the explanatory variable(s). For dependent or response variables with more than two categories, a multinomial logistic regression model is used. In case of an ordered dependent variable with more than two categories, an ordinal logistic regression model is more appropriate. Another type of commonly used dependent variable in health research is count. A Poisson regression model or a negative binomial regression model is used for count outcomes. These models together with examples are described in the upcoming sections.