Violent Crime and the Overmilitarization of US Policing
Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit
The Topic
The topic of the Project is based on the paper by Federico Masera, published in the Journal of Law and Economics in 2021 and available on our Moodle, entitled: “Violent Crime and the Overmilitarization of US Policing.” Your work in the Project will test your ability to explain the main ideas in this paper, to justify its econometric approach, to try to replicate some of its results, and to interpret some of its results.
The Files
Included in the files for the Project are:
county_data.dta (a Stata data file)
poldep_data.dta (a Stata data file)
replication.do (a Stata program file)
project.log (a file containing the Stata output of running replication.do)
Militarization_WEBSITE (the working-paper version of the manuscript, for which the replication.do file produces results)
The Questions to Answer
You will use the files provided and new Stata code that you write to address various dimensions of this research
endeavour, including motivation, method, estimation, and interpretation. For the empirical estimation, you should focus primarily on producing and analysing a version of what appears as “Table 1: Baseline Results” in the
working paper.
(a) What is the motivating reason to conduct this economic analysis? Why is the question important? What has been found previously, and why might our knowledge of this area still be incomplete or inaccurate?
(b) What population is of interest, what structural equation is of interest, and what does the sample of data consist of? You should provide and walk the reader through a table of basic summary statistics about the key
variables.
(c) Why is the IV strategy employed? Thoroughly describe the proposed instrument. Is the proposed instrument defensible conceptually? Empirically?
(d) What does “identification” mean in general, and specifically in this empirical context? Write out and explain all basic equations used to produce the main results.
(e) Show and thoroughly interpret the difference in estimates when you use OLS versus IV to estimate the causal effect of interest. Be sure to address explanatory power, statistical significance, economic significance, and any concerns about violation of the Gauss-Markov assumptions and any potential mis-specifications.
(f) Compare your results for Table 1 with the results in Table 1 of the working paper, produced by replication.do. Did you exactly replicate the paper’s results? If not, what differences do you see, and can you explain them?
The Exposition Format
Writeup your answers to the above questions in the style of an empirical report. Chapter 19 of Wooldridge, about how to proceed with an empirical project, will help you with this. You should limit your report to 8 pages,
excluding the cover sheet, the .do file, and the .log file. Your report should be made up of the following sections:
Introduction
Data
Conceptual Model
Econometric Model
Empirical Results
Comparisons with the Replicated Paper
Conclusions
References
Further Notes
• In preparing your Project, you may find it useful to draw on material covered in lectures up to the end of Week 7. However, you are encouraged to start working on the Project as early as possible in the term. You should work on your own answers and writeup your own Stata .do file and report, but you are encouraged to talk with other
students about the Project. What you submit for the Project should be your own independent work and no one else’s.
• The data required for the Project is provided in two Stata files: county_data.dta and poldep_data.dta. You may
use the Stata do-file “replication.do” as a template to carryout the required computations using these data. The log file produced from running replication.do is also provided (“project.log”). You are not expected to understand all of the Stata commands used in the replication.do file. However, you will find many elements from this .do-file
useful in building your own .do-file to perform your analysis. You may reference the replication.do file to assist
you with constructing variables, but you may not copy and paste code and submit it as your own (e.g., you must
create your own variable names and use different code to include year fixed effects, for example). You ARE
permitted to use the same estimation commands (e.g., reghdfe and ivreghdfe) to conduct your analysis. Please do NOT submit portions of the replication.do file (edited or otherwise) that you do not require in order to perform the empirical work for the Project.
• Parts (a)-(f) guide you through the steps to betaken, some of which require running Stata commands. You should construct your .do-file to carry out the computations required to complete these steps, and then write your answers to all of Parts (a) through (f) in the style of a report, as described above.
• Students who wish to be able to run the entire replication.do program on their Stata installations, which is not
required for completion of the Project, may find that running the following set of Stata commands just once at the start is needed – these commands install various packages that are drawn upon in the advanced commands
contained in replication.do:
ssc install reghdfe
ssc install ftools
ssc install ivreghdfe
ssc install ivreg2, replace
ssc install ranktest, replace
• You will submit the following items in this order, compiled into a single Word or PDF document, with a Cover Sheet (available on our Moodle) as the first page of the compilation:
Your .do file;
Your .log file;
Your project report.
Do not worry about fancy formatting. Simply paste your .do file and .log file output into a format suitable for appending to the front of your project report, directly after your Cover Sheet.
• The University regards plagiarism as a form of academic misconduct, and has very strict rules regarding
plagiarism. In this Project, using the same or very similar text in your work as is contained in the Masera paper will be considered plagiarism. To view UNSW’s policies and penalties, and information to help you avoid plagiarism, see:https://student.unsw.edu.au/plagiarism.
2023-11-09