Figure 9-10: STATA regression output using two interacted
dummy variables.
Segregate and Integrate: Testing for Significance
296
When using J – 1 dummy variables to represent a qualitative
characteristic that has multiple possible outcomes (as
discussed in the earlier section “Juggling multiple
characteristics with dummy variables”), you have to take into
account the collective significance of those variables. Their
effect can be collectively significant even if they are
individually insignificant (I discuss the difference between
individual and joint significance in Chapter 7). In the
following sections you find out two ways to determine
whether your dummy variables have joint significance.
Revisiting the F-test for joint significance
Testing the joint significance of a group of dummy variables
in a regression model is accomplished by generalizing the
F-test of overall significance to