2023 Undergraduate Research Showcase
Dr. Iryna Babik
In recent years, there has been a dramatic increase in polarization of public opinion on some topics of global importance, leading to increasing public reliance on partisan-based online news.1 However, news outlets may manipulate public opinion through misinformation or subtle modifications of news text’s semantic, stylistic, and affective characteristics.2,3 We hypothesized significant differences in news text characteristics between sources with different ideological orientations, levels of radicalism, and levels of public interest for highly divisive topics in our society: immigration, gun control, and vaccination. Data was collected from 360 articles on these topics from MSNBC News (left-leaning non-radical), Fox News (right-leaning non-radical), and Breitbart (right-leaning radical) during times of lowest and highest public interest. Text analysis with DICTION software identified characteristics of text, and a two-way ANOVA in SPSS 29 evaluated the effects of source and public interest on text characteristics. Results showed significant differences in text characteristics based on the source’s ideological leaning and radicalism level, as well as public interest in the topic. Right-leaning sources exhibited more categorical thinking, a greater tendency to overstate opinions, and a lower predisposition to invite readers to engage critically and evaluate the presented information than the left-leaning source. More radical sources displayed greater authoritarian inflexibility and robustness in their opinions. These differences suggest that news sources may aim to sway public opinion in a specific direction, potentially misrepresenting information. This study highlights the importance of critically evaluating news sources and the need for unbiased reporting to promote informed public opinion.
Eker, Emily and Babik, Iryna, "Text Analysis Identified Semantic Differences in Ideologically-Motivated News" (2023). 2023 Undergraduate Research Showcase. 21.