Read the paper "Using Electoral Cycles in Police Hiring to Estimate the Effect of Police on Crime" by Levitt (1997). You may also want to look at McCrary (2002) and Levitt (2002).
(i) What is the main question that this paper answers?
(ii) What is the equation of interest that the author wants to estimate?
(iii) Why is an instrument needed?
(iv) What do the authors use as an instrument? What two conditions must this variable satisfy for it to be a valid instrument? Can either of these conditions be checked? How? (v) What is the point of Figure 1 and columns 1-3 in Table 2? What should we check in these regressions? Is there any reason for concern?
(vi) What is(are) the estimated effect(s) of police on crime?
(vii) Do you think the findings in this paper are reliable? Refer to the papers by McCrary (2002) and Kovandzic et al. (2016) to justify your answer.
(viii) There are two major flaws in Levitt (1997). What are they? they are discussed in McCrary (2002) and Kovandzic et al. (2016).
Recently Asked Questions
- The probability that a person in the United States has type B+ blood is 10 %. Five unrelated people in the United States are selected at random.
- Cruz Video Center accumulates the following cost and net realizable data at December 31. Inventory Categories Cost Data Net Realizable Value Data
- The table below shows the results of a survey in which 143 men and 145 women workers ages 25 to 64 were asked if they have at least onemonth'sincome set