Introduction

In this assignment, you will demonstrate your ability to understand and conduct a difference-in-differences analysis by replicating the main analysis from the article “Violence and Voting in the United States: How School Shootings Affect Elections” (García-Montoya et al., American Political Science Review, 2022)

Abstract

Formalities

Your assignment must be a maximum of 18,000 characters.

All of the requirements below are standard in basically all working papers in political science and for submission to journals, so please stick to the following conventions:

  • 12-point font
  • 1 inch margins
  • Double-spaced
  • No Table of Content
  • 18,000 characters maximum

If you write your assignment in LaTeX (e.g. Overleaf) it’s therefore:

  • \documentclass[12pt, A4]{article}
  • \usepackage[margin = 1in]{geometry}
  • \doublespacing after your \begin{document}

If you want to use the Overleaf template that I use for my own academic manuscripts, you can find it here.

General expectations

Please note that although the technical side of your assignment should be well-executed, you should also focus your energy on the clarity of your writing. The reason that students often do not receive the grade that they might hope for is because the written part of their assignment is unclear. So please be sure to spend sufficient time ensuring that you clearly write up the goal of the study, the research design, the results, and the discussion.

Instructions

Data: Assignment_2B_Data.zip

Note 1: The authors use time (year) and unit (fip_code) fixed effects in all of their models. They cluster their standard errors by state (fip_state), and not fip_code. The numerical results in your assignment should exactly match the numerical results in the paper that you are replicating.

Note 2: The Stata code for the replication package is available in the replication data above. You are free to look at it to see how the authors run their models. I don’t know Stata myself, but still found it useful to look at the code when I replicated this myself.

Codebook (for the variables you will use only):

year: Year
fip_code: County ID
fip_state: State ID
n_deaths: Number of deaths from shootings in a given county-year
shooting: Number of shootings in a county-year
t_shooting: Is there a school shooting (in a given county-year)
t_fatalshooting: Is there a fatal school shooting (in a given county-year)
t_non_fatalshooting: Is there a non-fatal school shooting (in a given year in a county)
turnout: Presidential turnout
DemVotesTotalPercent: Democratic Party percent of the vote
population_total: Population
prop_non_white: Proportion of the population that is non-white
change_unem_rate: Unemployment rate
neighbor_shooting: Was there a shooting in a neighboring county in a given year
ever_shooting: Does a given county ever have a shooting at any year in the data

Your assignment must proceed as follows:

  1. Read the article: “Violence and Voting in the United States: How School Shootings Affect Elections”.
  2. For your Discussion section at the end, it will be very strongly recommended that you read “Navigating Potential Pitfalls in Difference-in-Differences Designs: Reconciling Conflicting Findings on Mass Shootings’ Effect on Electoral Outcomes” (Hassell & Holbein, Forthcoming)
    - As you will see after reading Hassell & Holbein (Forthcoming), I’ve chosen the article by García-Montoya et al. for replication for an important pedagogical reason.

Now, do the following for your written assignment and accompanying R code:

Introduction

  1. At the beginning of your assignment, briefly explain the goals of the paper:
    - What is the authors’ research question?
    - Why is it important?
    - What are their hypotheses?

Research design

  1. Explain the authors’ research design:
    - The empirical strategy and data
    - The identifying assumptions
    • i.e. how does their research design allow them to estimate a causal effect and what are the potential threats to causal inference?
    • Note: It should be clear from this section that you know how a difference-in-differences research design works.
    • Note: The authors suggest that school shootings need to be as-if random for a difference-in-differences design to identify the causal effect of school shootings: “Our research design treats school shootings as natural experiments. That is, we assume that the timing of a school shooting in a particular county is as-if random.” However, as-if randomness is not the identifying assumption of a difference-in-differences design. It’s nice to have as-if randomness—it can bolster the main assumptions in a difference-in-differences design—but it’s not required. In other words, treatment can be non-random and the design still be valid.

Analysis

  1. Replicate Figure 2 from the article.
    - Explain what the figure is showing and why the authors are showing it.
    - You are welcome to split this figure into, say, two side-by-side figures, one for each variable (number of shootings & number of deaths)
    - Your figure might look something like this, but make it look however you want:

Example figure

  1. Replicate Figure 3 from the article.
    - Note: The three “County-time FE” models include only rows in the data where change_unem_rate, population_total, and prop_non_white are not missing. i.e. the data you put into the model should exclude any row where any of those variables has a missing value
    - Explain the figure’s purpose and the results
    - Your figure might look something like this, but make it look however you want:
    Example figure
  2. Replicate Figure 4 from the article.
    - Explain its purpose and the results.
    - Note: The three “County-time FE” models include only rows in the data where change_unem_rate, population_total, and prop_non_white are not missing.
    - Your figure might look something like this, but make it look however you want:
    Example figure
  3. Create a new figure that is an event study of the effect of school shootings on Democratic share. As the treatment, use the first time that a shooting (t_shooting) occurs in a given county (fip_code). In other words, the treatment should turn on (and stay on) in the first year that a shooting occurs in that county.
    - This figure should be analogous to or the same as Figure 4 in Hassell & Holbein, Forthcoming
    - Explain the purpose of this figure and what it means for the results presented in the García-Montoya, Arjona, and Lacombe (APSR 2022).
    - Feel free to read Hassell & Holbein (APSR Forthcoming) for further explanation for this figure, although this is covered in class.
    - It might look something like this (as in Hassell & Holbein (Forthcoming), I fold observations whose time-to-treatment period is “9” and “10” into period “8”, by recoding “9” and “10” to “8”; you can leave these as if you want, however):
    Example figure
  4. OPTIONAL STEP: Whether you complete this step will not affect your grade.
    As is clear from the data, counties experience school shootings in different years (the timing of the treatment is “staggered”). Create an event study figure using a staggered difference-in-differences model. As in Step 4 above, assume treatment turns on and stays on from the very first shooting in a county.
    - Explain the results of this figure and very briefly explain the reason why using a staggered difference-in-differences model is important.
    - Note: In principle, you could also calculate the ATT (average treatment effect on the treated). However, given the results of the event study graph, you won’t believe the ATT estimate anyway, so don’t bother.
    - Your figure might look like this (given how the R library did works, “0” is the first year of treatment. This is unlike in Step 4, where my figure has the first treatment occurring in time-to-treatment period “1”):
    Example figure

Discussion

  1. Explain the overall findings of the article. One paragraph should suffice.
  2. Discuss the benefits and drawbacks of the research design as it is applied in the article.
    - You may want to read through “Navigating Potential Pitfalls in Difference-in-Differences Designs: Reconciling Conflicting Findings on Mass Shootings’ Effect on Electoral Outcomes” (Hassell & Holbein, Forthcoming).

Submission instructions

  1. Formatting requirements!

    • 12-point font
    • 1 inch margins
    • Double-spaced
    • No Table of Contents
    • 18,000 characters maximum

  2. Your code should be commented so that it is clear to me that you know what each piece of code does. You don’t need to be excessive, but comment your code in a way that you think is reasonable.

  3. If you are in a group, only one of you needs to submit the assignment. Just ensure that all of your student IDs are on the title page.

  4. Submit your assignment and R code separately. Please send your assignment as a PDF and use an equivalent file name across the files:

    • Assignment_2B.pdf
    • Assignment_2B.R

  5. Submit through Absalon under “Assignment 2B”.