1. Confounding arises in a two-sample t-test when the groups differ in ways other than the labeling that distinguishes the groups.
2. An analysis of covariance is another name for the use of randomization to avoid confounding.
3. A dummy variable is a numerical encoding that as- signs the value + to the members of a category and assigns the value 0 to others.
4. To build the interaction between X and a dummy variable D, we multiply X times D.
5. If the multiple regression implies parallel fits, the slope of the dummy variable is the difference between the two fitted lines.
6. A multiple regression with a numerical predictor and a dummy variable as two explanatory variables implies parallel fits to the two groups.
7. The purpose of an interaction is to force fits in the groups to be parallel.