The data file birthweight_smoking.xxs is from the 1989 linked National Natality-Mortality Detail files, which contains a census of infant births and deaths.
Variable Description:
- Birthweight: Birth weight of infant in grams.
- Smoking: Indicator equal to one if the mother smoked during pregnancy and zero otherwise.
- Mother's Attributes:
- Age: Age of the mother.
- Educ: Years of educational attainment (more than 16 years, coded as 17).
- Unmarried: Indicator if mother is unmarried.
- Alcohol: Indicator if mother drank alcohol during pregnancy.
- Drinks: Number of drinks per week.
- Triprel: Indicator of prenatal care visit in trimester 1.
- Tripre2: Indicator of prenatal care visit in trimester 2.
- Tripre3: Indicator of prenatal care visit in trimester 3.
- Nprevist: Total number of prenatal visits.
Regression Models:
1. Birthweight on Smoking.
2. Birthweight on Smoking, Alcohol, and Nprevist.
3. Birthweight on Smoking, Alcohol, Nprevist, and Unmarried.
(a) What is the value of the estimated effect of smoking on birth weight in each of the regressions?
(b) Construct a 95% confidence interval for the effect of smoking on birth weight using each of the regressions.
(c) Does the coefficient on Smoking in regression (2) suffer from omitted variable bias? Explain.
(d) Consider the coefficient on Unmarried in regression (3). Construct a 95% confidence interval for the coefficient. Is the coefficient statistically significant? Explain.
(e) Is the magnitude of the coefficient large? Explain.
(f) An advocacy group notes that the large coefficient suggests that public policies that encourage marriage will lead, on average, to healthier babies. Do you agree?
