Regression Analysis by Example, 4th Edition

CHAPTER 5

QUALITATIVE VARIABLES AS PREDICTORS

5.1 INTRODUCTION

Qualitative or categorical variables can be very useful as predictor variables in regression analysis. Qualitative variables such as sex, marital status, or political affiliation can be represented by indicator or dummy variables. These variables take on only two values, usually 0 and 1. The two values signify that the observation belongs to one of two possible categories. The numerical values of indicator variables are not intended to reflect a quantitative ordering of the categories, but only serve to identify category or class membership. For example, an analysis of salaries earned by computer programmers may include variables such as education, years of experience, and sex as predictor variables. The sex variable could be quantified, say, as 1 for female and 0 for male. Indicator variables can also be used in a regression equation to distinguish among three or more groups as well as among classifications across various types of groups. For example, the regression described above may also include an indicator variable to distinguish whether the observation was for a systems or applications programmer. The four conditions determined by sex and type of programming can be represented by combining the two variables, as we shall see in this chapter.

Table 5.1 Salary Survey Data

Indicator variables can be used in a variety ...