How News Outlets Frame LGBT Rights: Identifying and Differentiating Narratives

• In Event: Poster Session: Political Communication

Abstract

The LGBT rights movement has encountered a striking breadth in its victories and setbacks over the past decade. American society has not reached consensus on LGBT rights and inclusion, and this rift is reflected in divergent media coverage of LGBT political issues. A burgeoning body of scholarship has examined how belief in different attributional narratives of gay etiology has driven evolution and polarization of public opinion. I contribute to this research agenda by exploring the utility of computational content analysis methods to identify the most differentiating and ideologically-laden words and features print news coverage of LGBT issues.

I use generative computational modeling to perform unsupervised clustering of articles from several top-circulation American newspapers. Topic modeling and token-based ideal-point estimation using the WORDFISH algorithm allow the subdivision and relative scaling of news stories, based on compositional differences. Neural network derived word vectorization allows finer grained insight into specific changes in the typical context of ideologically important words and phrases. This succession of statistical methods supports both pre-identified hypothesis testing, and also the emergence of unexpected results, inaugurating new lines of inquiry. These insights reflect broadly on how LGBT political issues are covered in the news, which narratives about gay identity are used most frequently and by who, and how those narratives are framed – between sources and across time.

Author

• Jeff Paul Antsen, Temple University