The ORE of Politics will be available for subscription in late September. Speak to your Oxford representative or contact us to find out more.

Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, POLITICS ( (c) Oxford University Press USA, 2016. All Rights Reserved. Personal use only; commercial use is strictly prohibited. Please see applicable Privacy Policy and Legal Notice (for details see Privacy Policy).

date: 17 August 2017

Partisan Media and Polarization: Challenges for Future Work

Summary and Keywords

Four potential mechanisms explore the linkages between partian media outlets and attitudinal polarization, as well as discusses how such outlets cause polarization and influence American politics more generally: partisan media outlets can have direct effects on their audience, indirect effects on the broader population, effects on the news media, and effects on political elites. Some challenges and questions remain to be answered in each area in the hopes of spurring more, and broader, work on these media institutions.

Keywords: partisan media, media effects, polarization, social network


Over the past few years, there has been an explosion of interest in partisan media outlets, particularly in light of their potential role in polarization. Partisan media outlets are those that eschew standard journalistic norms of objectivity and balance in favor of a more one-sided presentation of the facts (see Levendusky, 2013, chapter 1, for more discussion). The main exemplars are cable television networks such as Fox News and MSNBC, talk radio hosts such as Rush Limbaugh, and Internet websites (especially blogs) such as Instapundit or Daily Kos (Levendusky, 2013; Jamieson & Cappella, 2008; Stroud, 2011).1 Such outlets, with their slanted presentation of the facts of the day, have an obvious and straightforward connection to polarization: if people watch these outlets, and hear only one side of the issue, then they may well become more polarized (Mutz, 2007).2

A number of studies have investigated this hypothesized relationship, and while they find that partisan outlets contribute to polarization, this effect is not as large as many had initially feared. Most Americans simply tune out these networks, and so they never hear these partisan messages (Arceneaux & Johnson, 2013). Most people who watch these outlets are already quite polarized even before they watch (Stroud, 2011). Data suggest that these outlets make their somewhat polarized viewers even more so. They do not, however, turn moderates into extremists (Levendusky, 2013).3 The partisan media audience was small and extreme before exposure, and it is small and somewhat more extreme afterward.

This pathway—increasing the polarization of its already interested and polarized audience, a consequence of direct exposure to these outlets—represents an important mechanism through which partisan media influence polarization and public opinion more generally. The audience for these outlets is small, but it is more politically engaged and politically influential than the general public. Affecting this small audience can therefore have a significant effect on important political outcomes (Prior, 2013).

To date, most of the published work on partisan media focuses on this general topic: how do these outlets shape the attitudes and behaviors of their audience? But this does not represent the only pathway for these outlets to impact American politics. There are at least three additional pathways to influence. First, partisan media messages can diffuse from this relatively modest audience to the broader public via social networks. Because the audience for these programs are more politically engaged and informed, they are likely to spread their messages to others via interpersonal discussion, as well as through Facebook, Twitter, and other online venues. So even people who themselves never watch these programs can be exposed to these messages indirectly, raising the possibility of much broader influence. Second, partisan media outlets may influence the broader news agenda, changing what stories other networks cover and how they cover them. Third, these outlets can also influence political elites, rather than just voters. In short, partisan media can shape political outcomes by doing far more than just directly influencing their audience.

Before beginning in earnest, two notes are in order. First, the article presumes readers are already familiar with the basic contours of the literature on this topic. Those who are not should see Prior (2013), which provides an outstanding overview of the main questions and works addressing the consequences of partisan media outlets. Second, in what follows, the author’s own work is discussed at some length. This reflects (hopefully) not self-absorption, but rather that, especially when criticizing work, it is easier to identify the flaws in one’s own work than in others’ work.

Identifying People Who Actually Use Partisan Media

A generation ago, Shanto Iyengar and Donald Kinder (1987) revolutionized the study of mass media and politics by using random assignment to experimentally explore how media shaped political attitudes. At the time of their classic studies (the late 1970s and early 1980s), the nation’s dominant source of information was television, particularly television news. So simply randomly assigning subjects to watch different news broadcasts allowed researchers to recover the key parameters of interest (i.e., how news shapes opinions). There was much less concern about self-selection and selective exposure because there were only three major networks, and they all broadcast the same type of neutral, objective news.

Today, however, the situation is quite different, especially when studying partisan media. It is a particular person who wants to watch partisan television news (or listen to talk radio, or visit partisan websites), and that person tends to be older, more politically informed and interested, and more partisan and ideologically extreme (e.g., Stroud, 2011). Simply bringing a random sample of subjects into a lab and showing them Fox News or MSNBC will not allow us to estimate the effects of interest. Many people in the sample would never voluntarily watch partisan media, so their reactions are not terribly interesting for understanding the political consequences of these outlets.

Scholars have attempted to modify their experimental designs to address these concerns. For example, Levendusky (2013) asks subjects for their preferred news outlet in a pre-test survey, and then estimates effects for those who prefer to watch partisan news vs. those who do not (and finds larger effects for those who like to watch). Arceneaux and Johnson (2013) employ a similar strategy, and then also very cleverly allow subjects to choose which programs they watch during the experiment (i.e., instead of being forced to watch a particular program, subjects choose which program they will watch). Both strategies are an effort to identify the effects of partisan media on those who would actually watch it in the real world (more generally, see Gaines & Kuklinski, 2011).

While these studies are a real advance, they still likely represent an overestimate on the number of people who would choose to consume these types of media outlets (Prior, 2013). As both Levendusky and Arcenaux and Johnson acknowledge, given the relatively limited choices offered to subjects in their experiments, it is perhaps not surprising that many people chose to watch partisan media. This could reflect a political Hawthorne effect: in the lab, when subjects know (or at least reasonably suspect) that their choices are being watched, they consume partisan media outlets (and news more generally). Whether they would continue to do so in the real world, with its vastly wider array of choices, is more difficult to say.

One might be tempted to try and address this problem by using passively observed data (e.g., web traffic data, as in Gentzkow & Shapiro, 2011). However, because such data often lacks individual-level covariates such as partisanship and ideology, they are often ill-suited to testing the sort of theories most of interest to political scientists.

Unfortunately, there are no easy answers to this problem, and without accurately identifying the correct population, it is difficult to know if scholars are estimating the correct experimental effects. Part of the solution may be to try to recruit experimental subjects who are more likely to be partisan media consumers (i.e., trying to recruit subjects from political organizations, or who fit the demographic profile of the partisan media audience). However, such individuals are typically not interested in taking part in university research studies, thereby exacerbating the issue. Another alternative would be to try and identify natural experiments that allow scholars to test relevant theories (e.g., Della Vigna & Kaplan, 2007), but such examples are relatively uncommon as well.

Issue Selection and Generalizing beyond the Lab

Even after identifying the correct sample of individuals, there are other important challenges to studying partisan media effects. When studying partisan media effects experimentally, many scholars use moderately salient issues where opinion is more malleable. This is a sensible strategy. As Bartels (1993) argued two decades ago, media effects are more likely on certain types of issues than others: they’re more likely on emerging issues, or issues where respondents have relatively weak prior opinions (see also Druckman & Leeper, 2012). So, for example, the odds of changing a respondents’ opinion on (say) abortion are likely quite low, but it should be possible to change them on (say) the risks of genetically modified crops. After all, one is a long-standing issue where the parties have staked out clear and salient positions, so most people’s attitudes will be firmly anchored in place, while the other is a new issue that has received little national attention from either the Democratic or the Republican Party, and most respondents probably barely have an opinion on the topic. The issued used in the stimulus matter.

But this, in turn, raises a more fundamental point: on what range of issues can partisan media shape attitudes? Further, how does the magnitude of the effect depend on the issue itself? One can imagine a theory of issues that explained why some issues are more susceptible to partisan media persuasion than others. Surely Bartels’s (1993) observations on attitude crystallization are crucial, but other dimensions are also central as well: whether the issue is easy or hard (Carmines & Stimson, 1980), whether the issue is owned by one of the major political parties (Petrocik, 1996), whether the issue can be framed in moral terms (Clifford & Jerit, 2013), and so forth. There are any number of factors that will shape why partisan media outlets can persuade audiences on one topic but not another. Exploring this sort of cross-issue heterogeneity is an important next step toward a more general understanding of these effects (for more on this point, see Levendusky, 2014; Arceneaux & Johnson, 2015).

Further, many existing studies are simply one-shot experimental designs, where subjects come to the lab and receive a single dosage of partisan media content. Given the difficulty of identifying subjects willing to participate even in a brief one-time experiment, this is perfectly understandable. However, in the real world, subjects are exposed to repeated doses over time. As the study of framing demonstrates, introducing message repetition can sharply change the contours of opinion and the effectiveness of political media (Chong & Druckman, 2010). There have been a few studies that look at repeated exposure, and they find that effects get smaller over time upon repeated exposure (see, e.g., Levendusky, 2013, chapter 4). But much more remains to be done, and over longer time frames. As Martin and Yurukoglu (2014) show, repeated brief exposure can have quite powerful effects, at least for certain audiences. So to actually understand the longer-term effects of partisan outlets, scholars need a different set of designs.

At some level, this discussion of the challenges of studying the partisan media audience is simply a particular discussion of external validity, and ensuring that lab/survey experiments generalize to the “real world” of interest. Much of the existing work establishes a key set of baseline effects. The challenge now is to extend them and consider how these other sorts of factors (like the audience or the issues used in studies of partisan media) condition these effects.

Effects for Those Who Don’t Watch Partisan Media

The arguments above focus on the effects of partisan media on its audience. But of course, as is well known, the audience for these programs in the U.S. is relatively modest, numbering a few million people a night in a nation of 300 million. But part of the reason these outlets are important is that their effects extend beyond their immediate audience. The partisan media audience is a well-informed and politically influential audience, and hence, is likely to spread its messages to others in their social networks both in person and online (Levendusky, 2013). This means that many people who themselves would never turn on left-leaning television host Rachel Maddow or right-wing host Sean Hannity are actually exposed to the messages they disseminate on their shows. Of course, this is not a new idea: it goes back at least fifty years to the idea of two-step communication flows (Berelson, Lazarsfeld, & McPhee, 1954; Katz & Lazarsfeld, 1955). What’s new here is the introduction of a partisan media message, which might spread through a network somewhat differently.

To the best of my knowledge, there is only one (unpublished) study that tackles this topic (Druckman, Levendusky, & McLain, 2015). There, the authors randomly assign some people to watch a partisan media stimulus and others to watch apolitical content. After watching their content, some respondents are then randomly assigned to discussion groups to talk about the media they just watched. This allows the authors to isolate not only the direct effect of exposure, but also the effects of indirect exposure as well (i.e., the effect of not watching partisan outlets but then discussing an issue with those who did). The study finds that those who do not watch partisan media outlets but then discuss them with others end up polarized, just like those who watch the programs. Discussing the programs can generate polarization, just as exposure itself can. While this is just one study, it establishes that the effects of partisan media can spread beyond its relatively modest audience.

The study above is an important first step, but it is just that: a first step. Druckman, Levendusky, and McLain (2015) study how the influence of partisan outlets spread through social networks via face-to-face discussion. But that is not the only way that people are exposed to partisan media messages. They are also exposed frequently via social media, especially Facebook and Twitter. Indeed, many partisan media stories are repeatedly shared on social media outlets, thereby amplifying the reach of these outlets. This brings up two important points. First, this gives a partial return to the old idea of incidental exposure to information (Downs, 1957). One of the hallmarks of the broadcast era was that people were exposed to political information even if they did not search it out: even if they disliked politics, they saw it when they read the paper to get the sports scores or watched the news to see the weather. As Prior (2007) notes, the dawn of the high-choice media environment largely ended this incidental exposure for the apolitical, but the rise of online social media could revive this concept. If your Facebook feed has political content, you will see it even if you yourself are apolitical and apathetic. What political consequences this has remain to be seen.

Second, not only are people exposed to political information online, they are also exposed to counter-attitudinal information (i.e., information that diverges from their prior beliefs, such as when a Republican is exposed to Rachel Maddow). Practically everyone has someone—a cousin, an old coworker, a friend from high school—who posts counterattitudinal information (much of it from partisan outlets) in their Facebook feed. While it is true that individuals try to avoid that information, and Facebook’s algorithms also aid in that process (Bakshy, Messing, & Adamic, 2015), people are not able to avoid it completely. Do people actually engage with this information, or do they simply tune it out? If they engage with this content, does it affect their attitudes? No extant research can really answer this question.

More generally, the study of exposure to partisan information on the Internet is a largely untapped resource for scholars. Much of what we know about partisan media comes from studies of cable TV news (Arceneaux & Johnson, 2013; Levendusky, 2013; Stroud, 2011; Baum & Groeling, 2010) or talk radio (Barker, 1999; Jamieson & Cappella, 2008). While there is some work addressing the consequences of partisan exposure to information on the Internet, it is far less common (though for important exceptions, see Garrett, 2009; Gentzkow & Shapiro, 2011; Lelkes, Sood, & Iyengar, in press). There are at least two major challenges here. First, there’s the problem of identifying who actually sees a message online. Measuring unique website visits is extremely difficult (Hindman, 2008), and even if someone visits a website, how do we know if they actually consumed the relevant partisan content? This is somewhat easier on social media platforms, where we know what is in someone’s feed (though again, we do not know if they actually read that content). Further, there is a difficulty in separating out the effects of site-generated content and user comments, which can have quite different dynamics (Suhay, Blackwell, Roche, & Bruggeman, 2015). In short, deciding on what the “information” is, and who sees it, is often not straightforward online.

Second, there’s the problem of understanding how online behavior translates into offline behavior. Scholars have made progress in using social media data to understand people’s political opinions as expressed online (e.g., Barbera, 2015; Bond & Messing, 2015). But do these behaviors correspond to their offline (real world) attitudes and behavior? Scholars need to do more to link the treasure trove of information from these outlets with key political variables measured elsewhere to push forward this type of research.

Effects on the Media Agenda

There is still another mechanism through which partisan outlets could shape opinion: by influencing other news organizations and shaping the news agenda. A number of classic studies argues that news outlets influence one another: if one outlet carries a story, others are more likely to pick it up as well (Roberts & McCombs, 1994; Boyle, 2001). Such studies, however, have typically focused on how (say) newspaper coverage influences television coverage. But no studies have examined how the decision of Fox or MSNBC—or other partisan sources, especially online—to discuss a story influences the decision of other outlets to pick up that story. We know that partisan outlets—particularly blogs—bias the news they present to viewers by selecting stories that are more favorable to their point of view (Baum & Groeling, 2008). But if partisan outlets cover a story, does that influence other media outlet’s behavior? Do partisan media outlets help shape the national news agenda?

This influence over the agenda is particularly important because of the slant of these outlets. If, say, ABC News influences NBC News or The Washington Post—all of them non-partisan—it is unlikely that new partisan frames will be introduced in the process. But if Fox News or MSNBC influences one of these outlets, new partisan perspectives and stories are most likely introduced into the mainstream media conversation. If partisan media outlets shape the mainstream news agenda, it is another mechanism through which they can influence those who never watch them.

Partisan media could (at least in theory) shape the broader news agenda in two ways. First, they could promulgate and promote a particular frame or perspective on an existing story. For example, during the health care debate in 2009, some at Fox News took to calling the “public option” the “government option” (Brock, Rabin-Havt, & Media Matters for America, 2012, pp. 85–91). While in this case the frame does not seem to have spread very widely, in other cases, it is certainly possible to imagine that a frame from partisan outlets could be picked up by other news outlets.

Second, not only could particular frames become part of the conversation, particular stories pushed on these networks could also spread to the broader mainstream media. For example, during the 2008 campaign, the Huffington Post reported President Obama’s remarks about “bitter” blue-collar voters who “cling to guns or religion,” and the story exploded from there (Fowler, 2010). Some have also argued that Fox News played an important role in promoting and disseminating the alleged wrongdoing at ACORN that lead to its downfall (Dreier & Martin, 2010). Similarly, when the Swift Boat Veterans for Truth initially aired their findings in the spring of 2004, the mainstream media largely ignored them. It was not until partisan media outlets, particularly Fox News, picked up the story over the summer that it really grew into a major news event (Shaw, 2006). We cannot know that coverage on partisan outlets caused these sorts of stories to spread—we never observe the counterfactual world where these outlets do not air these stories. But it seems reasonable to conclude that their prominence on these outlets played a role in these stories becoming as important as they became.

Even if a story does not begin on partisan news outlets, partisan outlets can play a key role in sustaining these stories moving forward. Skocpol and Williamson (2012) argue that conservative media outlets helped to sustain the rise of the Tea Party (see also Brock, Rabin-Havt, & Media Matters for America, 2012). One could argue that the ongoing conversation about media bias is another example of this phenomenon. While most academic studies conclude there is relatively little media bias (e.g., D’Alessio & Allen, 2000, though see Groseclose, 2011), such charges are relatively common in partisan outlets, especially on the right (see the discussion in Levendusky, 2013, chapter 2). Alleging media bias—and pointing it out to viewers—has two important ramifications. First, it reminds viewers of why they want to watch your outlet: to get the real story, away from the media spin, so it helps to boost (or at least maintain) ratings. Second, it forces mainstream outlets to respond to these charges of bias, a point long recognized by political operatives (Perlstein, 2008; Domke, Watts, Shah, & Fan, 1999). In effect, by charging that the mainstream media are biased, partisan media outlets can help to shape the broader conversation around media bias. This same pattern may play out on other issues as well.

Part of the difficulty in studying how partisan media outlets influence the broader news agenda is that it is difficult to observe “non-events.” It is easy to point to isolated examples where these outlets first report a story that becomes news on more convectional outlets. But finding examples of cases where these outlets try to push a story and it fails are more difficult to find for two reasons. First, many of these nonstories are likely ephemeral: partisan outlets try a few times to promote a story, but then if it does not work, the story likely fades from view. Second, it is quite difficult to determine when partisan outlets are simply following other outlets, and when they are having a genuine influence on the contours of the story. The influence of partisan outlets is typically more conditional and modest than in the examples above, but identifying and documenting these subtler cases will be difficult. Difficult, however, does not necessarily mean impossible. Especially with the advent of online repositories such as the UCLA Communication Studies Archive, and the increasing availability of full-text transcripts in archives such as Lexis-Nexis, it is now possible to do the sort of detailed content analysis that would be necessary to identify both successes and failures of partisan media outlets influencing the broader news agenda.

Effects on Political Elites

Partisan outlets can also affect the behavior of political elites as well. For example, several recent papers find that the entry of Fox News in the mid-1990s changed the voting behavior of members of Congress (Clinton & Enamorado, 2014; Arceneaux, Johnson, Lindstadt, & Vander Wielen, 2016). It is unclear from the data whether this is because of the effect of the network on voters (Della Vigna & Kaplan, 2007; Hopkins & Ladd, 2014) or because the network more directly changed politicians’ incentives.

These outlets shape elite politics in other ways as well. In particular, the rise of such stations permits political elites to “narrowcast” messages in a new way (Baum, 2011). For example, during the 2012 primary election season, nearly all of the Republican nominees for president made at least a few appearances on Fox News. Indeed, Sean Hannity hosted an event on his show he called the “Hannity Primary” where he interviewed potential Republican nominees (Stanley, 2011). During the week of December 5, 2011, the Republican nominees collectively made 21 appearances on Fox News (Peters, 2011). Similar patterns emerged in 2016, with leading Fox hosts like Hannity interviewing many of the candidates on their shows (Scarry, 2015). All of these appearances are not an accident: candidates know that many of the conservative voters who will knock on doors, work for candidates, and participate in caucuses are likely Fox watchers. This raises a host of interesting questions about the role of such media in the primary process, especially the invisible primary process, which is an understudied topic. How does exposure on Fox correlate with poll numbers? Does appearing on Fox boost poll numbers, especially among key constituencies? Or is Fox simply responding to changes in the polls: inviting on viable candidates, ignoring nonviable ones (unless they draw eyeballs to the network)?

But the narrowcasting does not stop during the primary. These shows give a ready outlet for members of Congress from the left and the right to promulgate their message to an attentive audience. Given the relatively modest number of voters who watch these shows across the nation, the number in any given congressional district or state would be especially small. This suggests that such appearances are not a mechanism to increase the member’s name recognition in his/her own state or district. Rather, these appearances are likely more about raising a member’s national profile, perhaps in a bid for a leadership position or higher office. This raises particularly interesting possibilities for future work on theories of progressive ambition.

Finally, there is one additional point that brings together many of the issues raised in this article. There has recently been work illustrating the linkages between partisan media and incivility (e.g., Berry & Sobieraj, 2014). Such incivility depicted in the media seems to increase uncivil behavior in the audience (Gervais, 2014), but this raises the question of how such effects spread more broadly as well. For example, do the effects of this incivility in the audience spread to others? While viewers are unlikely to be uncivil with those in their immediate social networks, perhaps it leads them to (say) make uncivil posts about disliked politicians on Facebook, which might lead others to perceive said politicians as illegitimate (Mutz, 2007). One might also consider linkages with other news outlets, or with elites. Few would argue that partisan outlets are a main cause of uncivil discourse at the elite level, but because uncivil behavior draws viewers, these outlets may incidentally encourage that behavior from some politicians. In short, there are potentially complex interrelationships among the guests on these shows (often, but not always, political elites, journalists, bloggers, and the like), the audience, the broader public, and other news outlets. While political scientists and communications scholars have ably documented many consequences on the audiences themselves, there are other important questions remaining about some of these other relationships.

Hopefully, then, this article has helped to raise some points for future research. In particular, I hope to encourage scholars to think about how to move beyond what we have done well—identifying effects on the partisan media audience—to considering how these outlets shape public opinion and American politics more generally.


Arceneaux, K., & Johnson, M. (2013). Changing channels or changing minds? Partisan News in an Age of Choice. Chicago: University of Chicago Press.Find this resource:

Arceneaux, K., & Johnson, M. (2015). How mainstream media can polarize America, too. Paper presented at the conference Political Polarization: Media and Communication Influences. Princeton, NJ: Princeton University.Find this resource:

Arceneaux, K., Johnson, M., Lindstadt, R., & Vander Wielen, R. (2016). The influence of news media on political elites: Investigating strategic responsiveness in Congress. American Journal of Political Science, 60(1), 5–29.Find this resource:

Bakshy, E., Messing, S., & Adamic, L. (2015). Exposure to ideologically diverse news and opinion on Facebook. Science, 348(6239), 1130–1132.Find this resource:

Barbera, P. (2015). Birds of the same feather tweet together: Bayesian ideal point estimation using Twitter. Political Analysis, 23(1), 76–91.Find this resource:

Barker, D. (1999). Rushed decisions: Political talk radio and vote choice, 1994–1996. Journal of Politics, 61(2), 527–539.Find this resource:

Bartels, L. (1993). Messages received: The political impact of media exposure. American Political Science Review, 87(2), 267–285.Find this resource:

Baum, M. (2011). Preaching to the choir or converting the flock: Presidential communication strategies in the age of three medias. In R. Fox, & J. Ramos (Eds.), iPolitics: Citizens, elections, and governing in the New Media Age. New York: Cambridge University Press.Find this resource:

Baum, M., & Groeling, T. (2008). New media and the polarization of American political discourse. Political Communication, 25(4), 345–365.Find this resource:

Baum, M., & Groeling, T. (2010). War stories: The causes and consequences of public views of war. Princeton, NJ: Princeton University Press.Find this resource:

Berelson, B., Lazarsfeld, P., & McPhee, W. (1954). Voting: A study of opinion formation in a presidential campaign. Chicago: University of Chicago Press, Midway Reprints.Find this resource:

Berry, J., & Sobieraj, S. (2014). The outrage industry: Political opinion media and the new incivility. New York: Oxford University Press.Find this resource:

Bond, R., & Messing, S. (2015). Quantifying social media’s public space: Estimating ideology from publicly revealed preferences on Facebook. American Political Science Review, 109(1), 62–78.Find this resource:

Boyle, T. (2001). Intermedia agenda setting in the 1996 presidential election. Journalism and Mass Communication Quarterly, 78(1), 26–44.Find this resource:

Brock, D., Rabin-Havt, A., & Media Matters for America. (2012). The Fox effect: How Roger Ailes turned a network into a propaganda machine. New York: Anchor Books.Find this resource:

Carmines, E., & Stimson, J. (1980). The two faces of issue voting. American Political Science Review, 74(1), 78–91.Find this resource:

Chong, D., & Druckman, J. (2010). Dynamic public opinion: Framing effects over time. American Political Science Review, 104(4), 663–680.Find this resource:

Clifford, S., & Jerit, J. (2013). How words do the work of politics: Moral foundations theory and the debate over stem cell research. Journal of Politics, 75(3), 659–671.Find this resource:

Clinton, J., & Enamorado, T. (2014). The national news media’s effect on Congress: How Fox News affected elites in Congress. Journal of Politics, 76(4), 928–943.Find this resource:

D’Alessio, D., & Allen, M. (2000). Media bias in presidential elections: A meta-analysis. Journal of Communication, 50(4), 133–156.Find this resource:

Della Vigna, S., & Kaplan, E. (2007). The Fox News effect: Media bias and voting. The Quarterly Journal of Economics, 122(3), 1187–1234.Find this resource:

Domke, D., Watts, M., Shah, D., & Fan, D. (1999). The politics of conservative elites and the “liberal media” argument. Journal of Communication, 49(4), 35–58.Find this resource:

Downs, A. (1957). An economic theory of democracy. New York: Harper & Row.Find this resource:

Dreier, P., & Martin, C. (2010). How ACORN was framed: Political controversy and media agenda setting. Perspectives on Politics, 8(3), 761–792.Find this resource:

Druckman, J., & Leeper, T. (2012). Learning more from political communication experiments: Pretreatment and its effects. American Journal of Political Science, 56(4), 875–896.Find this resource:

Druckman, J., Levendusky, M., & McLain, A. (2015). No need to watch: How the effects of partisan media can spread via social networks. Manuscript, Northwestern University.Find this resource:

Dvir-Gvirsman, S., Tzfati, Y., & Menchen-Trevino, E. (2016). The extent and nature of ideological selective exposure online: Combining survey responses with actual web log data from the 2013 Israeli elections. New Media and Society, 18(5), 857–877.Find this resource:

Fiorina, M., Abrams, S., & Pope, J. (2008). Polarization in the American Public: Misconceptions and Misreadings. Journal of Politics, 70(2), 556–560.Find this resource:

Fowler, M. (2010). Bittergate: The untold story behind the story that rocked the Obama campaign. Huffington Post, March 18. Available online at this resource:

Gaines, B., & Kuklinski, J. (2011). Experimental estimates of heterogeneous treatment effects related to self-selection. American Journal of Political Science, 55(3), 724–736.Find this resource:

Garrett, R. K. (2009). Echo chambers online? Politically motivated selective exposure among internet news users. Journal of Computer Mediated Communication, 14(2), 265–285.Find this resource:

Gentzkow, M., & Shapiro, J. (2011). Ideological segregation online and offline. Quarterly Journal of Economics, 126(4), 1799–1839.Find this resource:

Gervais, B. (2014). Following the news? Reception of uncivil partisan media and the use of incivility in political expression. Political Communication, 31(4), 564–583.Find this resource:

Groseclose, T. (2011). Left turn: How liberal media bias distorts the American mind. New York: St. Martin’s Press.Find this resource:

Hindman, M. (2008). The myth of digital democracy. Princeton, NJ: Princeton University Press.Find this resource:

Hopkins, D. & Ladd, J. (2014). The consequences of broader media choice: Evidence from the expansion of Fox News. Quarterly Journal of Political Science, 9(1), 115–135.Find this resource:

Iyengar, S, & Kinder, D. (1987). News that matters. Chicago: University of Chicago Press.Find this resource:

Jamieson, K. H., & Cappella, J. (2008). Echo chamber: Rush Limbaugh and the conservative media establishment. New York: Oxford University Press.Find this resource:

Katz, E., & Lazarsfeld, P. (1955). Personal influence: The party played by people in the flow of mass communications. New York: Free Press.Find this resource:

Lelkes, Y., Sood, G., & Iyengar, S. (in press). The hostile audience: The effect of access to broadband internet on partisan affect. American Journal of Political Science.Find this resource:

Levendusky, M. (2013). How partisan media polarize America. Chicago: University of Chicago Press.Find this resource:

Levendusky, M. (2014). Review of the book Changing Minds or Changing Channels? by Kevin Arceneaux and Martin Johnson. Perspectives on Politics, 12(2), 474–475.Find this resource:

Martin, G., & Yurukoglu, A. (2014). Bias in cable news: Real effects and polarization. Manuscript, Emory University.Find this resource:

Mutz, D. (2007). Effects of “in-your-face” television discourse on perceptions of a legitimate opposition. American Political Science Review, 101(4), 621–635.Find this resource:

Perlstein, R. (2008). Nixonland: The rise of a president and the fracturing of America. New York: Scribner.Find this resource:

Peters, J. (2011, December 5). Campaigns take ad war to TV after months of holding fire. New York Times.Find this resource:

Petrocik, J. (1996). Issue ownership in presidential elections, with a 1980 case study. American Journal of Political Science, 40(3), 825–850.Find this resource:

Prior, M. (2007). Post-broadcast democracy: How media choice increases inequality in political involvement and polarizes elections. New York: Cambridge University Press.Find this resource:

Prior, M. (2013). Media and political polarization. Annual Review of Political Science, 16(1), 101–127.Find this resource:

Roberts, M., & McCombs, M. (1994). Agenda setting and political advertising: Origins of the news agenda. Political Communication, 11(3), 249–262.Find this resource:

Scarry, E. (2015, April 7). Fox’s Hannity revitalized by GOP 2016 primary. Washington Examiner.Find this resource:

Shaw, D. (2006). The race to 270. Chicago: University of Chicago Press.Find this resource:

Skocpol, T., & Williamson, V. (2012). The Tea Party and the remaking of Republican conservatism. New York: Oxford University Press.Find this resource:

Stanley, A. (2011, December 10). The Republican primary campaign in Iowa is right at home on Fox News. New York Times.Find this resource:

Stroud, N. (2011). Niche news: The politics of news choice. New York: Oxford University Press.Find this resource:

Suhay, E., Blackwell, A., Roche, C., & Bruggeman, L. (2015). Forging bonds and burning bridges: Polarization and incivility in blog discussions about Occupy Wall Street. American Politics Research, 43(4), 643–679.Find this resource:


(1.) This article focuses on partisan media in the United States. For a discussion of partisan media’s political roles in other systems, see, for example, Dvir-Gvirsman, Tzfati, and Menchen-Trevino (2016).

(2.) Throughout this essay, when I discuss polarization, I primarily refer to ideological polarization: that is, a tendency for opinions to move from the ideological center toward the poles (for more on various definitions of polarization, Fiorina, Abrams, & Pope, 2008). That said, partisan media exposure can also influence affective polarization, or the tendency for ordinary Democrats to distrust and dislike one another (Lelkes, Sood, and Iyengar, in press).

(3.) Another fascinating possibility—not considered in this essay—is that even mainstream media outlets, at least for some people, can also increase polarization (Arceneaux & Johnson, 2015).