A variable, taking only the values 0 and 1, derived from a polytomous categorical variable. If the categorical variable has k categories then (k−1) dummy variables are required. For example, with four categories the three dummy variables (x1, x2, x3) could be assigned the values (1, 0, 0) for category 1, (0, 1, 0) for category 2, (0, 0, 1) for category 3, and (0, 0, 0) for category 4. Dummy variables enable the inclusion of categorical information in multiple regression models.