Dummy Variable

A dummy variable is a binary (0/1) variable created to indicate whether a case belongs to a particular category.  Typically a dummy variable will be derived from a multi-category variable. For example, an insurance policy might be residential, commercial or automotive, and there would be three dummy variables created:

 

 

  • Residential (yes/no)

  • Commercial (yes/no)

  • Automotive (yes/no)

Note that if a policy must belong to one of the three categories, there is duplicative information information in the dummies.  For example, if residential and commercial are both 0, then you know that automotive is 1. Logistic regression methods fail in such cases, if you include all the variables, so you would include only k-1 out of k variables.  

Other terms for binary dummies are indicator and flag, and sometimes the term dummy is applied to a naturally binary variable (e.g. bankrupt/not-bankrupt) that is converted from text to a 0/1 variable.