1.2. A researcher observes that the math scores of students…

1.2. A researcher observes that the math scores of students who attend charter schools in a district are 3.2% higher than the math scores of students who attend public schools in the district.  The researcher concludes that math scores would increase by 3.2% if all students in the district attended charter schools.  Using your knowledge of the bias decomposition in the standard estimator, give the two main reasons for why the researcher is likely wrong.

3.5. Unbeknownst to the sociologist, the trajectories of ave…

3.5. Unbeknownst to the sociologist, the trajectories of average happiness among the women who do, and the women who don’t, have a child are as depicted in the figure below. Here, F marks the timing of the first birth among women who do have a child during the observation period. The circles mark the timing of the happiness measures. Briefly state whether the DiD estimate for the effect of having a child on happiness would be positive or negative. Is this estimate credible?   

1.11. A researcher is interested in the effect of attending…

1.11. A researcher is interested in the effect of attending private school (vs public school) on students’ test scores.  He is worried that the effect of school choice on test scores is confounded by a number of unobserved factors, such as parental ambition.  The researcher also believes that driving distance between home and the nearest private school is randomly distributed in the population and has no direct effect on test scores.  If we believe these assumptions, what technique should the researcher use to estimate the effect of school choice on test scores, and why?