To create a valid logistic regression model, the data type for the attribute you wish to predict must be ______.

• A. Binary
• B. Numeric
• C. Categorical
• D. Text

Answer: (A)

Logistic regression is a method built for predicting a probability of one class in a two-class (binary) outcome. It works by modeling the log-odds of the event happening as a linear combination of the input features and then squashing that result with the logistic function to produce a probability between 0 and 1. Because it is fundamentally a binary classification tool, the target variable it predicts must be dichotomous—two possible outcomes such as yes/no or 0/1. If the target were a numeric value, you’d use linear regression instead, and if there are more than two classes, you’d turn to multinomial logistic regression or another multi-class method. So the data type the model is designed to predict is binary.