In 1985, neither Florida nor Georgia had laws banning open alcohol containers in vehicle passenger compartments. By 1990, Florida had passed such a law, but Georgia had not. (i) Suppose you can collect random samples of the driving-age population in both states, for 1985 and 1990. Let arrest be a binary variable equal to unity if a person was arrested for drunk driving during the year. Without controlling for any other factors, write down a linear probability model that allows you to test whether the open container law reduced the probability of being arrested for drunk driving. Which coefficient in your model measures the effect of the law? (ii) Why might you want to control for other factors in the model? What might some of these factors be? (iii) Now, suppose that you can only collect data for 1985 and 1990 at the county level for the two states. The dependent variable would be the fraction of licensed drivers arrested for drunk driving during the year. How does this data structure differ from the individual-level data described in part (i)? What econometric method would you use?

Question

In 1985, neither Florida nor Georgia had laws banning open alcohol containers in vehicle passenger compartments. By 1990, Florida had passed such a law, but Georgia had not.  (i) Suppose you can collect random samples of the driving-age population in both states, for 1985 and 1990. Let arrest be a binary variable equal to unity if a person was arrested for drunk driving during the year. Without controlling for any other factors, write down a linear probability model that allows you to test whether the open container law reduced the probability of being arrested for drunk driving. Which coefficient in your model measures the effect of the law?  (ii) Why might you want to control for other factors in the model? What might some of these factors be?  (iii) Now, suppose that you can only collect data for 1985 and 1990 at the county level for the two states. The dependent variable would be the fraction of licensed drivers arrested for drunk driving during the year. How does this data structure differ from the individual-level data described in part (i)? What econometric method would you use?

Tuthornt · Accepted Answer

(i) Consider FL be a dummy or binary variable which is equal to one if a person lives in Florida, and otherwise zero. Now, consider y90 be a dummy variable for the year 1990. Then, the linear probability model is, arrest=β0+β1y 90+β2 FL+β3y 90×FL+u The effect of the law is measured by β3 which is the variable of interest here since it describes the probability of drunk driving arrest due to the new law in Florida. (ii) Any factor that leads to different overall trends in both states could be relevant. That is, arrest due to the law of some other exogenous factors or merely an unexplained trend. These can include age, race, education, previous arrest or gender distributions may have changed. These factors are important to be considered as these might affect whether someone is arrested for drunk driving that makes them important to control. At the least, there are the chances of obtaining a more precise estimator of β3 by reducing the error variance. Essentially, any explanatory variable that affects arrest can be used for this purpose. (iii) According to the mentioned set up, the actual arrest rates are present, instead of only a sample, reducing the error from sampling. The interpretation of the coefficients will differ, because they represent averages across counties in a given state rather than state level averages. The individual level data allows the control of individual level variation that can potentially help in reducing the standard errors. The first difference can also be used because of the same set of counties in both years observed at two points in time.

In 1985, neither Florida nor Georgia had laws banning open alcohol containers in vehicle passenger compartments. By 1990, Florida had passed such a la

Answered question

Answer & Explanation

New Questions in College Statistics