Files in this item

FilesDescriptionFormat

application/pdf

application/pdfSMITH-DISSERTATION-2020.pdf (1MB)Restricted Access
(no description provided)PDF

Description

Title:I saw it on social media: Public opinion on race, crime, and immigration in the era of social media news consumption
Author(s):Smith, Marisa Ashley
Director of Research:Dixon, Travis L
Doctoral Committee Chair(s):Dixon, Travis L
Doctoral Committee Member(s):Bigman, Cabral A.; Tewksbury, David; Yang, JungHwan; Usher, Nikki
Department / Program:Communication
Discipline:Communication
Degree Granting Institution:University of Illinois at Urbana-Champaign
Degree:Ph.D.
Genre:Dissertation
Subject(s):racial stereotypes
Twitter
news
media effects
computational
Abstract:Political polarization among the American electorate continuously grows (Iyengar, Sood, & Lelkes, 2012), perniciously impeding political cohesion. Partisans refute policies simply because they do not align with their party’s stance (Bechtel, Hainmueller, Hangartner, & Helbling, 2015) and partisans distrust members of the opposing political party (Iyengar & Krupenkin, 2018; Iyengar, Lelkes, Levendusky, Malhotra, & Westwood, 2018). As America approaches the 2020 presidential election, it is ever evident that identity drives politics (Kreiss, 2018; Kreiss, Lawrence, & McGregor, 2020). Given the growing chasm between Republicans and Democrats, scholarship often examines how news available on social media contributes to polarized opinions (e.g., Allcott & Gentzkow, 2017; Freelon & Wells, 2020; Grinberg, Joseph, Friedland, Swire-Thompson, & Lazer, 2019; Lewandowsky, Ecker, & Cook, 2017). However, the majority of research examining the implications of social media on sociopolitical attitudes omits the underlying cause of political polarization – racism. Racial resentment towards African Americans drives polarization between Republicans and Democrats (Valentino & Zhirkov, 2017), as well as, voting behavior among White Americans (Reny, Collingwood, & Valenzuela, 2019). Furthermore, attitudes towards immigration or the Black Lives Matter movement predicted support for Trump during the 2016 presidential election (Drakulich, Wozniak, Hagan, & Johnson, 2020; Reny et al., 2019). Holistic understanding of how social media impacts sociopolitical attitudes requires understanding whether news encountered during social media use impacts perceptions of African Americans or Latinos. Drawing on framing theory, I examined how crime and immigration news potentially encountered during Twitter use influences stereotype endorsement, perceptions of immigrants as a threat, and support for punitive crime or immigration policies. Furthermore, I proposed the concept of decentralized framing effects. This proposition contends that opportunities for news gatekeeping afforded by social media shifts influential power away from offline and online news sources to the variety of sources users encounter during social media use. As a result, audience opinions will be less contingent on the cable, print, or online news audiences directly select, and more so dependent on the news they incidentally encounter on social media. Study 1 recruited 244 adult Twitter users and content analyzed the crime and immigration news participants potentially encountered on Twitter. Results indicated that crime news was typically framed as relating to crime and justice, while immigration news emphasized the political and policy aspects of immigration. Surprisingly, explicit mentions of African Americans and Latinos were not prevalent in crime and immigration news. When present, however, African Americans were represented as victims of crime and mentions of Latino immigrants ranged from authorized immigrants (i.e., asylum seekers) to unauthorized immigrants (i.e., illegal alien). The top shared news sources were legacy news outlets, such as CNN and the New York Times. Furthermore, left-leaning sources dominated participants’ Twitter networks irrespective of the participant’s party affiliation. Study 2 (N = 141) expanded upon Study 1, examining relationships between the crime and immigration news in participants’ Twitter networks and participant attitudes. Analysis demonstrated that partisanship was the greatest predictor of immigrant attitudes and support for anti-immigrant policy. Interestingly, crime news participants potentially encountered on Twitter was marginally associated with depressed support for punitive crime policy. Furthermore, immigration news participants potentially encountered were associated with their thoughts on immigration. Overall, Study 2 found support for decentralized framing effects. Participants’ consumption of offline (e.g., newspapers, cable news) and online (e.g., news websites) news was significantly associated with participants’ endorsement of African American stereotypes. Heavy news consumers reported greater endorsement in African American stereotypes. However, when accounting for the crime news users potentially encountered on Twitter, greater crime news reduced stereotype endorsement for heavy consumers of mainstream + liberal news. In other words, the crime news participants potentially encountered on Twitter was capable of overriding the impact of cable, print, or online news. These results suggest that social media are influential in reshaping perceptions of African Americans and suppressing stereotypes cultivated during mainstream media use. The implications of these findings are discussed considering social media and activism, as well as immigration and political polarization.
Issue Date:2020-08-19
Type:Thesis
URI:http://hdl.handle.net/2142/109554
Rights Information:Copyright 2020 Marisa A. Smith
Date Available in IDEALS:2021-03-05
Date Deposited:2020-12


This item appears in the following Collection(s)

Item Statistics