Goidel, Davis and Goidel (2021) used a two-wave module of the 2016 Cooperative Congressional Election Study (N = 801) to show that individual perceptions of media bias polarized over the 2016 U.S. presidential campaign: strong Democrats moved toward seeing coverage as biased in favor of Donald Trump, independent-leaning Republicans moved toward seeing it as biased against him, and the aggregate mean barely moved. Whether these within-person shifts were lasting or temporary — the authors conjectured, from indirect evidence such as the post-election surge in newspaper subscriptions and recovered media trust among Democrats and Independents, that the Democratic shift may have been short-lived — and whether the inferred mechanisms (sustained elite anti-media cues for Republicans; the surprising election outcome for attentive Democrats), rather than alternatives such as changing media content, partisan selective exposure, or social influence, actually account for the changes, remains unresolved. No published study has re-measured perceived-bias trajectories after 2016 or directly tested the mechanism attributions. The question is open.

Background

Research on the hostile media phenomenon — the finding, originating with Vallone, Ross and Lepper (1985), that partisans perceive the same news coverage as biased against their own side — has mostly examined perceptions at a single point in time or in short laboratory settings. Much less is known about how an individual's perception of media bias changes over the course of a real election campaign. Goidel, Davis and Goidel (2021) addressed this gap with a two-wave panel module fielded as part of the Cooperative Congressional Election Study (CCES), an annual U.S. opt-in Internet survey with sample matching: 801 respondents were interviewed before the 2016 presidential election (October 4 to November 6, 2016) and again after it (November 9 to December 12, 2016), each time rating news coverage of the campaign on an 11-point scale from 0 (biased in favor of Hillary Clinton) through 5 (fair and balanced) to 10 (biased in favor of Donald Trump). The 2016 context made the question especially pointed: Donald Trump repeatedly claimed that media bias was part of a rigged political system, and his victory surprised most observers who had expected a Clinton win.

The paper's central finding is polarization hidden behind aggregate stability. The sample mean was nearly unchanged from before to after the election (roughly 3.60 to 3.74), and about 60% of respondents changed their answer, but the individual-level changes pulled partisans in opposite directions: strong Democrats shifted most, on the order of +0.41 scale points toward perceiving pro-Trump bias (the largest and statistically significant movement), while independent-leaning Republicans — respondents who identify as political independents but say they lean toward the Republican Party — shifted toward perceiving anti-Trump bias. Political interest moderated these shifts. In a change-model specification (a regression estimated on individual-level change scores rather than levels), the strong-Democrat shift was significant only among the most attentive respondents, whereas the Republican-side movement appeared among less-attentive leaning Republicans. The authors read this pattern as consistent with two different mechanisms operating simultaneously: attentive Democrats updating their views in response to the surprising outcome, and less-attentive Republicans following sustained elite cues — that is, taking direction from statements by party leaders and figures, here Trump's repeated claims of a hostile, 'rigged' media. Crucially, these attributions were inferred from the moderation pattern, not tested directly: the study measured neither exposure to elite messaging nor media consumption.

The paper's discussion leaves the durability question explicitly unresolved. The authors note that available indirect evidence — increased subscription rates for major newspapers after the election and a recovery of media trust among Democrats and Independents — suggests the Democratic shift may have been short-term, while the elite-cue-driven Republican shift, on this reading, would persist as long as the cues do. This is the open question the present record addresses.

Problem Statement

Determine what became of the 2016 polarization in perceived media bias documented by Goidel, Davis and Goidel (2021), and why it happened. Two jointly posed components, as framed by the source study's own unresolved discussion: (1) Durability — did the within-person shifts observed between the pre-election and post-election waves of the 2016 CCES module (strong Democrats moving toward perceived pro-Trump bias, independent-leaning Republicans — respondents who identify as political independents but lean Republican — moving toward perceived anti-Trump bias) persist, revert, or continue to polarize over subsequent years and election cycles (2017-2024), within each partisan subgroup? The authors' conjecture that the Democratic shift was short-term, based on indirect evidence such as subscription surges and recovered Democratic and Independent media trust, is an untested hypothesis, not a finding. (2) Mechanisms — do sustained elite anti-media cues, i.e. repeated statements by party leaders attacking the news media (the proposed Republican-side channel), the unexpected 2016 outcome (the proposed Democrat-side channel), or instead alternative factors — changing media content, partisan selective exposure (the tendency to consume news mainly from outlets aligned with one's own party), affective polarization (dislike of the opposing party and its supporters), or social influence — account for the observed changes? The paper inferred the first two from an interest-moderation pattern without measuring exposure to elite messaging or media consumption, and the attribution remains untested. The durability component bears on bias-perception measurements of the same or demonstrably equivalent construct taken at repeated points after 2016, and the mechanism component bears on evidence capable of distinguishing the candidate mechanisms.

References

  1. Kirby Goidel, Nicholas T. Davis, and Spencer Goidel (2021), 'Changes in perceptions of media bias', Research & Politics 8(1). DOI: 10.1177/2053168020987441. https://doi.org/10.1177/2053168020987441
  2. Robert P. Vallone, Lee Ross, and Mark R. Lepper (1985), 'The hostile media phenomenon: Biased perception and perceptions of media bias in coverage of the Beirut massacre', Journal of Personality and Social Psychology 49(3), 577-585. DOI: 10.1037/0022-3514.49.3.577. https://doi.org/10.1037/0022-3514.49.3.577
  3. Emily Van Duyn and Jessica Collier (2019), 'Priming and Fake News: The Effects of Elite Discourse on Evaluations of News Media', Mass Communication and Society 22(1), 29-48 (first published online 12 September 2018). DOI: 10.1080/15205436.2018.1511807. https://doi.org/10.1080/15205436.2018.1511807
  4. Allison M. N. Archer (2023), 'The Effects of Elite Attacks on Copartisan Media: Evidence from Trump and Fox News', Public Opinion Quarterly 87(4), 887-910. DOI: 10.1093/poq/nfad042. https://doi.org/10.1093/poq/nfad042
  5. Allison M. N. Archer and Erik Peterson (2025), 'The Fragility of the Local News Trust Advantage: Evidence from Republican Attacks on Local News', PS: Political Science & Politics 58(2), 177-183. DOI: 10.1017/S1049096524000775. https://doi.org/10.1017/S1049096524000775
  6. Jonathan M. Ladd and Alexander R. Podkul (2020), 'Sowing Distrust of the News Media as an Electoral Strategy', in The Oxford Handbook of Electoral Persuasion, Oxford University Press, 426-454. DOI: 10.1093/oxfordhb/9780190860806.013.17. https://doi.org/10.1093/oxfordhb/9780190860806.013.17
  7. Jane Suiter and Richard Fletcher (2020), 'Polarization and partisanship: Key drivers of distrust in media old and new?', European Journal of Communication 35(5), 484-501. DOI: 10.1177/0267323120903685. https://doi.org/10.1177/0267323120903685
  8. Youngju Kim, Hyunseo Hwang, and Yonghwan Kim (2023), 'The "Us vs. Them" Mentality: The Role of Affective Polarization in Deepening the Partisan Divide in Media Bias Perception', Journalism & Mass Communication Quarterly (published online 25 November 2023; issue assignment pending at time of record). DOI: 10.1177/10776990231211804. https://doi.org/10.1177/10776990231211804
  9. Erik Peterson and Ali Kagalwala (2021), 'When Unfamiliarity Breeds Contempt: How Partisan Selective Exposure Sustains Oppositional Media Hostility', American Political Science Review 115(2), 585-598. DOI: 10.1017/S0003055420001124. https://doi.org/10.1017/S0003055420001124
  10. Weina Ran and Masahiro Yamamoto (2024), 'Neutral news from in- and out-party media and attitudes toward them: integration of expectancy violations theory and hostile media perceptions', Communication Studies 75(6), 880-899. DOI: 10.1080/10510974.2024.2353393. https://doi.org/10.1080/10510974.2024.2353393