The Red, White and Biased? - Media Framing of the U.S. Opioid Epidemic Based on News Outlets’ Political Partisanship – A Comparative Analysis of California and Texas

Abstract

Objective: This thesis investigates potential differences and similarities in media framing of the U.S. opioid epidemic in regard to partisan media bias in 16 liberal, centrist and conservative news outlets in the country’s two most populous states, California and Texas. It takes state-level differences related to demographics, state governments and laws, prescription policies or health insurance into account. The results of the hypothesis testing will then be assessed in the context of the two contrasting states. The aim of this procedure is to be able to draw conclusions whether one or both of the above-mentioned factors can be ruled out as as an explanation for possible patterns in media coverage, possibly hinting towards a potential impact of partisan media bias or state differences on framing and thus opening up a debate for further research. Theory: The analysis is based on the framing theory by Robert Entman (1993), which states that issues can be viewed from a range of perspectives and that individuals arrive at opinions or reconsider their attitudes based on the way a matter is portrayed in the media. Individuals are often guided towards pre-determined evaluations of issues by the media’s focus on specific issue attributes. Frames are thus tools helping human beings to grasp the world and their surroundings (Lecheler & de Vreese, 2019:7). Method: A comparative quantitative content analysis was conducted using five generic frames established by Semetko and Valkenburg (2000) and four issue-specific frames identified through the review of previous research on media narratives in the opioid crisis. The collected data was analyzed using the statistical tests of Pearson’s chi2, Cramér’s V and cross-tabulations. Intracoder reliability was ensured through Cohen’s kappa values. Material: The data was collected from March 30th, 2024 to April 12th, 2024 and retrieved through a search of pre-defined keywords from the news outlets’ respective websites and archives. The sample (n=401) includes articles published between March 1st, 2023 and July 31st, 2023, the time frame around the mortality peak of the crisis to date, May 2023. Only news items, no other journalistic text types, making clear reference to the current opioid crisis in the United States, were coded. The analysis focused on the textual parts of the articles only. Results: During the hypothesis testing, significant variations in media framing were found between newspapers of different political leanings. Conservative media use the criminalization narrative more frequently than liberal ones, while those make use of the medicalization, human interest and conflict frame more often than their conservative counterparts. Liberal and conservative news outlets do not differ in the use of the stigma frame and their portrays of victims and perpetrators within the crisis. When contextualizing these results in the Californian and Texan samples, the outcomes remain the same, indicating that the differences found may be due to the media’s partisan bias and that state-level contrasts investigated in this thesis can be dismissed as an explanation for the patterns in this case.