Communications of the ACM
Contributed articles: Virtual extension
The Internet Electorate
Credit: Election Resources on the Internet
The Internet was a prominent aspect of the 2008 U.S. presidential election, regularly noted for its role in the Obama campaign's successful fundraising and supporter-mobilization efforts9 and widespread use by interested voters.16 Here, we report on a national telephone survey we conducted in the weeks following that election to assess how Americans' experience of elections is changing in response to the increasing availability and use of the digital communication network. The Internet has long been heralded as an efficient means of acquiring political information,2 but the increasing presence of user-created content means the network is also becoming an important mode of political expression.3 Here, we examine these complementary roles, exploring how Americans used it to learn about the campaign, share political information, and voice their own opinions. We also analyze which individuals would be most likely to engage in online information acquisition and expression, concluding by examining the influence these practices have on voters. The analyses are based on the national random-digit-dial telephone survey of 600 adult Americans in the two weeks immediately following the 2008 election (November 620), with a response rate of 26.2%.
In the lead-up to the election, nearly two-thirds (64%) of Americans got campaign news online (see Figure 1), a marked increase from 2004, when only about one-quarter (27%) of Americans said they got campaign news online.13 Equally notable is the fact that in 2008 almost two-fifths (38%) of respondents reported seeking online campaign news almost every day. Use of online sources in 2008 was only slightly lower than Americans' information seeking on radio and even newspapers, though it continued to be a far less pervasive source than television news (96% use and 83% use almost daily).
What online sources do citizens prefer? Figure 2 provides evidence of five broad findings about online sources of political information during the 2008 election. First, about half (49%) of all Americans used Web sites associated with the major news organizations for campaign information. This appears to be a technological update of the traditional approach to seeking information about elections (and politics in general) from credible, established media institutions (especially television networks and major newspapers). Unlike many other types of Web-based information sources, these major media outlets are not defined solely by their online presence. The Internet is just one of the modalities by which these organizations disseminate news and information, though an increasingly important one, and the routin-ized use and reliability associated with their offline outlets probably contributes to their success online.
It is noteworthy that in 2008 most citizens accessed the online versions of mainstream news outlets rather than the more-partisan alternatives. A number of scholars have expressed concern that as Americans turn to the Internet for political information, they gravitate to partisan sites in order to reinforce their existing views and filter out contact with political ideas with which they might disagree, abandoning broad, less-biased news outlets.1,20 Yet compared to the 49% of survey respondents who used mainstream news sites, only 14%about one in sevenused explicitly partisan information sites (liberal or conservative) during the 2008 election cycle. This proportion is slightly higher but statistically indistinguishable from the 11% of Americans who sought information from partisan sites during the 2004 presidential election. Indeed, given that more than twice as many individuals sought campaign information online in the 2008 election cycle, it is arguable there was a shift away from partisan online sources. The proportion of Americans using partisan sites appears to have leveled off despite continued growth of the mainstream news audience.
Second, it is striking that the proportion of individuals (53%) reporting that email from friends and family was a source of political information during the 2008 presidential campaign is even higher than the proportion using major news sites online. As might be expected, a large proportion of respondents (38%) reported turning to both friends and family, as well as national news media, to learn about the campaigns. And some, perhaps most, of the information from these online-friends-and-family sources can be traced back to mainstream sources; for example, we regularly forward news stories we consider important. However, despite their dependance on content from mainstream outlets, such informal social networks, based on friendship and familial bonds, constitute a relationship between the person sending campaign information and the person receiving it that is very different from its counterpart between mainstream broadcast senders and news readers. A message traveling over a social network has been customized in the sense it has been selected by a significant other and targeted to the recipient. It is thus not the product of an information-seeking activity and, though it might have originated on a broadcast medium, has heightened salience to the recipient because it comes via a personal connection.
This pattern of information sharing is consistent with a classic finding of opinion researchthat individuals receive a significant share of their political information through a "two-step flow" of communicationwhereby an opinion leader selectively identifies and communicates relevant information to the recipient.15 But several aspects of this digitized communication process are notable, and perhaps new. Senders need not absorb, synthesize, and restate information in order to transmit rich and detailed content, because forwarding the original content or electronic pointer is so easy. Also, senders can transmit information to multiple recipients with minimal effort and need not establish a synchronous face-to-face connection to do so. Moreover, the recipient has near-instant access to a vast, diverse array of related information resources. The recipient, if interested, not only consumes the recommended information but can actively pursue the topic more extensively through online tools like search engines. When the recipient takes a more active role in processing and exploring an online political message, the impact of the message can be enhanced significantly.12
The third broad finding of our survey regarding online campaign information is that a substantial proportion of respondents (32%) received online information from political groups and from the candidates themselves. This targeted marketing is indicative of how formal political organizations are adopting and routinizing the public Internet as a means of reaching a vast population of potential voters. For political actors, online media offer clear strategic advantages: lower cost per contact; ability to customize specific messages to target groups; ability to integrate interactive technologies; and opportunity to encourage recipients to recirculate campaign information to large numbers of people via their friends-and-family networks. These features provide enormous potential benefits to political organizations and to individual politicians; it is reasonable to predict that the use of such online communications will only increase.
Fourth, among the "alternative" online news outlets, we found both partisan and nonpartisan sites had been consulted in 2008 by sizable audiences. As noted, about one in seven respondents (14%) used a partisan news site from either side of the political spectrum, with a greater proportion seeking information from conservative sites (11%, such as NewsMax.com) than from liberal sites (7%, such as DailyKos.com). Nonpartisan sites were even more widely used than their partisan counterparts. Voter information sites, including the rumor-checking FactCheck.org and Project Vote Smart, were used by about one-fifth (19%) of respondents. And independent (nonmajor) political news organizations (such as RealClearPolitics.com) were used by about one in six (15%) respondents.
Sites that aggregate news and information from across the Web (such as Google News and Digg) are another avenue to political content. Though not entirely comparable to the alternative news sites (because they do not generate their own content), they create opportunities for people to encounter crosscutting ideas, because unlike partisan sites, they explicitly seek to provide content from a range of sources. News aggregators are used more widely than any of the alternative outlets; 29% of respondents reported using them. Use of these news aggregators as alternatives to the mainstream news media suggests that when citizens seek other sources of political information, most value diversity and less bias.
Finally, the most "populist" modalities of the online world have also penetrated political campaigns. In particular, more than one in four Americans viewed campaign information at video-sharing sites (such as YouTube) during the 2008 presidential campaign. Some independently produced videos (most notably Obama Girl's "Crush on Obama") achieved national recognition, viewed more than 14 million times before the election and that earned invitations for its star to appear on several widely viewed television shows, including "The O'Reilly Factor" and "Saturday Night Live." Other videos captured revealing comments by campaigners and were often circulated by opponents to undermine a candidate's credibility and support. As one might expect, younger people were the heaviest users of the video-sharing sites, but such sites were not solely their domain. In fact, for those reporting they accessed campaign news at online video sites on a daily basis, the average age was 40.
Who acquires political information online? Having characterized the extent to which Americans turn to online sources, what distinguishes the information consumers from the information nonconsumers? Individual attributes associated with other types of Internet use are also important here: lower age and higher income and education level are each positively associated with acquisition of political information online (see Table A1 in the online appendix for supporting data). These consumers of online political information are also more extensive users of offline political news sources, as well. Moroever, we found several explicitly political factors play a role in shaping online political information acquisition among those with strong partisan ideology, as well as among those engaged in more extensive political activity. Finally, those who distrust the mainstream media are more likely to turn to Internet-based sources.
Political Expression Online
The Internet plays a key role for millions of Americans seeking political information online. But the 2008 election exemplified the growing importance of another use of the Internet by the electoratepolitical expression; that is, the Internet is a resource not only for those seeking and consuming political information but also for those producing and disseminating it. It serves as a conduit through which individuals share information and opinions and is thus a form of political expression and, potentially, persuasion; Figure 3 outlines at least five forms of online political expression.
How do citizens express themselves online? Email is the dominant Internet technology for expressing political ideas, as it is for acquiring political information. About one-third (34%) of respondents used it during the 2008 campaign, and more than one-fifth (21%) sent political information multiple times. While most information shared this way was produced by others, it was presumably extracted from online sources, thereby increasing the centrality of the Internet to the circulation of political information. Moreover, what was especially significant about these activities was that (in terms of the two-step communication flow) individuals used the Internet to enable themselves to be opinion leaders by disseminating information. Acting as an online disseminator of information is a far more active political role for individuals than being a recipient of political information or an information seeker. The role could potentially engage and mobilize individuals more extensively in the world of politics.
Email also facilitates communication with political leaders. About one-fifth (19%) of respondents sent email to an elected official or a political candidate during the 2008 U.S. presidential campaign. Contacting a political actor is consistently interpreted as one of the most important forms of citizen engagement in a representative democracy.4 There is also evidence that the Internet, as a mode of contact with elected officials, encourages political communication by individuals who previously did not engage in such contact.14
A third form of Internet-based political expression is signing online petitions, an activity reported by one-seventh (14%) of respondents, with most signing more than one petition. Online petitions are an integral part of many protest sites6 and constitute another means of communicating a political opinion to public officials and candidates. Because digitally signing these documents takes little effort, online petitions are sometimes viewed as having less impact on decision makers than hand-signed offline petitions. However, empirical evidence suggests that online petitions are influential in some cases10 and the act of signing can contribute to the formation of issue-based communities through recognition of shared political interest.7 At the very least, these petitions have symbolic communicative value to signers, enabling them to signal the salience of an issue to a political actor.
Fourth, an explicitly public form of political expression is posting comments online in response to political content authored by someone else. Nearly one in eight (13%) respondents reported posting a public comment on the Web at least once during the 2008 campaign, including about one in nine (12%) posting a comment on a news site and one in 12 (8%) posting a comment on a political blog or discussion board. This is a particularly interesting form of dialogic communication due to the level of commitment to political expression it represents. Moreover, our research suggests that some individuals view their participation in online political discussion as an opportunity to engage with those with whom they disagree, often valuing the diversity of views they encounter in the discussion.19 Other empirical analyses demonstrate that political discussions via blog posts regularly cross ideological boundaries and substantive exchanges are a central aspect of such crosscutting discussions.11 For advocates of deliberative democracy, it is thus perhaps encouraging that so many Americans engage in this mode of political expression.
Finally, only a tiny fraction (3%) of respondents shared their political views by creating a political video they posted online. Yet they might represent the cutting edge of online political expression, with early adopters experimenting with technologies that more Americans will likely embrace over time. YouTube, the first user-created video site to achieve a national (and now global) audience, was established in 2005. After only these few years, millions of Americans use it to express their political opinions. From serious critiques to humorous homage, hundreds of political videos were posted in the lead-up to the 2008 election, an action that is, as noted earlier, a potentially powerful form of populist political expression. Political organizations and major news organizations were quick to recognize this political potential, as, for example, when CNN in 2007 invited viewers to submit via YouTube questions for a series of presidential debates.8 In the coming years, it will be fascinating to track the evolution of political expression through user-created online media like video and social-networking services.
Who expresses political opinions online? How can we characterize the people most engaged in online political expression? Use of online news is the strongest predictor of online political expression, but political information acquisition and sharing via the Internet are not perfectly correlated. Other factors also have a distinct influence on online political expression. After controlling for their level of online news consumption, people who are older, have less money, and are more politically active are more likely to express themselves online (see Table A2 in the online appendix). Those with strongly held ideological beliefs are also marginally more likely to engage in online political expression than those less committed to an ideology.
Electoral consequences
Our analysis indicates the Internet was becoming an important part of the contemporary U.S. political communication environment. But do the online political communications described here have tangible electoral consequences? Our data reveals that use of the Internet to acquire political information and express political opinions is significantly associated with the act of voting. Voters sought political information from online news sources more often than nonvoters (t=2.21, p <.05) and were much more likely to express their opinions online than nonvoters (t=4.0, p <.001).
These correlations are partially explained by the fact that online political activity and voting share a common set of underlying causal factors. To demonstrate this, we used logistic regression, a statistical technique that allowed us to account for a variety of factors, including age, interest in the election, political activity, income, race, offline media use, and online political activity when predicting whether or not an individual would vote (see Table A3 of the online appendix for model coefficients). As expected, the relationship between online expression and voting is unchanged under conditions of multivariate control. Individuals who engage in online political expression are more likely to vote than those who do not engage (see Figure 4a), but, surprisingly, the direction of the association between acquiring campaign information online and voting is negative when we control for political expression. Among those with comparable levels of Internet-based information sharing, greater use of the Internet for acquiring political information actually reduces the likelihood that an individual would vote in the election (see Figure 4b).
What drives this pattern? One explanation springs from Anthony Downs' classic model5 of voting behavior, postulating that individuals seek information to guide a well-reasoned vote choice, as long as the perceived benefits from acquiring additional information outweigh the costs of gathering and managing it. Perhaps individuals continue to acquire political information online because they remain uncertain about whether and how to cast their ballots. However, even at the point the vote must be cast, they are still uncertain and thus do not vote.
A second explanation of the negative relationship between voting and greater explosure to online political information is based on a more dysfunctional version of Downs' voter model. Downs assumed the individual is strategic and will abandon the information search when costs outweigh benefits, but individuals might fail to accurately assess the costs of acquiring additional online information. Some people are unable to self-regulate their use of the Internet, transforming online-information seeking from a purposeful activity into something more problematic.17 If online political-information seeking becomes habitual, even addictive, extensive activity might no longer function to increase an individual's knowledge or motivation regarding voting.
A third possible explanation of the negative relationship is suggested by our findings that online political-information exposure was not heavily focused on content that reinforced the user's partisan viewpoint. Although an individual's exposure to a diverse marketplace of ideas has many desirable consequences in a democracy, it can also produce political ambivalence,18 even alienation. Thus, although individuals may seek information with the intention of casting a more-informed vote, the conflicting or nonreinforcing nature of the additional information could actually increase uncertainty or negative attitudes toward candidates or the electoral process in general, ultimately reducing the likelihood of their voting.
Use of online news is the strongest predictor of online political expression, but political information acquisition and sharing via the Internet are not perfectly correlated.
A fourth plausible explanation for the finding relates to the individual's motivations for seeking political information online. We've reported here that the negative relationship between online campaign-information exposure and voting holds only when we control for such factors as political expression and campaign interest. Perhaps increased online political-information seeking not accompanied by these forms of political engagement is indicative of non-vote-related motivation. It might be that some individuals are more interested in being entertained by online political information than they are in using it to shape a more informed vote choice they then exercise through the ballot box. Some political junkies may thus view politics as blood sport, seeking news about politics the way others attend to the sports pages. Yet others might view online political information as a compelling form of infotainment, seeking it for the same reasons they read gossip about Hollywood celebrities. Dedicated political sites (such as Politico, which offers exhaustive political coverage but sometimes emphasizes political game-playing and Beltway gossip over issues and candidate positions) could serve such online-political-information consumption. People especially interested in being entertained by politics are arguably both more likely to consume substantial amounts of political information and less likely to actually cast a ballot. While our data does not allow us to test all these alternative explanations, the decline of voting among a significant group that encounters more political information online merits further exploration.
Conclusion
Using the Internet as a conduit for political information has grown steadily over the past decade, with our data on the use of the Internet during the U.S. 2008 election campaign identifying two important modes:
As a news source. Fully two-thirds of Americans used the Internet as a source of political information during the 2008 campaign, thanks in part to the growing presence of online-information-exposure activities, comparable to the use levels of such traditional sources as newspapers and radio. We highlighted several particularly interesting aspects of this online information seeking; for example, networks of friends and family are a key source of online political information, creating a significant and novel form of the classic "two-step information flow." Also, online information from the more-established media is used much more extensively than its counterpart from nontraditional sources. Among the latter, nonpartisan sources are used more than partisan ones; and
As a mode of political expression. In the 2008 election, millions of Americans used email and the Web to share political information and opinions and even to create their own online political media. Thus, the Internet became an important avenue for political expression, with the potential to facilitate a much more democratic dissemination of information and empower individuals as information sources. People engaging in these practices might have found other ways to be politically involved without these technologies, but our analysis indicates that by 2008 the Internet had become a significant medium, reinforcing existing patterns while opening new lines of political communication.
Do the online political communications described here have tangible electoral consequences?
We also examined the linkages, at the individual level of analysis, between the two modes of Internet use and the probability of voting (in the 2008 presidential election). While more activity in each mode is independently correlated with a higher probability of voting, the introduction of controls revealed a more subtle and even surprising set of linkages. We offered some possible explanations for the lower probability of voting among those who were more active in online political-information seeking, given controls on online political expression and conventional predictors of voting behavior. The Internet electorate is still in its infancy, but our research suggests an increasing proportion of Americans will rely on these capabilities. The patterns of both political-information seeking and exposure and also of political expression within the Internet electorate will thus continue to evolve and contribute in significant ways to the democratic process.
Acknowledgments
We would like to thank Debbie Dunkle for her assistance preparing the survey and the data and the anonymous reviewers for their valuable feedback. This research is part of the People, Organizations, and INformation Technology (POINT) project of the Center for Research on Information Technology and Organizations at the University of California, Irvine. It is based on work funded by the U.S. National Science foundation under Grant No. SES-0121232. Any opinions, findings and conclusions reflected in the material are those of the authors and do not necessarily reflect those of the National Science Foundation.
References
1. Bennett, W.L. and Iyengar, S. A new era of minimal effects: The changing foundations of political communication. Journal of Communication 58, 4 (Dec. 2008), 707731.
2. Bimber, B. Information and American Democracy: Technology in the Evolution of Political Power. Cambridge University Press, New York, 2003.
3. Chadwick, A. and Howard, P.N. Routledge Handbook of Internet Politics. Routledge, New York, 2009.
4. Dalton, R.J. Citizen Politics: Public Opinion and Political Parties in Advanced Industrial Democracies. CQ Press, Washington, D.C., 2008.
5. Downs, A. An Economic Theory of Democracy. Harper & Row, New York, 1957.
6. Earl, J. Pursuing social change online: The use of four protest tactics on the Internet. Social Science Computer Review 24, 3 (Fall 2006), 362377.
7. Garrett, R.K. Protest in an information society: A review of literature on social movements and new ICTs. Information, Communication and Society 9, 2 (Apr. 2006), 202224.
8. Gough, P.J. CNN's YouTube debate draws impressive ratings. Thomson Reuters, 2007; http://www.reuters.com/article/idUSN2425835220070725
9. Green, J. The amazing money machine. The Atlantic Monthly 301, 5 (June 2008), 5263.
10. Gurak, L.J. Persuasion and Privacy in Cyberspace: The Online Protests Over Lotus Marketplace and the Clipper Chip. Yale University Press, New Haven, CT, 1997.
11. Hargittai, E., Gallo, J., and Kane, M. Cross-ideological discussions among conservative and liberal bloggers. Public Choice 134, 12 (Jan. 2008), 6786.
12. Holbert, R.L., Garrett, R.K., and Gleason, L.S. A new era of minimal effects: A response to Bennett and Iyengar. Journal of Communication 60, 1 (Mar. 2010), 1534.
13. Horrigan, J., Garrett, K., and Resnick, P. The Internet and Democratic Debate. Pew Internet & American Life Project, Washington, D.C., 2004.
14. Jensen, M., Danziger, J., and Venkatesh, A. Civil society and cyber society: The role of the Internet in community associations and democratic politics. The Information Society 23, 1 (Dec. 2007), 3950.
15. Katz, E. and Lazarsfeld, P.F. Personal Influence. Free Press, New York, 1955.
16. Kohut, A., Doherty, C., Dimock, M., and Keeter, S. Internet's Broader Role in Campaign 2008: Social Networking and Online Videos Take Off. The Pew Research Center for the People & the Press, Washington, D.C., 2008.
17. LaRose, R., Lin, C.A., and Eastin, M.S. Unregulated Internet usage: Addiction, habit, or deficient self-regulation? Media Psychology 5, 3 (Aug. 2003), 225253.
18. Mutz, D.C. Hearing the Other Side: Deliberative Versus Participatory Democracy. Cambridge University Press, New York, 2006.
19. Stromer-Galley, J. Diversity of political conversation on the Internet: Users' perspectives. Journal of Computer-Mediated Communication 8, 3 (Apr. 2003).
20. Sunstein, C.R. Republic.com. Princeton University Press, Princeton, NJ, 2001.
Figures
Figure 1. Use of political information sources.
Figure 2. Seeking political information online.
Figure 3. Sharing political information online.
Figure 4. Probability of voting by online political activity.
Figure. Website screen grabs of election results from the morning of November 5th 2008.
©2011 ACM 0001-0782/11/0300 $10.00
Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and full citation on the first page. Copyright for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, to republish, to post on servers, or to redistribute to lists, requires prior specific permission and/or fee. Request permission to publish from [email protected] or fax (212) 869-0481.
The Digital Library is published by the Association for Computing Machinery. Copyright © 2011 ACM, Inc.
No entries found