Perform a regression analysis to predict the number of correct words in vocabulary (WORDSUM) on the basis of both television viewing (TVHOURS) and highest year of education (EDUC).
a. Test the null hypothesis that the number of words correct is not related to hours per day of television viewing, holding years of education constant.
b. What are the regression slopes and the intercept?
c. Which variables are significant at the .05 level?
d. Find the multiple coefficient of determination (R²). What does it mean? Did it improve over the amount explained in the two-variable regression model?
e. What is the predicted number of correct words in vocabulary for a person who watches two hours of television per day and has 14 years as their highest level of education? (Remember to look for the values on each of the scales before doing your calculation.)