Dummy-Variable Regression

← Back to document page

7Dummy-VariableRegressionOne of the serious limitations of multiple- Regression analysis, as presented in Chapters 5and 6, is that it accommodates only quantitative response and explanatory variables. In thischapter and the next, I will explain how qualitative explanatory variables, calledfactors, can beincorporated into a linear current chapter begins with an explanation of how adummy-variable regressorcan becoded to represent adichotomous( , two-category) factor. I proceed to show how a set of dummyregressors can be employed to represent apolytomous(many-category) factor.

Here, gender is a qualitative explanatory variable (i.e., a factor), with categories male and female. The dummy variable D is a regressor, representing the factor gender. In contrast, the quantitative explanatory variable education and the regressor Xare one and the same. Were we to transform education, however, prior to entering

  Gender, Female