In regression analysis the dependent variable is often not influenced just by ratio scale variables but also by variables that are qualitative or nominal scale that are not measurable in quantitative terms such as sex, race, and nationality. Indeed, Dummy variables are used to capture the effects of these qualitative factors.
There are numerous social science applications in which dummy variables play an important role. For example, any regression analysis involving information such as race, age group, would use dummy variables. For example, holding all other factors constant, female workers are found to earn less than their male counterparts. This pattern may result from sex discrimination, but whether the reason, qualitative variable such as sex seem to influence the regressand and clearly should be included among the explanatory variables, or regressors.
Dummy variables indicate the presence or the absence of an attribute, such as male or female. One way we could “quantify” such attributes is by constructing artificial variables that take on values of 1 or 0, 1 indicating the presence (or the possession) of that attribute and 0 otherwise. There are two types of dummy variables: intercept dummies and slope dummies.
- Intercept dummy variables pick up a change in the intercept of the regression
- Slope dummy variables that pick up a change in the slope of the regression
For example :
- You are investigating the impact of years of work experience on earnings. In your sample you have information on a number of individual characteristics including gender and type of occupation (manual, non manual, or mix)
D1i=1 if the individual is female, 0 otherwise.
D2i=1 if the individual has a manual occupation, 0 otherwise.
D3i=1 if the individual has a non-manual occupation, 0 otherwise