Dummy variable (statistics)

As a dummy variable (rarely dummy variable or proxy variable, Eng. Dummy variable, even for a short dummy) is known in the statistical data analysis a binary variable with the forms 1 and 0 (yes -no variables) as an indicator of the existence of an expression of a multistage variable serves. This dummy variable underlying variable can have any scale level.

Applications and Examples

For statistical analysis, it may be helpful to know whether an investigation unit has a particular value of a categorical variable or not. To this end, forming a dummy variable with the forms 1 and 0:

Example: At an election poll indicates a categorical variable, would choose which party the respondent. To determine the proportion of CDU voters, one uses a dummy variable with the forms 1 = CDU voters and 0 = not CDU voters.

For the logistic regression analysis, it may be of interest to operationalize the probability for the occurrence of a variable. To provide some statistical programs, the possibility of a categorical variable automatically for any expression a dummy variable to produce. With a total of k occurrences of a categorical variable, in principle, k dummies are conceivable. However, it also comes with (k -1) test dummies to fully map the k forms: the (omitted ) k-th occurrence recognized by the fact that all dummies have a value of 0.

In interval- scaled variables, dummies are often used to indicate whether a value is dichotomous below or above a certain limit.

For example, the dummy variable gets the value 1 if the respondent is younger than 50 years, and otherwise the value 0