4.12 Lab 3: Birthweight and Smoking

For this lab, we’ll use the data Birthweight_Smoking and study the relationship between infant birthweight and mother’s smoking behavior.

  1. Run a regression of \(birthweight\) on \(smoker\). How do you interpret the results?

  2. Use the datasummary_balance function from the modelsummary package to provide summary statistics for each variable in the data separately by smoking status of the mother. Do you notice any interesting patterns?

  3. Now run a regression of \(birthweight\) on \(smoker\), \(educ\), \(nprevisit\), \(age\), and \(alcohol\). How do you interpret the coefficient on \(smoker\)? How does its magnitude compare to the result from #1? What do you make of this?

  4. Now run a regression of \(birthweight\) on \(smoker\), the interaction of \(smoker\) and \(age\) and the other covariates (including \(age\)) from #3. How do you interpret the coefficient on \(smoker\) and the coefficient on the interaction term?

  5. Now run a regression of \(birthweight\) on \(smoker\), the interaction of \(smoker\) and \(alcohol\) and the other covariates from #3. How do you interpret the coefficient on \(smoker\) and the coefficient on the interaction term?

  6. Now run a regression of \(birthweight\) on \(age\) and \(age^2\). Plot the predicted value of birthweight as a function of age for ages from 18 to 44. What do you make of this?

  7. Now run a regression of \(\log(birthweight)\) on \(smoker\) and the other covariates from #3. How do you interpret the coefficient on \(smoker\)?