Answer Happy

Consider the model of birth defects discussed in class.
Specifically, • Y = baby’s birth weight (in pounds) • X =1 if
mother drinks coffee during pregnancy 0 if not • Z =1 if mother
uses cocaine while pregnant 0 if not.
Unlike the real world, suppose we know exactly how birth defects
are determined. Specifically, • Y = 8 − 5Z •
10% of mom’s who drink coffee also use cocaine • .
1% of mom’s who don’t drink coffee use cocaine Suppose we
collect data on 100 thousand mothers, half of which drank coffee
during pregnancy.
Problem 2. Compute the average birth weight of babies born to
mothers in each of the following groups: (1) coffee drinkers and
cocaine users, (2) coffee drinkers and non-cocaine users, (3)
non-coffee drinkers and cocaine users, (4) non-coffee drinkers, and
non-cocaine users. What is the difference in average birth weight
among groups (1) and (3)? (2) and (4)? What does this tell us about
the effect of coffee consumption on the baby's birth weight?
Problem 3. Instead of collecting observational data, suppose we
randomly assigned half of the mothers to drink coffee and the other
half to abstain. What would be the average birth weight in each
group?