1. Short answer: What is statistical interaction? (You should be able to answer this question without looking at your notes!)
 

2. Interaction effects between categorical variables

Data: K:\nd.edu\user22\yli\Public\593sp01\Data\ordgss77.dta
    Source: Long, J. Scott. 1997. Regression Models for Categorical and Limited Dependent Variables.
    Content: Data from 1977 General Social Survey on education and occupational prestige.

Research questions:

• What is the effect of race (white versus non-white) on years of education?
• What is the effect of gender on years of education?
• Does the effect of race on education differ between men and women?

In Stata, use ANOVA and regression techniques to answer the questions above. Discuss your findings by

1) interpreting the b coefficents and their significance;
2) testing whether the interaction model is better than the model without the interaction term -- i.e. does the interaction term explains significantly more variation in the dependent variable?
3) predicting the mean education for each race-gender group, using your regression equation; and
4) illustrating the interaction effects of race and gender on education with graphs (in Stata or Quattro).

Hand in your typed discussion with tables, graphs, and Stata log.

Stata Commands to Learn:

anova
regress
generate
egen
graph

* Note: For an example of graphing interaction effects in Stata, check out lab handout on "Graph Interaction in Stata" (http://www.nd.edu/~dmyers/courses/593sp01/interactiongraph1.html).