Firing Blanks? The following is analysis from my senior undergraduate thesis from 2015 (when I was an Anteater; Zot Zot) examining the effect of California's assault weapon's ban on crime rates. This paper uses a modified difference-in-difference estimator. The treatment group is Non-CA states, who let the federal assault weapons ban in 2004 lapse while California maintained a state assault weapons ban, serving as our control. The data examines per capita crime, various controls, as well as treatment/control and pre/post treatment for counties that lie exactly on the border of CA with Nevada, Oregon, and Arizona. I estimated the following regression: In(Murder)it = -8.36-0.02Povertyit - 0.16Treatit - 0.40 Postit (0.03) (0.5) (0.012) -0.06CountyUnemploymentit - 0.14(POST × Treat) (0.35) (5.57) (0.08) In(Murder) the logged murder rate per 10,000 people. Poverty is the percentage of the population within each county that is below the Federal poverty line. Treat is a binary variable = 1 for states that let the federal assault weapons ban expire (non-CA counties), O for CA. Post is a binary variable = 1 for years after the federal assault weapons ban expire. (2004 onward), or for pre-2004. County Unemployment is the unemployment rate. The interaction term, Post X Treat measures the interaction between counties within the treatment states. That is, the effect of being a Round to 2 decimal places and include a negative sign if necessary. A. Should I have run RTO instead? Assume at the 95% confidence level and a Z distribution. Answer "YES" or "NO". B. What is the T-statistic associated with the unemployment rate? Round to 2 decimal places and include a negative sign if necessary. C. What is the t-critical value associated with the 95% confidence level and a Z distribution? Round to 2 decimal places. D. What is the effect of being a treated state, before the treatment year? That is, what is the effect of being a county that removed an assault weapons ban before the year the ban expired. Include a negative value if the answer is negative, round to 2 decimal places. E. Is unemployment rate a determinant of the murder rate in this data? Answer "YES" or "NO" F. Is the treatment effect statistically significant at the 95% confidence level? That is, did removing the assault weapons ban have a statistically significant effect on the murder rate? Hint: Think about which coefficient we care about in a diff-in-diff set up. Answer "YES" or "NO". G. Does running this as a two-way fixed effects model make sense? Answer by inputting the number that is associated with the correct statement:

The authors of a paper describe an experiment to evaluate the effect of using a cell phone on reaction time. Subjects were asked to perform a simulated driving task while talking on a cell phone. While performing this task, occasional red and green lights flashed on the computer screen. If a green light flashed, subjects were to continue driving, but if a red light Flashed, subjects were to brake as quickly as possible. The reaction time (in msec) was recorded. The following summary statistics are based on a graph that appeared in the paper. n = 49 x = 530 S = 65 USE SALT (a) Construct a 95% confidence interval for u, the mean time to react to a red light while talking on a cell phone. (Round your answers to three decimal places.) Interpret a 95% confidence interval for μ, the mean time to react to a red light while talking on a cell phone. We are 95% confident that the mean time to react to a red light while talking on a cell phone is between these two values. There is a 95% chance…

20 Total 58 11. The following summary table presents the results from an ANOVA comparing four treatment conditions with n = 10 participants in each condition. Complete all missing values. (Hint: Start with the df column.) 15 Source SS df MS Between treatments 10 F =_ Within treatments Total 16. 174 opment of language skills from age 2 to age 4. Three Lis en are obtained, one for each 12. A is the devel-

Education influences attitude and lifestyle. Differences in education are a big factor in the "generation gap." Is the younger generation really better educated? Large surveys of people age 65 and older were taken in n1 = 38 U.S. cities. The sample mean for these cities showed that x1 = 15.2% of the older adults had attended college. Large surveys of young adults (age 25 - 34) were taken in n2 = 37 U.S. cities. The sample mean for these cities showed that x2 = 18.1% of the young adults had attended college. From previous studies, it is known that ?1 = 6.4% and ?2 = 4.8%. Does this information indicate that the population mean percentage of young adults who attended college is higher? Use ? = 0.05. What is the value of the sample test statistic? (Test the difference ?1 − ?2. Round your answer to two decimal places.)=__(c) Find (or estimate) the P-value. (Round your answer to four decimal places.)=__