Does partisan news exposure change how citizens vote, or does it merely reflect preexisting preferences? Dilliplane (2014, American Journal of Political Science) used multiwave 2008 U.S. panel data with within-person fixed-effects models to find that like-minded partisan television raises the odds that wavering partisans return to their party's candidate (activation) and lowers conversion to the opposing candidate, while counter-attitudinal exposure pushes in the opposite direction. The causal status of these activation and conversion effects remains unresolved: later experiments that let subjects choose their media and quasi-experimental studies of channel availability reach conflicting conclusions; effects on vote choice (as opposed to attitudes) have not been causally identified under realistic, self-selected exposure; the microprocesses behind counter-attitudinal conversion are unknown; moderator evidence is mixed; and replication beyond 2008 U.S. television is sparse (a Brazilian web-tracking panel finds no vote-choice effect). The central question is whether the activation and conversion effects documented in that observational panel analysis survive credible causal identification, with related open sub-questions — how conversion occurs, for whom effects are conditioned, and in which media environments they generalize — remaining open.

Background

Research on mass communication and elections has long swung between 'minimal effects' and 'powerful effects' views. In panel studies of the 1940 U.S. election, Lazarsfeld and colleagues introduced the typology still used today: activation (moving undecided or defecting partisans back to their own party's candidate), conversion (moving a supporter of one candidate to the opposing candidate), and reinforcement (strengthening an existing preference without changing the vote). Their conclusion — that campaign media mostly reinforce — defined the minimal-effects era.

The modern high-choice media environment reopened the question. Television channels and later digital outlets with explicit ideological slant ('partisan news', e.g., Fox News and MSNBC in the United States) make both like-minded and opposing content readily available, while selective exposure — the tendency of audiences to choose congenial content — limits the counter-attitudinal exposure that classical persuasion requires. Aggregate evidence that Fox News's introduction raised Republican vote share (DellaVigna and Kaplan 2007) suggested media power, but could not say which individuals changed or why.

Dilliplane (2014) supplied the leading individual-level evidence. She used a multiwave panel from the 2008 National Annenberg Election Study (NAES), a national survey that re-interviewed the same respondents across the campaign. Exposure was measured as the number of regularly watched programs among 45 top-rated TV news programs, each classified as Democratic-, Republican-, or neutral-slanted using a separate national perception survey (about 45,105 respondents) plus content searches; the program-list measure's reliability and validity were established in companion work (Dilliplane, Goldman, and Mutz 2013). The analysis used within-person fixed-effects logit models — regressions that use only changes within the same individual over time and thus control for all stable unobserved traits — with activation and conversion (vote switches between adjacent waves) as outcomes, and linear fixed-effects models of change in candidate favorability on a 0–100 feeling thermometer (a survey warmth rating) for reinforcement. The findings: only modest reinforcement, but robust effects on vote change — exposure to own-party-slanted news increased activation odds and decreased conversion odds, while counter-attitudinal (opposing-party) exposure pushed the other way.

The study itself flags what remains unresolved, as five linked issues: (i) causal direction — reverse causation, in which campaign events or attitude change drive exposure changes, cannot be excluded without experimental manipulation that preserves realistic selective-exposure patterns; (ii) the microprocesses by which counter-attitudinal programming converts (cognitive acceptance, affective arousal, subtle cues, or differential perceptions of bias) are unspecified; (iii) whether political knowledge or partisan strength conditions susceptibility remains mixed and underpowered; (iv) exposure measurement (monthly program counts, static slant coding) limits precision; and (v) generalizability to other cycles, media (radio, web, social platforms), and non-U.S. contexts is untested.

Subsequent designs attacked parts of this agenda without settling it. Experiments incorporating subject choice — Arceneaux and Johnson's (2013) forced-exposure versus free-choice studies and Levendusky's (2013) experiments showing partisan media polarize mainly by making already-extreme viewers more extreme — and the Preference-Incorporating Choice and Assignment (PICA) design of de Benedictis-Kessner et al. (2019), which jointly estimates polarization caused by selective exposure and the persuasive effect of forced exposure, measure attitude polarization rather than vote choice. Quasi-experiments reach conflicting conclusions: Hopkins and Ladd (2014) find Fox News availability reinforces co-partisans with an average persuasion effect indistinguishable from zero, whereas Martin and Yurukoglu (2017), using cable channel positions as instruments (variables that shift viewership but are plausibly unrelated to political preferences), estimate nonzero persuasion effects on vote shares. No existing design identifies effects on individual vote change under realistic, self-selected media diets.

Problem Statement

The unresolved question is whether partisan news exposure causally changes citizens' vote choices during a campaign — via activation (wavering partisans returning to their party's candidate), conversion (switching to the opposing candidate), or reinforcement (intensifying unchanged preferences) — rather than merely reflecting prior preferences or campaign dynamics. The central open empirical target is to identify, with a design — experimental or quasi-experimental — that preserves realistic, self-selected media diets and rules out reverse causation (campaign events or preference change driving exposure change), whether the activation and conversion effects documented in observational panel analysis (Dilliplane 2014) survive credible causal identification, distinguishing activation from conversion. Here 'credible causal identification' denotes a design in which exposure variation is exogenous to the vote outcome conditional on explicitly stated, in-principle testable assumptions, with the known threats to these panel estimates — notably exposure measurement error and time-varying confounding — addressed. For activation and for conversion separately, the target is two-sided: a design under which that effect survives credible causal identification, or the finding that no design satisfies both constraints — with the verdict in either case assumption-conditional on the design's explicitly stated, in-principle testable assumptions. Related open sub-questions remain open: whether, how, for whom, and in which media environments partisan news exposure changes votes — the microprocesses by which counter-attitudinal exposure produces conversion, separating cognitive acceptance of arguments, affective arousal, source cues, and differential perceptions of bias; whether susceptibility is systematically conditioned by political knowledge and partisan strength, and how large any such moderation is; and how far the 2008 U.S. television findings extend to other election cycles, media platforms (radio, web, social media), and media systems.

References

  1. Dilliplane, Susanna (2014). Activation, Conversion, or Reinforcement? The Impact of Partisan News Exposure on Vote Choice. American Journal of Political Science 58(1): 79–94. DOI: 10.1111/ajps.12046. https://doi.org/10.1111/ajps.12046
  2. DellaVigna, S.; Kaplan, E. (2007). The Fox News Effect: Media Bias and Voting. The Quarterly Journal of Economics 122(3): 1187–1234. DOI: 10.1162/qjec.122.3.1187. https://doi.org/10.1162/qjec.122.3.1187
  3. Dilliplane, Susanna; Goldman, Seth K.; Mutz, Diana C. (2013). Televised Exposure to Politics: New Measures for a Fragmented Media Environment. American Journal of Political Science 57(1): 236–248. DOI: 10.1111/j.1540-5907.2012.00600.x. https://doi.org/10.1111/j.1540-5907.2012.00600.x
  4. Prior, M. (2013). The Challenge of Measuring Media Exposure: Reply to Dilliplane, Goldman, and Mutz. Political Communication 30(4): 620–634. DOI: 10.1080/10584609.2013.819539. https://doi.org/10.1080/10584609.2013.819539
  5. Goldman, Seth K.; Mutz, Diana C.; Dilliplane, Susanna (2013). All Virtue Is Relative: A Response to Prior. Political Communication 30(4): 635–653. DOI: 10.1080/10584609.2013.819540. https://doi.org/10.1080/10584609.2013.819540
  6. Arceneaux, K.; Johnson, M. (2013). Changing Minds or Changing Channels? Partisan News in an Age of Choice. University of Chicago Press. DOI: 10.7208/chicago/9780226047447.001.0001. https://doi.org/10.7208/chicago/9780226047447.001.0001
  7. Levendusky, Matthew S. (2013). Why Do Partisan Media Polarize Viewers? American Journal of Political Science 57(3): 611–623. DOI: 10.1111/ajps.12008. https://doi.org/10.1111/ajps.12008
  8. Hopkins, Daniel J.; Ladd, Jonathan M. (2014). The Consequences of Broader Media Choice: Evidence from the Expansion of Fox News. Quarterly Journal of Political Science 9(1): 115–135. DOI: 10.1561/100.00012099. https://doi.org/10.1561/100.00012099
  9. Conroy-Krutz, Jeffrey; Moehler, Devra C. (2015). Moderation from Bias: A Field Experiment on Partisan Media in a New Democracy. The Journal of Politics 77(2): 575–587. DOI: 10.1086/680187. https://doi.org/10.1086/680187
  10. Smith, Glen (2016). The Timing of Partisan Media Effects during a Presidential Election. Political Research Quarterly 69(4): 655–666. DOI: 10.1177/1065912916656409. https://doi.org/10.1177/1065912916656409
  11. Martin, Gregory J.; Yurukoglu, Ali (2017). Bias in Cable News: Persuasion and Polarization. American Economic Review 107(9): 2565–2599. DOI: 10.1257/aer.20160812. https://doi.org/10.1257/aer.20160812
  12. de Benedictis-Kessner, Justin; Baum, Matthew A.; Berinsky, Adam J.; Yamamoto, Teppei (2019). Persuading the Enemy: Estimating the Persuasive Effects of Partisan Media with the Preference-Incorporating Choice and Assignment Design. American Political Science Review 113(4): 902–916. DOI: 10.1017/S0003055419000418. https://doi.org/10.1017/S0003055419000418
  13. Bergan, Daniel E.; Lacy, Stephen; Carnahan, Dustin; Stamm, Michael; Krier, Daniel H. (2021). Reinforcement in the Aggregate: Partisan Newspaper Circulation and the Presidential Vote, 1900–1928. Journalism Studies 22(14): 1911–1929. DOI: 10.1080/1461670X.2021.1976074. https://doi.org/10.1080/1461670X.2021.1976074
  14. Wlezien, Christopher (2024). News and Public Opinion: Which Comes First? The Journal of Politics 86(1): 1–17. DOI: 10.1086/726940. https://doi.org/10.1086/726940
  15. Mont’Alverne, Camila; Arguedas, Amy Ross; Banerjee, Sayan; Toff, Benjamin; Fletcher, Richard; Nielsen, Rasmus Kleis (2025). Exposure to Partisan News and Its Impact on Social Polarization and Vote Choice: Evidence From the 2022 Brazilian Elections. The International Journal of Press/Politics 30(4): 904–930. DOI: 10.1177/19401612241292700. https://doi.org/10.1177/19401612241292700
  16. Baum, Matthew; Berinsky, Adam J.; de Benedictis-Kessner, Justin; Jeong, Joohye; Markovich, Zachary; Yamamoto, Teppei (2026). Dynamic persuasion: decay and accumulation of partisan media persuasion. Political Science Research and Methods (first view): 1–17. DOI: 10.1017/psrm.2026.10095. https://doi.org/10.1017/psrm.2026.10095