The media’s coverage of mass public shootings in America: fifty years of newsworthiness

ABSTRACT

This study uses a media distortion analysis to examine the New York Times coverage of mass public shooting incidents occurring in the United States from 1966 to 2016. A comparison between media coverage and actual incidents is used to identify the characteristics influencing the newsworthiness of mass public shootings. This work expands the breadth and depth of media and mass public shooting research, strengthening the validity of previous findings, and identifying new characteristics influencing newsworthiness. Findings indicate significant predictors of newsworthiness include higher casualties and injuries, as well as perpetrators that are young, Middle Eastern, and ideologically motivated. School shootings are more likely to receive any coverage, and all non-workplace shootings receive salient levels of coverage. An incident is also more likely to receive any coverage if a combination of weapons is used. These findings have important implications for public knowledge and perceptions of mass public shootings.

KEYWORDS:

• Mass public shooting
• mass media
• media distortion analysis
• school shooting
• terrorism

Notes

1. This research used a similar data collection strategy as Schildkraut et al. (2017), but did not use the same data set.

2. See Schildkraut and Elsass (2016) for an in-depth discussion of the debate over defining mass shootings.

3. Given the nature of the dependent variables (i.e. ratio level), ordinary least squares regression would normally be sufficient for the analysis. However, the distribution of the dependent variable (news salience) is severely skewed, which can threaten the validity of the estimates by dramatically changing the magnitude of regression coefficients.

4. The iterations for models with the number of general articles and words would not converge and therefore were omitted from the results. This problem is the result of the large share of general articles/words written about extreme cases. This large proportion affects weight assignments and the coverage’s of Huber iterations.

Additional information

Notes on contributors

Jason R. Silva

Jason R. Silva is a Doctoral Candidate in Criminal Justice at John Jay College of Criminal Justice / Graduate Center, CUNY. Silva’s research examines media and crime, specifically focusing on mass shootings and terrorism. He is currently the Project Manager for a National Institute of Justice funded project developing the U.S. School Shooting Database. He has previously worked as a Research Assistant for the Extremist Crime Database and Active Shooter Incident Project. His research has been published in the Security Journal.

Joel A. Capellan

Joel A. Capellan is an assistant professor in the Law and Justice Studies department at Rowan University. Capellan specializes in statistics and spatial analysis. Substantively, his research interests are broad.  He has conducted and published research on state-sponsored repression, segregation, lone wolf terrorism, policing bias, and criminological theories. Currently, he is devoting most of his attention to the study of mass public shootings.