1. Download the childIQ data. Consider multiple linear regression where child IQ is the dependent variable and mother IQ is the predictor variable. Recall that in the model lm(child.iq ~ mother.iq) that the regression coefficient is 0.69. This means that a +1 IQ point increase for mom is associated with a +0.69 increase for the child.
Re-scale mother IQ so that the mean is 100 and the standard deviation is 15. Call this new variable mother.iq.rescaled.
Fit a linear model with child.iq as the dependent variable and mother.iq.rescaled as the predictor variable. Summarize and intepret the results.
Answer
2. Interactions in the childIQ data. Consider multivariable regression where child IQ is the dependent variable, and the predictor variables are mother.age and mother.highschool.
a. Is there a significant interacrtion between mother.highschool and mother.age?
Answer: _________
b. Estimate the effect of mother.highschool on child.iq when mother’s age is equal to 30.
Answer: _____________
c. Using the fitted model from question 2b, calculate the mean child IQ when mother’s age is equal to 30 and when the mother did not complete high school.
Answer: ___________