A Market-Based Model of State Supreme Court News: Lessons from Capital Cases

Abstract

In this article, we present and test a market-based model of news content about state courts of last resort. We test our theory by examining newspaper coverage of decisions in death penalty cases. Our empirical results indicate that news elements of drama, novelty, and sensationalism influence coverage of state high courts' death penalty cases rather than traditional indicators of legal salience. News content either anywhere in a newspaper or on its front page is influenced by similar factors, but front-page coverage is more sensitive to dramatic conviction reversals and the rarity of executions in a given state. Our results suggest that traditional explanations of the relationship between crime and newsworthiness have limited impact on media attention given to state supreme courts. Media coverage of state high courts is instead associated primarily with the behavior of the court and its justices.

A Social Media Platform Model of Supreme Court News

“…With this, outlets often restrict their coverage to only those cases with robust degrees of latent salience that would promote readership. Collectively, this dynamic underpins the economic model of Supreme Court news (Vining and Marcin 2014;Vining, Wilhelm, and Collens 2015).…”

“…A final question, and perhaps most important, is discerning how media-affiliated accounts decide what to cover. Through the traditional lens of the economic model (Hamilton 2004;McManus 1988;Vining and Marcin 2014;Vining, Wilhelm, and Collens 2015), choices to cover the Supreme Court often lead to sensationalized reporting and only provide for cases that can facilitate engagement with audiences (Krewson, Lassen, and Owens 2018;Zilis 2015). Considering how Americans are increasingly likely to view the Court and its Justices through a partisan lens (Pew Research Center 2022a, 2022b; Vining and Bitecofer 2023), it is unsurprising that popular media coverage of decisions would be most pronounced when they concern prominent social or political issues.…”

Estimating logit models with small samples

“… 6 King and Zeng (2001), Zorn (2005), and Rainey (2016) each mention, but do not focus on, the small-sample properties of PML. Although we are aware of many published papers that use PML to address the problem of separation (e.g., Barrilleaux and Rainey, 2014; Leeman and Mares, 2014; Bell and Miller, 2015; Vining et al , 2015), and one published paper (Chacha and Powell, 2017) and one unpublished paper (Kaplow and Gartzke, 2015) that use PML to address the related problem of rare events (King and Zeng, 2001), we are aware of no published study in political science using PML for bias reduction specifically in small samples in logit models. We are aware of one published paper using PML for bias reduction: Betz (2018) uses PML to estimate a Poisson regression model.…”