rtburg / prosecutors_and_politics_media_pilot

Media Coverage of Prosecutors and Their Elections: Results of a Pilot Study

Media Coverage of Prosecutors and Their Elections: Results of a Pilot Study

Part of the the Prosecutors and Politics Project at the UNC School of Law

Full report: https://law.unc.edu/wp-content/uploads/2023/02/REPORT-FINAL-2.15.23.pdf

Code and repository by Ryan Thornburg ([email protected])

Project goal

From the full report:

"Prosecutors play an important role in the criminal justice system. Their discretion to decline charges and to plea bargain give prosecutors the power to greatly affect how many people go to prison and for how long. Politics --- specifically elections --- are one of the few checks on prosecutorial power. Voters can vote out local prosecutors who abuse their power or who do not act in the public interest.

"The ability of elections to serve as a check on prosecutors' power depends on what voters know. If voters are unaware of or uninformed about what their local prosecutor is doing, then they cannot make informed choices about whether to retain an incumbent prosecutor or whether to vote that prosecutor out of office. But most voters do not have the time or the ability to assess their prosecutors' performance; instead, they rely on media outlets to inform them. ...

"This pilot study aims to improve the discussion surrounding media coverage of the criminal justice system more generally and of prosecutors in particular. It provides a glimpse into the quantity and the quality of the media coverage that prosecutors receive during an election year in print news articles, as well as a limited number of local and national television news broadcasts. Specifically, the study quantifies the amount of media coverage, the content of coverage, and the tone of coverage that prosecutors and candidates for prosecutor receive. The study includes data from a sample of five to 10 prosecutor elections in four different states. In total, the study examines 27 prosecutor elections in jurisdictions of varying populations; some of those elections were contested, while others were uncontested. In the contested elections, the study also includes information about the coverage that candidates for the office of prosecutor received. All told, the study examined nearly 2,000 articles--- every article that mentioned the elected prosecutor or a candidate for the office in the calendar year 2020."

Project notes

Staff involved

Questions about this report should be directed to the PPP director, Professor Carissa Byrne Hessick at [email protected].

Associate Professor Ryan Thornburg, UNC School of Journalism and Media, played a key role in research design. He also performed the data analysis and wrote portions of the report.

Amy Ullrick did significant work obtaining data, refining spreadsheets, and managing the data flow for the entire project.

This report would not have been possible without the hard work of many students at the University of North Carolina School of Law, including Michael Griffith, Kate Kozain, Meighan Parsh, Abigail Perdew, Lydia Shelley, Jacob Showers, Gabrielle Supak, Tyler Ventura, Anna Washa, and Rachel Weisz. These students read through and coded thousands of news articles and contributed research necessary to produce this report.

Data source

See Methodology section of full report for data collection methods.

Data is published in the public domain on UNC Dataverse hosted by the Odum Institute for Research in Social Science. Dataverse Community Norms as well as good scientific practices expect that proper credit is given via citation.

Citation: Hessick, Carissa; Thornburg, Ryan, 2023, "Pilot Study of Media Coverage for Prosecutors and Prosecutor Elections, 2020", https://doi.org/10.15139/S3/3SWIKQ, UNC Dataverse, V1

Technical

This project skeleton was built using AP DataKit and is based on the AP R Cookiecutter project template.

Project setup instructions

After cloning the git repo:

Open prosproj_github_repo.Rproj in RStudio.

source("etl/load_data.R") will pull two spreadsheets from the UNC DataVerse and create one dataframe called incumbents and another for all the candidates who were not incumbents.

This calls etl/competitive_seats.R, which will create three data frames -- one for incumbents, one for non-incumbents and one combined -- that shows for each contest the number of incumbent candidates, the number of non-incumbent candidates and whether the seat is "uncontested" (an incumbent with no challengers), "contested" (at least one incumbent and one challenger), or "open" (no incumbent).

It also calls etl/binding_incumbents_challengers.R, which creates data frame called all_mentions. This data frame includes all variables that the "incumbent" and "non-incumbent" data frames have in common.

Data Analysis Notebooks

• Executive Summary
• Amount of Coverage
• Content of Coverage
• Tone of Coverage