Selection,Trust,and the Effects of Cable News Consumption

Abstract

Using two experimental studies, we examine how the selection and consumption of cable news influences news consumers’ cognitive processing, attitudes, and policy preferences. As expected, participants overwhelmingly self-selected into an ideologically aligned cable news network. Then, ideologically congruent messaging from Fox News and MSNBC was likely to prompt higher levels of agreement and lower levels of disagreement for those with mid- and high levels of trust in their selected cable news network. Our findings indicate a reinforcing spiral effect among both MSNBC and Fox News consumers who have high levels of trust in the cable news network they select.

Keywords

cable news trust Fox News MSNBC selective exposure

Cable news is the primary source of political information for many Americans, with viewership increasing in recent years (Pew Research Center, 2021). Audiences for both Fox News and CNN grew by more than 60% from 2019 to 2020, and MSNBC’s audience increased by nearly 30% during this period (Pew Research Center, 2021). In addition to a substantial market share, these “big three” cable news outlets have very different audiences. Americans consistently sort into different cable news networks by ideological leaning, with conservatives preferring Fox News and liberals selecting MSNBC and CNN in both real-world and experimental settings (Gil de Zúñiga et al., 2012; Holcomb, 2014; Iyengar & Hahn, 2009; Levendusky, 2013; Mitchell et al., 2014; Stroud, 2011). Yet, although partisan preferences for political news offerings are clear, the question remains how these networks influence Americans’ political attitudes and policy preferences.

Fox News has a particularly wide reach; it was the top-watched cable news network for every month from 2014 to 2019 (Flood, 2019). After a slow start to 2021 following U.S. President Donald Trump’s departure from office, Fox News still finished 2021 as the dominant cable news network in viewership (Katz, 2022). The Fox News franchise has successfully made itself the preeminent voice of conservative Americans (Jamieson & Capella, 2008), and Fox News consumers have markedly different political attitudes when compared with other cable news networks’ audiences (e.g., Gil de Zúñiga et al., 2012; Hoewe, Brownell, & Wiemer, 2020; Morris, 2007). The influence of an individual news outlet, however, may not hold for non-Fox News consumers. Compared with conservative Americans, liberal Americans tend to be more diversified in their news consumption habits, preferring CNN, MSNBC, and NPR, while avoiding Fox News (Iyengar & Hahn, 2009; Mitchell et al., 2014). Likewise, although polls consistently demonstrate that Republicans trust Fox News considerably more than any other news source, Democrats trust a broad range of sources including CNN, CBS, PBS, ABC, and NBC (Jurkowitz et al., 2020; Sanders, 2022). Compared with Fox News, there is less evidence of a relationship between consumption of CNN and MSNBC and individuals’ political attitudes and policy preferences (Hoewe, Brownell, & Wiemer, 2020; Hoewe, Peacock, et al., 2020). Thus, although selective exposure has long been connected to different perceptions of political reality (Gil de Zúñiga et al., 2012; Morris, 2007), the effects of cable news may be asymmetrical.

This study examines the potential differences in political attitudes and policy preferences that arise when Americans select news content from Fox News or MSNBC, with a consideration of the role of trust in the network selected. These two networks are similarly trusted among partisans (53% of Republicans trust Fox News and 52% of Democrats trust MSNBC; Sanders, 2022), making them an apt comparison. In two experiments using four different political issues, we confirm that—when given the opportunity in an experimental setting—participants overwhelmingly self-select into the ideologically consistent cable news network. Then, using those sorted samples, we examine how self-selected partisan programming shapes participants’ processing of issue content, particularly in terms of expressions of agreement and disagreement, and subsequent changes to political attitudes and policy preferences. We also consider a key mechanism in the potential effects of cable news use: trust in the network selected. We find that congruent arguments from both cable news networks—Fox News and MSNBC—are influential in reinforcing the political positions of those participants with high levels of trust in their selected cable news network. This research expands understanding of the roles of cable news selection and trust in information processing and changes in attitudes and policy preferences related to four distinct, high-salience topics.

Source Trust and Content Cues in Cable News Use

News sources have been conceptualized as brands, and the content they produce influences perceptions of that brand (Arendt et al., 2019; Gentzkow & Shapiro, 2006). Likewise, consumers’ attitudes toward a particular news source set expectations about whether that source aligns with preexisting values and political views (Westerwick et al., 2017). Considering media use, individuals seek to avoid the psychological discomfort caused by receiving dissonant information and tend to select information consistent with their personal attitudes or beliefs (Iyengar & Hahn, 2009; Stroud & Muddiman, 2012). Prior research has identified the source and content of news information as two distinct but interrelated constructs that influence whether and how a person decides to consume news content (Westerwick et al., 2017). Both are used as cues by news consumers and, thus, have implications for how that information is processed and its effects on audience members.

Likewise, perceptions of the credibility or trustworthiness of a given news source shape the influence of a particular source’s messaging on both attitudes and media use (Strömbäck et al., 2020). Credibility and trust are used somewhat interchangeably (Strömbäck et al., 2020), and since Hovland et al.’s (1953) research, scholars have noted how the source of a message markedly influences its impact on the consumer. Trust of a news source can be understood as the relationship between individuals (the trustor) and the media source (the trustee) where the individual, in situations of uncertainty, expects that interactions with that source will lead to positive rather than negative outcomes (Strömbäck et al., 2020), and based on that relationship, sources perceived to be more credible—and, therefore, more trustworthy—are more likely to persuade their audience (Pornpitakpan, 2004).

Source cues, then, are a powerful predictor of perceived source credibility and news consumption behavior, with partisans preferring and trusting news sources perceived as ideologically consistent with their existing beliefs (e.g., Mitchell et al., 2014; Pornpitakpan, 2004). Both Democrats and Republicans prefer sources that lean toward their respective party (Knobloch-Westerwick & Meng, 2009; Stroud, 2011), and staunch political partisans rely heavily on political source cues compared with nonpartisans (Gunther et al., 2012).

As such, it is important to consider the impact of trusted source cues in the study of cable news and its effects. In this study, we also incorporate a consideration of content cues (i.e., the argument presented in the news story) to determine how ideologically congruent or incongruent information interacts with the source cue provided by the cable news network.

Processing of Cable News: Agreement and Disagreement

One approach to measuring the processing of a political argument—particularly to assess evidence of motivated reasoning—is to evaluate the extent to which one chooses to agree or disagree with that information. Individuals may counter-argue against new information if it does not align with their existing opinions, even if factually accurate (Kunda, 1990; Taber & Lodge, 2006). In situations where media consumers can respond to an argument presented in the news, they have the choice to either support the argument or counter-argue against it (Xu & Wyer, 2012). This ability to engage—either positively or negatively—with the argument is powerful. As one example, even the mere presence of agreeable and disagreeable comments under news articles influences multiple types of user engagement with the story (Segesten et al., 2022). Therefore, the role of agreement and disagreement in news consumption and processing merits deeper investigation.

Individuals’ expectations are important factors for expressions of agreement and disagreement as an outcome of news consumption. If news consumers expect a particular point of view from a source, the perceived chances of experiencing resistance to the presented information are relatively low (Guess & Coppock, 2020). In other words, when given the chance to choose a familiar and/or trusted information source, consumers expect that the newly presented information will align with previous information from that source. Compared with a reader seeing information they expect, if the information is not what a reader anticipates and their evaluation of that information changes from agreement to disagreement, it is more likely for that reader to experience change and/or destabilization in their initial attitudes (Michaels et al., 2013). News consumers expecting to agree with news content may experience incoherence when they encounter unexpected counter-attitudinal information (Ewoldsen et al., 2022). In this way, expressions of agreement and disagreement are worth measuring as separate metrics to capture the full breadth of possible responses to news and information, particularly when individuals’ expectations about that news source or content differ from what is presented to them.

Expressions of agreement or disagreement may also relate to the reinforcing processes undertaken in the selective consumption of ideologically aligned news. The reinforcing spirals model (RSM) suggests that selective exposure works alongside the effects of that media content to establish and reinforce individuals’ attitudes and behaviors (Slater, 2007, 2015). In this way, the RSM proposes that media use and attitude formation, solidification, and change are part of a circular—or reinforcing—process, where media use functions as both a predictor and an outcome. One’s attitudes can predict the selection of media content, but consumption of that media content can also predict one’s attitudes (Slater, 2015). Alongside indicators of cognitive processing, as considered through expressions of agreement and disagreement, this process illustrates the potential reinforcing nature of selection into a particular cable news environment.

Thus, coupled with what is known about cable news networks and their effects on consumers, more work is needed to examine the outcomes of individuals choosing to receive news from a trusted partisan source yet instead receiving counter-attitudinal information. This study was designed to test the effects of individuals selecting cable news content that aligns with their political beliefs (source cues) and then whether that content either reinforces or contradicts those beliefs (content cues). It may be that source cues are prioritized over the expectation of source-congruent information, making room for a source cue effect where the news source is more influential than the information’s incongruence. However, it is possible that correcting the dissonance associated with consuming counter-attitudinal information is a higher priority than relying on the source of that information. For this reason, we consider the interplay between source and content cues in this study, where the ideological position of the argument (i.e., conservative- or liberal-leaning argument) will be varied in relation to the source (i.e., Fox News or MSNBC).

H1: Level of trust in the news network selected will moderate the relationship between the news source label and the ideological slant of the story, such that individuals who have higher levels of trust in the news network they selected and read a news story that is ideologically congruent with the news source will express (a) more agreement and (b) less disagreement with the news story.

RQ1: Will there be a direct relationship between news source selection and content cues, as moderated by level of trust, and changes in attitudes and policy preferences?

RQ2: Will greater amounts of agreement or disagreement be related to changes in attitudes and policy preferences?

Cable News Selection: The Fox News Effect?

When considering the relative power of source and content cues, different media outlets may have different effects. As was previously mentioned, Fox News may influence its consumers differently than other cable news networks. Fox News covers events from a perspective favorable to a Republican agenda (Kull et al., 2003; Morris, 2005; Weaver & Scacco, 2013), and that creative direction has paid off in the form of loyalty and trust from conservative viewers (Gramlich, 2020; Mitchell et al., 2014). Its viewers are more likely to hold policy preferences aligned with those espoused on the network regarding a broad range of issues, including immigration, climate change, and COVID-19 (Hoewe, Brownell, & Wiemer, 2020). Fox News use also has been linked to political behavior, including voting for Trump (Gottfried et al., 2017) and reductions in social distancing during the COVID-19 pandemic (Simonov et al., 2021). Fox News use was related to policy preferences regarding and more negative attitudes toward immigrants and refugees, even after controlling for political ideology, while there was no relationship for MSNBC or CNN consumers (Hoewe, Peacock, et al., 2020). Finally, conservatives who chose Fox News content were more likely to engage in like-minded political discussions, while liberals who chose MSNBC were more likely to be motivated into political participation (Choi, 2022).

Considering that prior research has illustrated the unique relationship between Fox News and its consumers, continued research should identify the implications of source and content cues on these effects. Thus, the present research considers the source cues of the information (i.e., Fox News or MSNBC) as well as the content cues (i.e., conservative- or liberal-leaning argument) and these cues’ effects on audiences’ processing of the news they consume. But in combination, we also expect the interplay between these cues to change based on the relationship between the cable news network and the consumer. Considering the distinctions in cable news networks’ content and effects noted previously, our research examines the differences in news consumers’ responses to the coverage provided by the cable news network from which they choose to read. Based on loyalty to the individual brand, it is possible that those who select Fox News will be more responsive to news label (i.e., source cue) than those that consume MSNBC, who we expect to respond more to the ideological leaning (i.e., content cue) of the news story.

RQ3: Will individuals who select Fox News and indicate trust in Fox News be more likely than those who select MSNBC to express more agreement and less disagreement and experience changes in their attitudes and policy preference in alignment with the argument presented?

To address these hypotheses and research questions, we conducted two experimental studies, where participants selected their preferred cable news network (Fox News or MSNBC) and then were randomly assigned to a conservative- or liberal-leaning news story. Study 1 examines the issue of immigration, and Study 2 utilizes stimulus sampling and includes three issues: the federal deficit, critical race theory in K–12 schools, and COVID-19 vaccine mandates.

Study 1: Method

Stimuli

The stimuli for this study were created from existing news stories (published by NBC News). They were edited to contain opposite arguments: one liberal-leaning that argued to increase immigrant work permits and one conservative-leaning that argued to decrease immigrant work permits. The news stories utilized the same format, where the headline and subhead were manipulated to reflect the liberal or conservative lean of the argument. Small edits were made to make the ideological lean clear (primarily in the latter portion of the article), but the stories were kept as similar as possible. The topic of immigration was selected based on its high salience and perceived importance among Americans (Gallup, 2018). Each news story was approximately the same length (506–511 words) and was presented to reflect the length and journalistic tone of actual stories published on FoxNews.com and MSNBC.com (see Supplemental Materials for stimuli).

Pretest

Participants were recruited from Mechanical Turk to take part in a pretest examining the news story stimuli and the salience of the issues selected for this study. Using the procedures outlined by Burleigh et al. (2018) to exclude bots and fake IP addresses, only individuals living in the United States were included. This sample of 102 participants included 39% women with an average age of 34.63 (SD = 11.31). In addition, 74.5% reported being White/Caucasian, 8.8% as Asian/Pacific Islander, 9.8% as Black/African American, 6.9% as Hispanic/Latino, 1.0% as Native American, and 2.9% as some other race/ethnicity (participants could select all options that applied). The majority had completed at least some college (83.3%), and the median income was US$50,000 to US$75,000.

Participants were randomly assigned to read the liberal-leaning (N = 50) or conservative-leaning news story (N = 52). Of those who read the liberal-leaning story, 70.0% correctly identified that it argued to allow immigrant work permits. Of those who read the conservative-leaning story, 82.2% correctly identified that it argued to restrict immigrant work permits. Participants also were asked to rate their perception of bias in the news story (1 = Very Conservative, 7 = Very Liberal). There was a significant difference in participants’ ratings, where the liberal-leaning condition (M = 4.94) was perceived as significantly more liberal than the conservative-leaning condition (M = 2.90), β = .65, t(100) = 8.47, p < .001.

Final Study Participants

Participants for the final study were recruited in November 2019 from Prolific Academic, where they were paid for their participation.¹ Prolific Academic recruited a sample of 298 participants meant to closely resemble the U.S. population based on age, gender, and race/ethnicity (see Prolific Team, 2019). This group of participants included 50% women with an average age of 44.28 (SD = 15.68). In addition, 69.9% reported being White/Caucasian, 16.8% as Black/African American, 10.0% as Asian/Pacific Islander, 9.4% as Hispanic/Latino, 1.9% as Native American, and 1.0% as some other race/ethnicity (participants could select all options that applied). The majority had completed at least some college (89%), and the median income was US$50,000 to US$75,000.

Procedure

Participants in the final study consented to participate and then answered questions regarding issue salience and their attitudes and policy preferences regarding immigrant work permits. They were then told they would read a news story. To increase external validity by simulating selective exposure, participants were asked to select the cable news network from which they would read. That is, participants were asked to choose between Fox News and MSNBC and were subsequently shown a story from the news source that they selected. They were required to stay on the news story page for at least 30 seconds to ensure they read the story. Participants were randomly assigned to a news story that was liberal-leaning or conservative-leaning in its arguments. This process produced four conditions: MSNBC story with liberal-leaning argument (N = 99, 32.0%), MSNBC story with conservative-leaning argument (N = 99, 32.0%), Fox News story with liberal-leaning argument (N = 56, 18.1%), and Fox News story with conservative-leaning argument (N = 55, 17.8%). The liberal-leaning stories were identical except for the cable news outlet heading, as were the conservative-leaning stories. Following stimulus consumption, participants completed open-ended prompts and then answered the same attitude and policy questions that were asked prior to stimulus exposure. Finally, they completed manipulation checks and demographic questions.

Variables

Participants’ levels of agreement and disagreement with the news story were assessed using open-ended prompts (e.g., Dillard & Shen, 2005; Hwang et al., 2018). Immediately after reading the news story, they were asked on separate pages to “list the information in the news story that you agreed [disagreed] with.” The ordering of these agree/disagree open-ended prompts was randomized to reduce ordering effects. Participants’ responses in agreement ranged from 0 to 131 words (M = 20.38, SD = 16.19), and their responses that disagreed ranged from 0 to 108 words (M = 18.84, SD = 17.88). These open-ended comments were examined by coders who were blind to the condition to which participants were assigned. Coders were asked to identify “how much agreement [disagreement] did the participant indicate?” The codebook (see Supplemental Materials) used a 4-point scale ranging from 1 None to 4 A lot. Two coders examined 20% of the sample to establish intercoder reliability, and acceptable levels were achieved (agreement responses: 92% agreement and Cohen’s kappa = .82; disagreement responses: 97% agreement and Cohen’s kappa = .94). The completed coding showed that participants expressed more overall agreement (M = 2.14, SD = 0.75) than disagreement (M = 1.94, SD = 0.86).

To assess possible changes in attitudes and policy preferences that may have resulted from stimulus exposure, participants were asked these questions twice—once before reading the stimulus news story (to establish baseline levels) and again after reading the story. Each item used a 7-point scale (1 Strongly Disagree to 7 Strongly Agree). The prestimulus scores were subtracted from the poststimulus scores to produce a d-score of attitude or policy preference change. The attitude questions consisted of four items, two reverse coded, where higher scores indicated more positive attitudes toward immigrant work permits. These items included “Immigrant work permits benefit people”; “Immigrant work permits hurt people”; “I would recommend allowing immigrant work permits”; and “I would recommend restricting immigrant work permits.” These items formed a reliable measure (prestimulus: M = 5.02, SD = 1.44, α = .93; poststimulus: M = 5.02, SD = 1.45, α = .92; d-score: M = −0.01, SD = 0.61). The policy preference questions were adapted from Hoewe (2018) and included six items, three reverse coded, where higher scores indicated a more favorable position toward immigrant work permits. These items included, for example, “the U.S. should allow more immigrant work permits” and “immigrant work permits in the U.S. need fewer restrictions.” These items also formed a reliable measure (prestimulus: M = 4.45, SD = 1.52, α = .93; poststimulus: M = 4.47, SD = 1.59, α = .95; d-score: M = 0.03, SD = 0.66).

Participants also were asked about their levels of trust in the cable news source that they selected. Using five questions about whether the source is trustworthy, accurate, unfair (reverse coded), biased (reverse coded), and tells the whole story (Meyer, 1988), participants rated their trust in either Fox News or MSNBC on 7-point scales (1 Strongly disagree to 7 Strongly agree), which formed a reliable measure of trust in the news source selected (M = 4.24, SD = 1.38, α = .92).

Participants’ political party is included as a control variable in the subsequent models. Participants who identified as Democrat or Lean Democrat were coded as 0 (N = 161) and those who identified as Republican or Lean Republican were coded as 1 (N = 70). Those who did not identify with a political party were coded as 1 (N = 28) to form a dummy-coded variable.

Manipulation Check

Although the stimulus news stories were already pretested, they were reexamined using a manipulation check for the ideological leaning of the news story to ensure that participants recognized the slant of the story. Of those who were randomly assigned to the liberal-leaning story, 84.8% correctly identified that it argued in favor of allowing immigrant work permits. Ninety-one percent of those who read the conservative-leaning story correctly said it advocated against allowing immigrant work permits. Participants who failed this manipulation check were removed from the sample; therefore, only participants who recognized the liberal or conservative slant to the news story are included in the subsequent analyses.

Study 1: Results

The analyses began by determining if participants chose to read a news story from the cable news network that aligned ideologically with their political party. A crosstabs analysis (χ² = 45.30, df = 3, p < .001, Cramer’s V = .42) showed that 81.0% of Democrats selected MSNBC, compared with 28.8% of Republicans, 66.2% of Independents, and 76.9% of those with no party affiliation. Similarly, 71.2% of Republicans selected Fox News, compared with 19.0% of Democrats, 33.8% of Independents, and 23.1% of those with no party affiliation. As expected, the results of this test show that the majority of partisan participants selected the news outlet that agreed ideologically with their political stance, which provides evidence of selective exposure. Diving further into actual cable news usage, 63 participants reported using MSNBC, 42 reported using Fox News, 46 reported using both Fox News and MSNBC, and 108 did not use Fox News or MSNBC. Of those, 90.5% of MSNBC users selected MSNBC, and 81.0% of Fox News users selected Fox News. For participants who reported using both networks, 39.1% selected Fox News and 60.9% selected MSNBC. Of those who said they do not use either network, 73.1% selected MSNBC and 26.9% selected Fox News. These data suggest selective exposure based on both partisanship and prior cable news use.

To test this study’s hypotheses and research questions, a moderated-mediation model was created with the SPSS PROCESS macro (Model 12; Hayes, 2017), using 5,000 bootstrapped samples and 95% bias-adjusted confidence intervals. The two condition variables (cable news source selected X ideological lean of the news story) and the moderating variable (trust in the cable news source selected) formed the three-way interaction. The mediating variables were the levels agreement expressed and disagreement expressed after reading the news story, the control variable was political party, and the dependent variable was either changes in attitudes or policy preferences.

The three-way interaction predicting levels of agreement expressed was statistically significant, b = 0.35, t(248) = 2.54, p = .01. Participants with high levels of trust in their selected cable news network who were randomly assigned to read a news story that was ideologically congruent with that news network were more likely to express agreement with the position taken in the news story. That is, when participants’ trust in the news network selected was at the mean value and at one standard deviation above the mean, there was a significant difference in their levels of agreement based on the type of ideological argument they saw. More specifically, for participants at the mid- and high levels of trust, those who selected Fox News and read a conservative-leaning argument expressed significantly more agreement than did participants who selected MSNBC and saw that argument (b = 0.30, p = .04; b = 0.47, p = .03, respectively). This relationship is illustrated in Figure 1. In addition, participants at high levels of trust who selected MSNBC and read a liberal-leaning argument expressed more agreement than did participants who selected Fox News and saw that argument, and this relationship—illustrated in Figure 2—approached statistical significance (b = −0.39, p = .07).

Figure 1.

Levels of Agreement for Conservative-Leaning Argument in Study 1.

Figure 2.

Levels of Agreement for Liberal-Leaning Argument in Study 1.

There was not a significant three-way interaction predicting expressions of disagreement, b = −0.18, t(248) = −1.12, p = .26. These results provide support for H1a regarding expressions of agreement, but not H1b which considered expressions of disagreement. Addressing the direct effect examined in RQ1, the three-way interaction (produced by the condition variables and the moderating variable of trust) was not related to attitude change, b = 0.06, t(246) = 0.50, p = .62, or policy preference change, b = 0.04, t(246) = 0.30, p = .76.

The second portion of this study’s moderated-mediation model addressed RQ2, which considered whether levels of agreement and/or disagreement with the news story were related to changes in attitudes and policy preferences (where each dependent variable was tested in a separate model). There were no indirect effects within these models, nor was there a significant direct effect from disagreement to changes in attitude change or policy preference. However, there was a positive relationship between agreement and attitude change, b = 0.14, t(246) = 2.60, p = .01, as well as policy preference change, b = 0.17, t(246) = 3.05, p = .003. These findings indicate that postexposure processing and evaluation of news content operates differently according to the positive or negative valence of that evaluation. Here, we find that expressions of agreement with news content are positively related to changes in attitudes and policy preferences to a more substantial degree than expressions of disagreement.

Addressing RQ3, expressions of agreement with the ideologically congruent argument were illustrated for both Fox News and MSNBC consumers who expressed high levels of trust in the network they selected. These findings may indicate a reinforcing spiral among cable news consumers with high levels of trust in the network they select to consume. In that case, having trust in sources to provide the anticipated ideologically congruent information may help explain the processing of news content and changes in attitudes and policy preferences. That said, there may be additional factors at play beyond met expectations about the ideological bent of a news story, as informed by trust in the news source. For example, since this study examined one specific topic, these results could be contingent on that topic. To resolve this issue, we replicate these results using stimulus sampling with three additional topics in Study 2.

Study 2: Method

Pretest

Before exposing participants to news story stimuli, we conducted a pretest to ensure that the topics presented in each story aligned with actual issues that both Republicans and Democrats view as important, but for ideologically different reasons. To achieve this, participants were recruited from Prolific Academic to take part in a pretest that examined attitudes toward and perceived importance of particular topics. This sample of 102 participants included 52% women with an average age of 35.73 (SD = 14.73). In addition, 78.4% reported being White/Caucasian, 7.8% as Asian/Pacific Islander, 7.8% as Black/African American, 16.7% as Hispanic/Latino, 2.9% as Native American, and 2.9% as some other race/ethnicity (participants could select all options that applied). The majority had completed at least some college (90.1%), and the median income was US$50,000 to US$75,000.

Since we are interested in how these topics are considered by partisan audiences, this pretest identified topics that showed significant differences between Democrats and Republicans in terms of their evaluations of the topics (three-item scale including good/bad, like/dislike, and beneficial/harmful) and also were not significantly different in terms of their perceived importance “to you” and “to the United States.” In this way, we attempted to identify topics that Republicans and Democrats both feel are important to themselves and to the United States, but that they evaluate very differently in terms of their attitudes. The three topics that fit these criteria included “the current federal deficit,” “COVID-19 vaccine mandates,” and “critical race theory being taught in K-12 schools,” and these topics were used in the news story stimuli.

Final Study Participants

A new group of participants was recruited from Prolific Academic, and data collection was completed in May 2022. This group of 601 participants included 300 people who identified as conservative and 301 who identified as liberal. This sample was 52.2% women with an average age of 39.80 (SD = 14.40). In addition, 84.9% reported being White/Caucasian, 8.2% as Asian/Pacific Islander, 6.8% as Black/African American, 6.3% as Hispanic, 1.0% as Native American, and 0.8% as some other race/ethnicity (participants could select all options that applied). The majority had completed at least some college (85.2%), and the median income was US$50,000 to US$75,000.

Stimuli

The stimuli for this study were again created from actual news stories, using the same base news story (from the Associated Press) with added partisan arguments. That is, these news stories were edited to contain opposite arguments for each topic: one liberal-leaning and one conservative-leaning. The overall content of the news stories was kept as similar as possible, and the headline, subhead, and portions of the argument were manipulated to reflect the liberal or conservative lean of the argument. Also, stimulus sampling—using the three topics identified in the pretest—provides greater generalizability for this study. Each news story was approximately the same length (192–207 words) and was presented look like an actual story published on FoxNews.com and MSNBC.com (see Supplemental Materials for stimuli examples).

Procedure

Participants consented to participate and then answered questions regarding their attitudes and policy preferences about the current federal deficit, COVID-19 vaccine mandates, and critical race theory being taught in K–12 schools. Identical to Study 1, they were then told they would read a news story, and they were asked to select the cable news network from which they would like to read (Fox News or MSNBC). They were then shown a story from the news source that they selected, and they were required to stay on the news story page for at least 30 seconds to ensure they read the story. Participants were randomly assigned to read a news story about one of the three topics. They also were randomly assigned to either liberal-leaning or conservative-leaning arguments within the news story. This process produced 12 news stories that were condensed into the four manipulations of interest: MSNBC story with liberal-leaning argument (N = 176, 29.3%), MSNBC story with conservative-leaning argument (N = 180, 30.0%), Fox News story with liberal-leaning argument (N = 126, 21.0%), and Fox News story with conservative-leaning argument (N = 119, 19.8%). Following stimulus consumption, participants completed open-ended prompts and then answered the same attitude and policy questions that were asked prior to stimulus exposure but only for the topic they read about in the news story. Finally, they completed manipulation checks, questions unrelated to this study, and demographic questions.

Variables

Participants’ levels of agreement and disagreement with the news story were assessed using open-ended prompts and closed-ended measures. Immediately after reading the news story, they were asked on separate pages to “list the information in the news story that you agreed [disagreed] with.” Participants were asked to type each thing they agreed [disagreed] with in a separate textbox, and if they did not agree [disagree] with anything in the news story, they wrote “nothing” in the final textbox. The order of these agree/disagree open-ended prompts was randomized to reduce ordering effects. Each textbox entry was counted as 1, with the maximum number of entries being 7 (individuals who wrote “nothing” were coded as 0; individuals who wrote “all,” “everything,” or some other indication of total agreement/disagreement were coded as 7). Participants expressed more overall agreement (M = 1.52, SD = 1.33) than disagreement (M = 0.95, SD = 1.24). In addition, participants were asked to identify their levels of agreement and disagreement on two different Likert-type scales. They were asked to think about the parts of the news story they agreed [disagreed] with and identify how much they agreed [disagreed] with the news story (7-point scale from 1 Did not agree [disagree] at all to 7 Agreed [Disagreed] with entire story). This resulted in two items: agreement (M = 4.58, SD = 2.18) and disagreement (M = 3.36, SD = 2.29). Importantly, these measures of agreement and disagreement were highly correlated (agreement: r = .60, p < .001; disagreement: r = .74, p < .001), indicating the validity of these items.

As in Study 1, to assess possible changes in attitudes and policy preferences that may have resulted from stimulus exposure, participants were asked these questions twice—once before reading the stimulus news story and again after reading the news story. In this study, participants were asked attitude and policy preference questions about all three topics before the stimulus, helping reduce sensitization effects. After seeing the stimulus, they were only asked to readdress the questions about the topic covered in the news story they read. Once again, prestimulus scores were subtracted from the poststimulus scores to produce a d-score of attitude or policy preference change. The attitude and policy preference questions mimicked those from Study 1 as closely as possible, and the conservative position was coded as lower and the liberal position as higher. The attitude items formed reliable measures: (a) the federal deficit prestimulus: M = 4.58, SD = 1.74, α = .95; poststimulus: M = 4.78, SD = 1.94, α = .95; d-score: M = 0.18, SD = 0.86; (b) critical race theory in K–12 schools: M = 4.10, SD = 2.25, α = .96; poststimulus: M = 4.05, SD = 2.34, α = .95; d-score: M = 0.02, SD = 0.69; and (c) COVID-19 vaccine mandates: M = 4.47, SD = 2.17, α = .97; poststimulus: M = 4.56, SD = 2.17, α = .97; d-score: M = 0.08, SD = 0.69. The policy preference items formed reliable measures as well: (a) the federal deficit prestimulus: M = 4.33, SD = 1.82, α = .97; poststimulus: M = 4.60, SD = 2.07, α = .98; d-score: M = 0.19, SD = 0.89; (b) critical race theory in K–12 schools: M = 4.17, SD = 2.32, α = .98; poststimulus: M = 4.09, SD = 2.39, α = .99; d-score: M = 0.02, SD = 0.64; and (c) COVID-19 vaccine mandates: M = 4.27, SD = 2.20, α = .97; poststimulus: M = 4.37, SD = 2.26, α = .98; d-score: M = 0.11, SD = 0.71.

Participants also were asked about their levels of trust in the news source they selected. Using the same questions as in Study 1, participants rated their trust in either Fox News or MSNBC. This again formed a reliable measure of trust in the news source selected (M = 4.29, SD = 1.45, α = .93).

Once again, participants’ political party was assessed to be included as a control variable. Participants who identified as Democrat or Lean Democrat were coded as 0 (N = 298) and those who identified as Republican or Lean Republican were coded as 1 (N = 276). Those who did not identify with a political party were coded as 1 (N = 27) to facilitate dummy coding.

Manipulation Check

A manipulation check examined the ideological leaning of the news story to ensure that participants recognized the slant of the story. Twenty-two participants identified the liberal argument incorrectly (7.4%), and 21 participants identified the conservative argument incorrectly (7.2%). In addition, 12 participants did not answer this question. As in Study 1, participants who failed this manipulation check (or failed to answer it) were removed from the sample, so only participants who recognized the liberal or conservative slant to the news story are included in the subsequent analyses.

Study 2: Results

The analyses for Study 2 were completed as closely to those of Study 1 as possible. First, a crosstabs analysis (χ² = 188.34, df = 3, p < .001, Cramer’s V = .59) showed that 89.5% of Democrats selected MSNBC, compared with 25.9% of Republicans, 63.0% of Independents, and 66.7% of those with no party affiliation. Similarly, 74.1% of Republicans selected Fox News, compared with 10.5% of Democrats, 37.0% of Independents, and 33.3% of those with no party affiliation. Examining actual cable news usage, 102 participants reported using MSNBC, 113 reported using Fox News, 88 reported using both Fox News and MSNBC, and 243 did not use Fox News or MSNBC. Of those, 95.1% of MSNBC users selected MSNBC, and 86.7% of Fox News users selected Fox News. For those who used both networks, 56.8% selected Fox News and 43.2% selected MSNBC. For those who said they do not use either network, 74.9% selected MSNBC and 25.1% selected Fox News. Thus, Study 2 also illustrates selective exposure based on both partisanship and prior cable news use.

The SPSS PROCESS macro (Model 12; Hayes, 2017) using 5,000 bootstrapped samples and 95% bias-adjusted confidence intervals was again used to test the hypotheses and research questions. We tested both sets of agreement and disagreement measures as mediators, and they performed similarly. For the sake of parsimony, we only report the results of the closed-ended measures below.

The three-way interaction predicting levels of agreement expressed was statistically significant, b = 0.97, t(537) = 4.39, p < .001. Replicating the effect from Study 1, for participants at the mid- and high levels of trust, those who selected Fox News and read a conservative-leaning argument expressed significantly more agreement than did participants who selected MSNBC and saw that argument (b = 1.45, p < .001; b = 1.75, p < .001, respectively). This effect is shown in Figure 3. In addition, for participants at mid- and high levels of trust, those who selected MSNBC and read a liberal-leaning argument expressed more agreement than did participants who selected Fox News and saw that argument (b = −1.56, p < .001; b = −2.31, p < .001, respectively), as is shown in Figure 4.

Figure 3.

Levels of Agreement for Conservative-Leaning Argument in Study 2.

Figure 4.

Levels of Agreement for Liberal-Leaning Argument in Study 2.

Unlike in Study 1, there was a significant three-way interaction predicting expressions of disagreement, b = −0.92, t(537) = −4.03, p < .001. These results are illustrated in Figures 5 and 6. For participants at the mid- and high levels of trust, those who selected Fox News and read a conservative-leaning argument expressed significantly less disagreement than did participants who selected MSNBC and saw that argument (b = −1.33, p < .001; b = −1.91, p < .001, respectively). In addition, for participants at mid- and high levels of trust, those who selected MSNBC and read a liberal-leaning argument expressed less disagreement than did participants who selected Fox News and saw that argument (b = 1.65, p < .001; b = 2.35, p < .001, respectively). Taken together, the results of Study 2 provide support for both H1a and H1b.

Figure 5.

Levels of Disagreement for Conservative-Leaning Argument in Study 2.

Figure 6.

Levels of Disagreement for Liberal-Leaning Argument in Study 2.

Addressing the direct effect examined in RQ1, the three-way interaction was related to attitude change, b = −0.31, t(535) = −3.47, p = .001. Participants who selected MSNBC, had mid- and high levels of trust in MSNBC, and were randomly assigned to a liberal argument illustrated direct attitude change ( b = 0.19, p = .07; b = 0.49, p = .001, respectively). This three-way interaction predicting policy preference change approached statistical significance, b = −0.17, t(535) = −1.85, p = .07. Similarly, participants who selected MSNBC, had mid- and high levels of trust in MSNBC, and were randomly assigned to a liberal argument illustrated direct policy preference change (b = 0.26, p = .02; b = 0.43, p = .006, respectively). There were no indirect effects within these models, nor was there a significant direct effect from agreement or disagreement to changes in attitude change or policy preference, which addresses RQ2.

Regarding RQ3, the effects detailed above were found for participants who selected both MSNBC and Fox News, had high levels of trust in the network they selected, and saw an argument that aligned with the network’s ideological slant. Moreover, when the model was rerun with only participants who reported consuming Fox News and/or MSNBC in their actual news use, the results reported here were maintained. Thus, taken together with the results of Study 1, these findings indicate a reinforcing spiral effect among both MSNBC and Fox News consumers who have high levels of trust in the cable news network they select, where their political positions are reinforced through ideologically consistent arguments.

Discussion

Cable news is a critical source of political information for many Americans, shaping individuals’ perspectives of many important issues. To gain a better understanding of if and how cable news influences its consumers, we examined how selective exposure to cable news impacts expressions of agreement and disagreement as well as changes in attitudes and policy preferences among news consumers, particularly considering the role of trust in the selected cable news network. In both experiments, partisan participants selected the cable news network that ideologically matched their political party, confirming the basic premise of selective exposure (e.g., Knobloch-Westerwick & Meng, 2009; Stroud, 2011).

Using this self-sorted sample, we were able to test the effects of cable news selection and consumption in relation to the strength of source cues versus content cues in the assessment of received information and potential differences between two well-known cable news networks: Fox News and MSNBC. To do so, we evaluated how levels of trust in the selected news network impacted the reception of ideologically slanted news content. We found that trust in the cable news source leads to more expressions of agreement and fewer expressions of disagreement with the news story, depending on the ideological slant of the story. These findings provide context for the conditions under which the source and content cues, in the case of Fox News and MSNBC, alter consumers’ reactions to news content.

Our results suggest several overarching contributions. The first is regarding the cognitive processing that occurs during the consumption of cable news, particularly with regard to expressions of agreement and disagreement with the news story. Ideologically congruent messaging from Fox News and MSNBC prompted higher levels of agreement and lower levels of disagreement, particularly for those with high levels of trust in the cable news network they selected. Conversely, ideologically incongruent messaging prompted lower levels of agreement and higher levels of disagreement among those with greater trust in the network they selected. Importantly though, this relationship was more consistent for expressions of agreement than for disagreement. These results indicate that both source cues (the cable news network) and content cues (the ideological slant of the news story) are part of the news consumption process for individuals with high levels of trust in cable news.

Next, impacts on attitudes and policy preferences were more likely for individuals who saw an ideologically congruent argument from a cable news source they trust. These results are in line with recent research suggesting that the confirmation of preexisting attitudes serves as a stronger incentive during news consumption than does a desire to be accurate (Mourão et al., 2023). Both of our experimental studies showed that getting information from a trusted cable news source contributes to the reinforcement of congruent attitudes about high-salience political issues. In Study 1, we found a relationship between expressions of agreement and changes in attitudes and policy preferences. In Study 2, participants who selected MSNBC, saw a liberal argument, and had high levels of trust in MSNBC were more likely to change their attitudes and policy preferences in line with the news story. These results are perhaps best explained by the RSM, where individuals’ beliefs and attitudes are strengthened through consumption of congruent messages (Slater, 2007, 2015). This process then spirals as individuals continue to select content that supports their political stances. That is, consuming ideologically congruent news content from these trusted cable news sources led to more expressions of agreement and fewer expressions of disagreement among participants, with the potential to prompt changes in attitudes and policy preferences in line with that ideologically congruent argument.

In sum, our research shows that congruent arguments about high-salience issues, including immigration, the federal deficit, critical race theory, and COVID-19 vaccines, are influential for individuals who trust and choose to consume news content from cable news sources. Past research has shown that one cable news network—Fox News—has the propensity to lead news consumers to policy preferences and behaviors in line with that network’s content (Gottfried et al., 2017; Simonov et al., 2021). Fox News is also a regularly leading cable news channel among television news viewers, adding to expectations that Fox News may be especially influential among those who select it (Flood, 2019; Katz, 2022). However, we found that Fox News was not uniquely influential in this experimental setting. Instead, we found influences among participants who selected Fox News and MSNBC. The powerful factor in this relationship was the trust consumers placed in the cable news source they selected. Thus, to better understand the real-world influence of cable news on its consumers, it is extremely important to examine news consumers’ loyalty and continued use of that cable news network, particularly when they consider it a trusted news source.

The results of this study suggest many avenues for further research. Future studies should consider the medium of news consumption when studying the impact of brands like Fox News and MSNBC, which could be done by testing video stimuli to assess effects. Using such stimuli might increase the influence of the news story, particularly for those with high levels of trust in a particular show or commentator. Also, future research should more closely address potential moderating variables in these relationships, including need for cognition, tolerance for ambiguity, and need for closure (Hollander, 2018; Young, 2020). Testing and controlling for more established constructs related to persuasion research, such as initial motivation and ability to argue or counter-argue (e.g., Silvia, 2006), would add to the richness of our explanations as well. We also made the decision to measure participants’ attitudes and policy preferences before and after the stimulus to facilitate the calculation of a difference score. However, this decision is not without consequences. It is possible that asking these questions before the stimulus activated particular responses to the stimuli (i.e., sensitization). To mitigate some of these potential effects, participants were also asked attitude and policy preference questions that were not related to the stimulus news story they received. Future research should consider asking these questions poststimulus only to see whether differences emerge. Finally, more research should address the effects of cable news, particularly its ability to reinforce and strengthen cognitions, attitudes, and policy preferences among partisans.

Supplemental Material

sj-docx-1-jmq-10.1177_10776990221142877 – Supplemental material for Selection, Trust, and the Effects of Cable News Consumption

Supplemental material, sj-docx-1-jmq-10.1177_10776990221142877 for Selection, Trust, and the Effects of Cable News Consumption by Jennifer Hoewe, Janel Jett, Amber Lusvardi and Eric Wiemer in Journalism & Mass Communication Quarterly

Footnotes

Acknowledgements

Janel Jett is now affiliated to University of Missouri, Columbia, MO, USA.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Jennifer Hoewe

Eric Wiemer

Supplemental Material

Supplemental material for this article is available online.

Notes

Author Biographies

Jennifer Hoewe is an associate professor in the Brian Lamb School of Communication at Purdue University. She studies media psychology and political communication. Her research focuses on how political issues and groups of people are depicted in media content and how those depictions influence media consumers, particularly in terms of their cognitive processing, their attitudes, and their own identity.

Janel Jett is a Postdoctoral Research Scholar with the New Carbon Economy Consortium and the Harry S. Truman School of Government and Public Affairs at the University of Missouri. She specializes in political communication and political psychology, particularly in the context of science, technology, and environmental policy.

Amber Lusvardi is an assistant professor of political science at the University of Wisconsin Oshkosh. Her work focuses on agenda setting and issue framing of gender justice issues.

Eric Wiemer is a Jr. Data Analyst at Citizen Data. His work is concerned with strengthening democracy in America through social science research and natural language processing techniques.

References

Arendt

Northup

Camaj

(2019). Selective exposure and news media brands: Implicit and explicit attitudes as predictors of news choice. Media Psychology, 22(3), 526–543. https://doi.org/10.1080/15213269.2017.1338963

Burleigh

Kennedy

Clifford

(2018). How to screen out VPS and international respondents using Qualtrics: A protocol. https://ssrn.com/abstract=3265459

Choi

(2022). How partisan cable news mobilizes viewers: Partisan media, discussion networks and political participation. Journal of Broadcasting & Electronic Media, 66, 129–152. https://10.1080/08838151.2021.2013220

Dillard

J. P.

Shen

(2005). On the nature of reactance and its role in persuasive health communication. Communication Monographs, 72(2), 144–168. https://doi.org/10.1080/03637750500111815

Ewoldsen

Hoewe

Grady

(2022). A cognitive processing framework for media interpretation. Journal of Media Psychology, 34(2), 65–76. https://doi.org/10.1027/1864-1105/a000326

Flood

(2019, December 12). Fox News boasts best primetime ratings ever, finishes on top of basic cable for 4th straight year. Fox News. https://www.foxnews.com/media/fox-news-ratings-top-primetime

Gallup. (2018). Most important problem. https://news.gallup.com/poll/1675/most-important-problem.aspx

Gentzkow

Shapiro

(2006). Media bias and reputation. Journal of Political Economy, 114(2), 280–316. https://doi.org/10.1086/499414

Gil de Zúñiga

Correa

Valenzuela

(2012). Selective exposure to cable news and immigration in the US: The relationship between Fox News, CNN, and attitudes toward Mexican immigrants. Journal of Broadcasting & Electronic Media, 56(4), 597–615. https://doi.org/10.1080/08838151.2012.732138

10.

Gottfried

Barthel

Mitchell

(2017, January 28). Trump, Clinton voters divided in their main source for election news. Pew Research Center. https://www.journalism.org/2017/01/18/trump-clinton-voters-divided-in-their-main-source-for-election-news/

11.

Gramlich

(2020). 5 facts about Fox News. Pew Research Center. https://www.pewresearch.org/fact-tank/2020/04/08/five-facts-about-fox-news/

12.

Guess

Coppock

(2020). Does counter-attitudinal information cause backlash? Results from three large survey experiments. British Journal of Political Science, 50(4), 1497–1515. https://doi.org/10.1017/S0007123418000327

13.

Gunther

A. C.

Edgerly

Akin

Broesch

J. A.

(2012). Partisan evaluation of partisan information. Communication Research, 39(4), 439–457. https://doi.org/10.1177/0093650212441794

14.

Hayes

A. F.

(2017). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. Guilford Publications.

15.

Hoewe

(2018). Coverage of a crisis: The effects of international news portrayals of refugees and misuse of the term “immigrant.” American Behavioral Scientist, 62(4), 478–492. https://doi.org/10.1177/0002764218759579

16.

Hoewe

Brownell

Wiemer

E. C.

(2020). The role and impact of Fox News. The Forum: A Journal of Applied Research in Contemporary Politics, 18(3), 367–388. https://doi.org/10.1515/for-2020-2014

17.

Hoewe

Peacock

Kim

Barnidge

(2020). The relationship between Fox News use and Americans’ policy preferences regarding refugees and immigrants. International Journal of Communication, 14, 2036–2056. https://ijoc.org/index.php/ijoc/article/view/12402

18.

Holcomb

(2014). Five facts about Fox News. Pew Research Center. http://www.pewresearch.org/fact-tank/2014/01/14/five-facts-about-fox-news/

19.

Hollander

B. A.

(2018). Partisanship, individual differences, and news media exposure as predictors of conspiracy beliefs. Journalism & Mass Communication Quarterly, 95(3), 691–713. https://doi.org/10.1177/1077699017728919

20.

Hovland

C. I.

Janis

I. L.

Kelley

H. H.

(1953). Communication and persuasion. Yale University Press.

21.

Hwang

Kim

(2018). Influence of discussion incivility on deliberation: An examination of the mediating role of moral indignation. Communication Research, 45(2), 213–240. https://doi.org/10.1177/0093650215616861

22.

Iyengar

Hahn

K. S.

(2009). Red media, blue media: Evidence of ideological selectivity in media use. Journal of Communication, 59(1), 19–39. https://doi.org/10.1111/j.1460-2466.2008.01402.x

23.

Jamieson

K. H.

Capella

J. N.

(2008). Echo chamber: Rush Limbaugh and the conservative media establishment. Oxford University Press.

24.

Jurkowitz

Mitchell

Shearer

Walker

(2020, January 24). Democrats report much higher levels of trust in a number of news sources than Republicans. Pew Research Center. https://www.pewresearch.org/journalism/2020/01/24/democrats-report-much-higher-levels-of-trust-in-a-number-of-news-sources-than-republicans

25.

Katz

A. J.

(2022, January 3). 2021 ratings: After a slow start, Fox News completes another year as the dominant network on cable news. AdWeek. https://www.adweek.com/tvnewser/2021-ratings-after-slow-start-fox-news-completes-another-year-as-the-dominant-network-on-cable-news/496934/

26.

Knobloch-Westerwick

Meng

(2009). Looking the other way: Selective exposure to attitude-consistent and counterattitudinal political information. Communication Research, 36(3), 426–448. https://doi.org/10.1177/0093650209333030

27.

Kull

Ramsay

Lewis

(2003). Misperceptions, the media, and the Iraq War. Political Science Quarterly, 118(4), 569–598. https://doi.org/10.1002/j.1538-165X.2003.tb00406.x

28.

Kunda

(1990). The case for motivated reasoning. Psychological Bulletin, 108(3), 480–498. https://doi.org/10.1037/0033-2909.108.3.480

29.

Levendusky

(2013). Partisan media exposure and attitudes toward the opposition. Political Communication, 30(4), 565–581. https://doi.org/10.1080/10584609.2012.737435

30.

Meyer

(1988). Defining and measuring credibility of newspapers: Developing an index. Journalism Quarterly, 65(3), 567–574. https://doi.org/10.1177/107769908806500301

31.

Michaels

J. L.

Vallacher

R. R.

Liebovitch

L. S.

(2013). Volatile psychological dynamics in social interactions: Attitudes and emotions react asymmetrically to interaction shifts between agreement and disagreement. Social Psychological and Personality Science, 4(6), 705–713. https://doi.org/10.1177/1948550613482985

32.

Mitchell

Kiley

Gottfried

Matsa

K. E.

(2014). Political polarization & media habits. Pew Research Center. https://www.journalism.org/2014/10/21/political-polarization-media-habits/

33.

Morris

J. S.

(2005). The Fox News factor. Harvard International Journal of Press/Politics, 10(3), 56–79. https://doi.org/10.1177/1081180X05279264

34.

Morris

J. S.

(2007). Slanted objectivity? Perceived media bias, cable news exposure, and political attitudes. Social Science Quarterly, 88(3), 707–728. https://doi.org/10.1111/j.1540-6237.2007.00479.x

35.

Mourão

R. R.

Thorson

Kryston

Tunney

(2023). Predicting audience verification intention: The impact of partisanship, source, importance, and information familiarity on willingness to verify headlines. Journalism & Mass Communication Quarterly, 100(2), 308–331. https://doi.org/10.1177/10776990221130992

36.

Palan

Schitter

(2018). Prolific.ac—A subject pool for online experiments. Journal of Behavioral and Experimental Finance, 17, 22–27. https://doi.org/10.1016/j.jbef.2017.12.004

37.

Peer

Brandimarte

Samat

acquisti

(2017). Beyond the Turk: Alternative platforms for crowdsourcing behavioral research. Journal of Experimental Social Psychology, 70, 153–163. https://doi.org/10.1016/j.jesp.2017.01.006

38.

Pew Research Center. (2021). Cable news fact sheet. https://www.pewresearch.org/journalism/fact-sheet/cable-news/

39.

Pornpitakpan

(2004). The persuasiveness of source credibility: A critical review of five decades’ evidence. Journal of Applied Social Psychology, 342(2), 243–281. https://doi.org/10.1111/j.1559-1816.2004.tb02547.x

40.

Prolific Team. (2019, March 5). Representative samples on Prolific. Prolific Help Centre. https://researcher-help.prolific.co/hc/en-gb/articles/360019236753-Representative-Samples-on-Prolific-limited-release

41.

Sanders

(2022, April). Trust in media 2022: Where Americans get their news and who they trust for information. YouGov. https://today.yougov.com/topics/politics/articles-reports/2022/04/05/trust-media-2022-where-americans-get-news-poll

42.

Segesten

A. D.

Bossetta

Holmberg

Niehorster

(2022). The cueing power of comments on social media: How disagreement in Facebook comments affects user engagement with news. Information, Communication & Society, 25, 1115–1134. https://doi.org/10.1080/1369118X.2020.1850836

43.

Silvia

P. J.

(2006). Reactance and the dynamics of disagreement: Multiple paths from threatened freedom to resistance to persuasion. European Journal of Social Psychology, 36, 673–685. https://doi.org/10.1002/ejsp.309

44.

Simonov

Sacher

Dubé

J. P.

Biswas

(2021). The persuasive effect of Fox News: Noncompliance with social distancing during the COVID-19 pandemic. Marketing Science, 41, 230–242. https://doi.org/10.1287/mksc.2021.1328

45.

Slater

M. D.

(2007). Reinforcing spirals: The mutual influence of media selectivity and media effects and their impact on individual behavior and social identity. Communication Theory, 17(3), 281–303. https://doi.org/10.1111/j.1468-2885.2007.00296.x

46.

Slater

M. D.

(2015). Reinforcing spirals model: Conceptualizing the relationship between media content exposure and the development and maintenance of attitudes. Media Psychology, 18(3), 370–395. https://doi.org/10.1080/15213269.2014.897236

47.

Strömbäck

Tsfati

Boomgaarden

Damstra

Lindgren

Vliegenthart

Lindholm

(2020). News media trust and its impact on media use: Toward a framework for future research. Annals of the International Communication Association, 44(2), 139–156. https://doi.org/10.1080/23808985.2020.1755338

48.

Stroud

N. J.

(2011). Niche news: The politics of news choice. Oxford University Press.

49.

Stroud

N. J.

Muddiman

(2012). The American media system today: Is the public fragmenting? In Ridout

T. N.

(Ed.), New directions in media and politics (pp. 6–23). Routledge.

50.

Taber

C. S.

Lodge

(2006). Motivated skepticism in the evaluation of political beliefs. American Journal of Political Science, 50(3), 755–769. https://doi.org/10.1111/j.1540-5907.2006.00214.x

51.

Weaver

D. A.

Scacco

J. M.

(2013). Revisiting the protest paradigm: The Tea Party as filtered through prime-time cable news. The International Journal of Press/Politics, 18(1), 61–84. https://doi.org/10.1177/1940161212462872

52.

Westerwick

Johnson

B. K.

Knobloch-Westerwick

(2017). Confirmation biases in selective exposure to political online information: Source bias vs. content bias. Communication Monographs, 84(3), 343–364. https://doi.org/10.1080/03637751.2016.1272761

53.

A. J.

Wyer

R. S.

Jr. (2012). The role of bolstering and counterarguing mind-sets in persuasion. Journal of Consumer Research, 38(5), 920–932. https://doi.org/10.1086/661112

54.

Young

D. G.

(2020). Irony and outrage: The polarized landscape of rage, fear, and laughter in the United States. Oxford University Press.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

3.48 MB