Psychology Help

Selectively Social Politics: The DifferingRoles of Media Use on Political Discussion J. D. Ponder and Paul Haridakis Department of Studies Studies Kent State University In the modern media environment, people are afforded a variety of options for political information. In addition, people now use multiple media sources (e.g., television, radio, blogs) to obtain information about all aspects of politics (Eveland, 2004; Pew Research Center for the People and the Press, 2009).

The purpose of this study was to examine how use of particular media sources influenced the frequency of political discussion with people from the same political party (political in-group members) and people from a different polit- ical party (political out-group members). Guided by a uses and gratifications perspective, which emphasizes the role of the user in media effects, we exam- ined how specific user background characteristics (e.g., age, sex, political opi- nion leadership, political social identity, political content affinity), motives for using traditional and social media for political information, and use of different media sources work together to influence discussion with political in-group and out-group members. Our results allowed us to identify several distinct differences between people who talk to political in-group and out- group members.

J.D. Ponder (Ph.D., Kent State University, 2012) is an Assistant Professor in the Department of Studies Studies at Kent State University. His research interests include media uses and effects, identity, learning, and political communication.

Paul Haridakis (Ph.D., Kent State University, 2000) is a Professor in the Department of Studies Studies at Kent State University. His research interests include media uses and effects, new communication technologies, freedom of expression and media history.

Correspondence should be addressed to J. D. Ponder, Department of Studies Studies, Kent State University, 135 Taylor Hall, 300 Midway Drive, P.O. Box 5190, Kent, OH 44240. E-mail:

[email protected] Mass Communication and Society, 18:281–302, 2015 Copyright#Mass Communication & Society Division of the Association for Education in Journalism and Mass Communication ISSN: 1520-5436 print =1532-7825 online DOI: 10.1080/15205436.2014.940977 281 INTRODUCTION In the contemporary media environment, people have a variety of different media sources from which to acquire political information. People interested in learning about politics or gaining political information can watch tele- vision; listen to the radio; read newspapers, magazines, or books; in addition to using online sources (e.g., blogs, social networking sites, video sharing sites). In most cases, people now use multiple media sources (e.g., television, radio, blogs) to obtain information about politics, political actors, and polit- ical issues (Eveland, 2004; Pew Research Center for the People and the Press, 2009).

Perhaps even more important, people also talk with others about politics.

They seek out people who can help them make sense of mediated political information (Eveland & Shah, 2003; Mutz, 2002). Although the attitudinal and behavioral effects of political discussion have been well documented, especially how it can positively affect political knowledge or voting (e.g., Delli-Carpini & Keeter, 1996; Kwak, Williams, Wang, & Lee, 2005), the latter in particular, is related to social capital. There have been claims that some media use (e.g., television) detracts from civic debate and participation (Putnam, 2000). On the other hand, newer social media have played an instrumental role in the creation of political movements (i.e., Occupy Wall Street, Tea Party movement), and even revolutionary movements (i.e., Arab Spring). Although such outcomes of media use and discussion have received much attention, predictors of discussion are relatively unstudied. In addition, a few scholars have found that people use media differently based upon the political affiliation of their discussion partners (e.g., Eveland & Shah, 2003; Mutz, 2002). However, without considering how media users’ individual differences work in conjunction with their media selection to influence postviewing discussion, we can only speculate about how people use media content from different media sources to talk about politics with their political allies versus those who hold opposing political views, or how viewers of certain types of programs may be more likely to debate poli- tics with political friends and =or foes than are viewers of other programs.

The purpose of this study is to examine how media use influences the frequency of political discussion with people from the same political party and people from a different party. This research is guided by a uses and gratifications perspective. Uses and gratifications researchers posit media effects, such as post-media-use discussion, as a product of user personal characteristics, motives for using media, and media use. More important, this is a process that builds on itself in that personal characteristics influence motives for using media, which influence media use, which in turn influence the effects of that use. 282 PONDER AND HARIDAKIS THE ROLE OF DISCUSSION IN POLITICS Denton and Woodward ( 1998) explained that political communication research focuses on the production, dissemination, processing, and effects of information, both through the media and interpersonally. In this light, political discussion is a key component in both disseminating information and helping people make sense of the information presented to them via mass media (Delli-Carpini & Keeter, 1996; Eveland & Hively, 2009). People not only turn to others to help them make sense of the information but also differentiate between people in their own party (political in-group members) and people from a different party (political out-group members) as conver- sation partners (Eveland & Shah, 2003; Eveland & Hively, 2009). Further, they seek out politically similar others more frequently than politically dissimilar others when trying to process political information presented to them via mass media (Borgida, Federico, & Sullivan, 2009; Eveland & Hively, 2009).

USES AND GRATIFICATIONS Uses and gratifications is an audience-centered, media effects perspective that focuses on what people do with the media as opposed to what the media do to people. Uses and gratifications scholars suggest that individual social and psychological differences; motives for media use; use of media; and levels of activity at various points before, during, and after exposure to media all work together to influence the effects of media exposure (Rubin, 2009).

Uses and gratifications has been applied to examine the use of media for political information (Eveland, 2004; Hanson, Haridakis, Wagstaff, Sharma, & Ponder, 2011), motives for using media for political information (Kaye & Johnson, 2004; McLeod & Becker, 1974), and the effects of polit- ical media use, such as political discussion (Eveland, 2004). In the uses and gratifications framework, discussion has been identified as a common post- activity manifestation of audience activity (Levy & Windahl, 1984; Rubin, 2009 ). In the context of politics, political discussion allows people to reflect on and integrate that information into their lives (Eveland, 2004; Rubin & Perse, 1987).

As just referenced, uses and gratifications is based on the notion that media users’ social and psychological circumstances influence their media use. Certain personal characteristics have been particularly salient in the study of politics and political discussion. In the following sections, we discuss the role that these particular characteristics play in influencing use of media for political information and =or political discussion. IN-GROUP AND OUT-GROUP 283 Demographics Multiple scholars have examined the role that age and gender play in explaining differences in using media for political information, media use, and political discussion. Historically, age and gender have been associated with different motives people have for using media, the media sources people use, and the amount of political discussion in which people engage (Mendez & Osborn,2010; Mutz, 2002). Specifically, some research has sug- gested that younger voters tend to be less engaged and less willing to discuss politics with others than are older voters (Hardy & Scheufele, 2009; Shah, Kwak, & Holbert, 2001). In addition, there is a substantial body of literature outlining the differences between men and women in relation to political participation (Mendez & Obsorn, 2010). For example, Lazarsfeld, Berelson, and Gaudet (1948) found that women in 1940 tended to vote largely in line with their husbands. Historically, although women have been activists and leaders of social movements like prohibition and suffrage, they have largely been reticent to engage in conversations about the state of the political environment (e.g., Mendez & Osborn, 2010). Therefore, it is important to include these characteristics in any model that predicts political discussion.

Political Opinion Leadership In terms of politics, scholars have suggested that opinion leaders are less likely to attend to messages from politically different people and institutions and less likely to speak to out-group members than members of the public (e.g., Guerin et al., 2004). Conversely, scholars have also found that opinion leaders are more likely to talk to members of their own political party than the general public (e.g., Eveland & Shah, 2003; Shah & Scheufele, 2006). In addition, research on political opinion leadership has consistently suggested that opinion leaders use more media content in their specific interest area, are more active during this use, and disseminate it to others in their political in-group (Shah & Scheufele, 2006). Therefore, it is important to understand how a person’s degree of opinion leadership could influence motives for using media for political information, which media sources he or she uses, and how this use can influence political discussion with political in-group and out-group members.

Political Social Identity Previous uses and gratifications researchers have examined how a person’s membership in a group may influence media use. However, the studies have not considered the psychological or emotional connection that members feel with that particular group. This connection, called a person’s social identity, 284 PONDER AND HARIDAKIS has been linked to motives for media exposure (Huddy,2001), use of specific media sources (Stroud, 2008), and even the choice of discussion partners (Borgida et al., 2009; Stroud, 2008). For instance, Stroud ( 2008) found that people who identify with a party selectively use cable television, newspapers, and the Internet as sources for political information. Further, scholars have suggested that a person’s identification with a particular social group is highly influential in determining which discussion partners people choose after media exposure (Borgida et al., 2009).

Political Content Affinity Affinity with particular content refers to the level of importance placed upon the content of media messages by the user (Rubin, 2009). In the uses and gratifications perspective, people’s affinity with particular content has been conceptualized as an indicator of audience activity, suggesting that audience members are more active in their media use and what they subsequently do with content consumed when they enjoy a specific type of programming (Holbrook & Hill, 2005). Uses and gratification scholars have linked media content affinity to more purposive motives for using media and engaging in postexposure discussion (Holbrook & Hill, 2005; Levy & Windahl, 1984).

Motives for Using Media for Political Information Uses and gratifications researchers emphasize that people’s social and psychological characteristics are sources for the needs and desires people seek to satisfy through using media and other communication channels. These needs and desires are manifested in motives that people have for using media. Researchers have explored a variety of possible motives that people have for seeking political fare (McLeod & Becker, 1974) and for using specific media such as the Internet for political fare (Kaye & Johnson, 2004). These include seeking information, entertainment, relaxation, social utility, and vote guidance, among others. In addition, when it comes to using the Internet or specific Internet functions such as YouTube, some users are motivated to use these newer media sources, in part, because of the convenience of online use (Hanson et al., 2011; Papacharissi & Rubin, 2000). Other scholars have suggested that users may be motivated to use media fare to increase their positive feelings attached to group memberships (Harwood, 1999; Huddy, 2001 ). According to uses and gratifications, the motives that people have for using particular media content also influence their decisions to select and use particular media. IN-GROUP AND OUT-GROUP 285 Media Use In today’s media landscape, people are afforded a variety of media options for political information. In such a diverse media environment, people are given the opportunity to choose the medium (or variety of media) that best satisfies their needs. Political uses and gratifications researchers have begun to focus on how people use specific media sources to satisfy their needs related to politics. Understanding what affects people’s media use for polit- ical information has been a common focus of political uses and gratifications research. For instance, Bennett, Flickinger, and Rhine (2000) found that Americans who obtained political information from magazines, the news- paper, and television were more likely to engage in political discussion than those who used more traditional media sources. Conversely, Moy and Gastil ( 2006 ) suggested that the use of newspaper, television network =cable news, political talk radio, the Internet, and late-night comedy programs tended to promote political discussion among their users. In light of these varying conclusions, it was our aim to seek to examine how people selectively seek out and use a variety of different media to acquire political information, and ultimately how that media use can influence discussions about politics with others. In addition, although previous scholars have sought to uncover the use of various media in the political conversation process, these investiga- tions have been limited in the breadth of media sources measured. Typically, investigations in this vein use a limited number of media analyzed. Thus in the fragmented but rich media environment currently available, scholars need to examine the role of newer media in concert with more traditional forms in facilitating discussion among citizens. Indeed, scholars have criti- cized media studies for this very reason, calling for further investigations into how all media work together to influence political behaviors (e.g., Pfau, Houston, & Semmler, 2005). In this investigation, we expanded this approach to multiple media types (18 in all) to examine the media system more thoroughly.

Elaboration Elaboration plays an important role in facilitating political discussion.

According to Kwak et al. ( 2005), elaboration is an important contributor to political engagement, including discussion. Existing research has demon- strated a strong relationship between elaboration on political information and the ability to recall and use that information in later discussions (Eveland, 2004; Scheufele, 2002). In addition, Eveland and Dunwoody ( 2000 ) explained that cognitive elaboration on mediated political infor- mation occurs when people connect the media information to information 286 PONDER AND HARIDAKIS in existing schemas. Eveland (2004) explained that this process occurs when people anticipate engaging in discussions after their use of media for political information. Therefore, it is evident that cognitive elaboration should play a key role in predicting postexposure political discussion with others.

Political Discussion Scholars consistently have found that people turn to the media to provide them with topics and information to use in later discussions and how that dis- cussion leads to lasting effects on audience members (e.g., Bennett et al., 2000 ; Kim, Wyatt, & Katz, 1999). For instance, Kim et al. ( 1999) suggested that political discussion serves as a way for citizens to bridge their personal experiences to the larger political world. Political discussion was also one of the major foci in early media effects research that positioned the user as an active participant (e.g., Lazarsfeld et al., 1948). More recently, researchers have found that people discuss important news and political issues after exposure to media content (Eveland, 2004; Kim et al., 1999). These scholars have argued that discussion (or even the anticipation of it) is an important part in how people interact with and make sense of political information available to them through the media. In terms of in-group and out-group behaviors, researchers have found that once people identify with a political group, they seek out politically similar others to make sense of mediated political information (Borgida et al., 2009; Hogg & Reid, 2006). At the same time, people do talk to politi- cally different others about political information, albeit at a less frequent rate. In addition, out-group discussion has not been studied as in-depth as discussion with in-group members (Eveland & Hively, 2009; Mutz, 2002).

RESEARCH QUESTIONS AND HYPOTHESES In sum, according to the basic suppositions of uses and gratifications, media effects are a result of much more than sheer exposure. Effects of media use are the result of a variety of different factors working together. In the context of the study, uses and gratifications suggests that (a) individual personal characteristics of media users (i.e., political opinion leadership, political social identity, political content affinity) (b) influence motives for using media for political information, which in turn (c) influence selection of media for political information, (d) elaboration on political media content, and ulti- mately the outcome of interest, (e) political discussion. The main research questions of this particular study asked about the relationship between the personal characteristics, motives for using media for political information, IN-GROUP AND OUT-GROUP 287 media use, and elaboration, in predicting political discussion with two distinct types of political discussants—in-group members and out-group members. We examined these relationships by posing two research questions. RQ1: How do user personal characteristics, motives, types, and amount ofmedia use for political information and elaboration on political content obtained predict discussion with political in-group members?

RQ2: How do user personal characteristics, motives, types, and amount of media use for political information and elaboration on political content obtained predict discussion with political out-group members? METHOD Sample Participants composed a nationwide sample solicited through an online survey provider—Qualtrics. Qualtrics’ partners recruited a pool of 1,516 participants (approximating the demographic makeup of U.S. voters) through a variety of online survey sample websites that enable users to sign up to take surveys. Participants were offered an incentive of $2.25 for filling in the survey, which took approximately 15 minutes to complete. Three hundred fifty-one people initiated the survey, for a response rate of 23.1% based on the American Association of Public Opinion Research’s RR3 cal- culation. This response rate was within the acceptable range for panel-based surveys (e.g., Curtin, Presser, & Singer, 2000; Keeter, Kennedy, Dimock, Best, & Craighill, 2006). In addition, of those who did start the survey, 93% (327) completed it. However, as the sample was based on those who initially self-selected for participation in the panel, rather than a true random probability sample, no estimates of sampling error can be calculated. Of the 327 respondents, those who incorrectly answered the quality control questions (e.g., those who did not click Strongly Disagreewhen prompted) were removed from the sample pool. The final sample ( N ¼279) consisted of 142 male and 137 female participants (male coded 0, female coded 1). The minimum reported age was 18 years old and the maximum reported age was 70 years old ( M¼47.42, SD¼12.80). This gender and age breakdown of the sample approximated the average age and gender breakdown of the average voter in the United States (U.S.

Census Bureau, 2013). When asked to report their education levels, 1.4% reported they had completed some high school, 22.9% reported they were high school graduates, 31.1% said they had completed some college, 26.4% said they were college graduates, 4.6% reported they had completed some graduate school, and 13.6% reported having graduate degrees. When 288 PONDER AND HARIDAKIS asked about their political party affiliation, 115 reported they were Democrats, 84 reported they were Republican, 75 reported they were Independent, and five reported ‘‘Other’’ (1 Green Party, 1 Liberal, 1 non- party affiliate, 1 politically neutral, 1 undecided).

Measures Political Social Identity We used Greene’s (1999) partisan social identity scale to measure political social identity. This scale has been applied in numerous studies to measure a person’s level of identification with a political party (Blais, Gidengil, Nadeau, & Nevitte, 2001). A typical item for this measure reads, ‘‘I have a number of qualities typical of members of this group’’ (Greene, 1999, p. 396). This scale consisted of 10 items. Each item asked the respondent to rate his or her agreement with each item on a 5-point scale ranging from 1( strongly disagree )to5(strongly agree ). The responses were summed and averaged to create an index of a person’s level of identification with a political party ( M¼2.98, SD¼0.60, a¼ .81).

Political Opinion Leadership We used the opinion leadership measure developed by Flynn, Goldsmith, and Eastman ( 1996) to identify a person’s level of political opinion leadership. A typical item reads, ‘‘I often influence people’s opinions about [PRODUCT CATEGORY]’’ (Flynn et al., 1996, p. 146). For the current study, PRODUCT CATEGORY was replaced withpolitics. The six-item measure focused on participants’ perceived influence over other people’s opinions with regard to specific topics of interest (e.g., music, cars—and in this study, politics). Participants responded to each of the six items of mea- sure using a 7-point Likert-type scale ranging from 1 ( strongly disagree)to7 ( strongly agree ). Participants’ responses were summed and averaged to create a single opinion leadership score for each participant ( M¼3.74, SD¼1.08, a ¼ .83).

Political Content Affinity We used an adapted version of the political humor affinity scale developed by Hmielowski, Holbert, and Lee ( 2011). Items related to political humor or satire were adapted to measure the broader category of political information.

A typical item reads, ‘‘I appreciate political information because it can make me feel more knowledgeable about politics’’ (Hmielowski et al., 2011, p. 114).

This measure consisted of 11 items scaled on a 5-point scale ranging from 1 IN-GROUP AND OUT-GROUP 289 (strongly disagree )to5(strongly agree ); higher scores indicated a greater affinity for political content. Participants’ scores for all 11 items in the measure were summed and averaged to create a single affinity score for each participant ( M¼3.46, SD¼0.63, a¼ .90).

Motives We measured motives for using media for political information with a 51-item scale drawn from previous research, which identified various motives for interpersonal communication, media use for political information, social identity, sports, and online media use (Flanigan & Metzger, 2001; Kaye & Johnson, 2004; Rubin, 1983; Trail & James, 2001; Wann, 1995). Some motive items were adapted to reflect reasons for using media for political information (e.g., to feel like I have won when my team wins was changed to to feel like I have won when my party wins to reflect reasons for using media for political information).

Respondents were asked to indicate how much each of the 51 motive statements reflected their reasons for using media for political information, ranging from 1 ( not at all)to5(exactly ). Principal components factor analy- sis with varimax rotation was used to analyze respondents’ statements.

Three different motives for using media for political information were ident- ified. These factors explained 66.51% of the total variance after rotation.

The resulting 34-item index resulted in three motive factors (see Table 1 for items and loadings). The three motive factors were (a) idiomatic-use, (b) political utility, and (c) political in-group achievement.

Factor 1, idiomatic-use , contained 15 items ( M¼1.99, SD¼0.88, a¼ .96), reflecting how a person uses mediated political information for personal utility. Factor 2, political utility, contained 14 items ( M¼3.31, SD¼0.86, a ¼ .95), reflecting the use of media to gain information for making political decisions. Factor 3, political in-group achievement , contained four items ( M ¼2.53, SD¼0.98, a¼ .89). This motive reflected how people use media to increase positive feelings associated with their membership in their polit- ical party.

Media use. Respondents were asked to indicate how often (1 ¼never , 5 ¼ very often ) they used each of 14 media sources for acquiring political information about politics. These sources were adapted from the Political Media Diet scale, with four additional sources (wikis, political fact checking sites, home pages, political action websites) added to represent the current state of the information environment available to voters (e.g., Kaid, Fernandes, & Painter, 2011; Rill & McKinney, 2011). Means and standard deviations are reported in Table 2.

290 PONDER AND HARIDAKIS TABLE 1 Factor Loadings for Motives for Using Media for Political Information Items Motive items Idiomatic–use Political utility Political in-group achievement Idiomatic-use So I can get away from what I’m doing .87 .10 .05 So I can get away from the rest of the family or others .84 .01 .15 When there’s no one else to talk to or be with .83 .06 .17 To feel less lonely .83 .04 .15 So I won’t be alone .82 .07 .12 So I can forget about school, work, or other things .82 .03 .01 Because it helps me relax .80 .11 .25 Because it’s a pleasant rest .80 .17 .18 Because it passes time when bored .79 .05 .02 Because it allows me to unwind .79 .15 .24 Because it makes me feel less lonely .79 .03 .20 When I have nothing better to do .77 .02 .06 Because it’s a habit, just something I do .75 .20 .16 Because it peps me up .73 .20 .34 To learn more about myself .72 .25 .20 Political utility To see how the candidates stand on various issues .00 .90 .03 To help me make up my mind about the important issues .04 .85 .09 To see what a candidate would do if elected .03 .85 .06 To find out about issues affecting people like myself .02 .81 .11 To keep up with the main issues of the day .03 .80 .01 To make up my mind how to vote in an election .01 .78 .06 To access political information from home .07 .77 .23 To judge the personal qualities of candidates .11 .77 .08 To access political information any time .10 .75 .21 To access political information quickly .14 .74 .16 To learn more about others .22 .73 .10 To judge who is likely to win an election .26 .63 .25 To remind me of my candidates strong points .12 .63 .31 To get unbiased viewpoints .30 .61 .14 Political in-group achievement To feel like I have won when my party wins .37 .31 .74 To feel a personal sense of achievement when my party does well .43 .37 .67 To feel proud when my party does well .24 .47 .66 To get to know others in my party .45 .35 .63 Eigenvalue 13.64 7.06 1.25 Variance explained 41.32 21.40 3.80 Cronbach’s a .96 .95 .89 M 1.99 3.31 2.53 SD 0.88 0.86 0.98 IN-GROUP AND OUT-GROUP 291 Elaboration.The Perse (1990) five-item cognitive elaboration scale was used to measure respondents’ level of elaboration on political content.

Participants reported how often they have thoughts referenced in the five statements, ranging from 1 ( never)to5( very often ). Responses to the five items were summed and averaged to create an index of cognitive elaboration ( M ¼3.45, SD¼0.82, a¼ .90).

Discussion. Eveland and Shah (2003) measured in-group discussion with one item (e.g., asking liberals how often do they talk with other liberals) and out-group discussion (e.g., asking liberals how often they talk with con- servatives). To obtain a more nuanced measure of in-group and out-group discussion, we expanded this measure to obtain a broader approximation of in-group discussion across five different groups of people from the same political party (coworkers, neighbors, friends, family, and acquaintances) and out-group discussion across those same groups.. Previous researchers have found that a majority of political discussion occurs between people in TABLE 2 Media Use for Political Information Means and Standard Deviations Medium M SD Television news 3.71 1.15 Home pages (e.g., MSN, Yahoo, Google) 3.03 1.26 Internet news sites (e.g., CNN.com, CBS.com, Reuters.com, Politico, NYTimes.com) 2.84 1.20 Radio news 2.74 1.24 Newspapers (e.g., USA Today, New York Times, Wall Street Journal ) 2.62 1.22 Nonsatirical political commentary programming (e.g., Hardball with Chris Matthews, The O’Reilly Factor, The Rachel Maddow Show )2.51 1.31 Satirical late-night programming (e.g., Saturday Night Live,The Colbert Report ,The Daily Show, The Tonight Show with Jay Leno, The Late Show with David Letterman ) 2.40 1.26 Magazines (e.g., Time, Newsweek, People, Fortune ) 2.34 1.21 Social networking sites (e.g., MySpace, Facebook) 2.32 1.24 Political talk radio (e.g., Rush Limbaugh, Glenn Beck, Dennis Miller) 2.15 1.28 Video-sharing websites (e.g., YouTube, Hulu) 2.12 1.17 Books (e.g., biographies =autobiographies of candidates, books about politics or political issues) 2.08 1.15 Political websites (e.g., MoveOn.org, candidate websites, Snopes, FactCheck.org) 2.05 1.14 Blogs =Microblogs (e.g., Drudge Report, The Republican Guru, Fidler’s Politics, Daily Kos, Twitter) 1.96 1.17 Note . Means are based on a 5-point scale. Scores ranged from 1 ( never)to5( very often ).

N ¼279. 292 PONDER AND HARIDAKIS these groups (e.g., Mutz,2002; Mutz & Martin, 2001). To score the amount of discussion, we summed these measures, as the number of discussions was additive, to develop a general political in-group discussion measure ( M ¼6.68, SD¼10.47, a¼ .80) and political out-group measure ( M¼4.45, SD ¼6.54, a¼ .78).

RESULTS In each regression analysis, demographic variables (e.g., age, gender) and personal characteristics (e.g., political opinion leadership, political content affinity, and political social identity) were entered on the first step. Motives for using media for political information were entered on the second step.

Media use was entered on the third step. Finally, elaboration was entered on the fourth step. This order is in line with the theoretical assumptions of the uses and gratifications model that guided this study. To reiterate, the theoretical model suggests that personal characteristics influence motives for using media for political information, which in turn influence media use, which influences elaboration on political information, and ultimately these variables work in concert with each other to influence media effects.

Political Discussion with Political in-Group Members Demographics entered in the first step accounted for 26% of the variance in in-group discussion. Male gender ( b¼ .23, p< .01), political opinion lead- ership ( b¼ .29, p< .001), and political content affinity ( b¼ .22, p< .01) were significant predictors of discussion with political in-group members. Motives for using media for political information, entered on Step 2, explained an additional 4.5% of variance in political discussion with political in-group members. The Fchange was significant ( p< .01). Political in-group achievement ( b¼ .20, p< .05) and political utility ( b¼ .16, p< .05) were sig- nificant predictors. Political opinion leadership and gender remained signifi- cant predictors on this step. Political content affinity was no longer significant. Media use was entered on Step 3. This resulted in an additional 5.2% of the variance explained. The Fchange was not significant ( p¼ .33). Political website use ( b¼ .17, p< .05) and nonsatirical political commentary programming viewing ( b¼ .15, p< .05) were the only significant predictors.

Political opinion leadership and gender remained significant predictors on this step. The political in-group affiliation motive ceased to be a predictor.

Elaboration was entered on Step 4. This accounted for only .3% of the variance in political discussion with political in-group members. The F change was not significant ( p¼ .25), and elaboration was not a significant IN-GROUP AND OUT-GROUP 293 predictor. Political opinion leadership, gender, political websites, and nonsatirical political commentary programming remained significant predictors of political discussion with political in-group members.The final equation accounted for 36.0% of the variance in political discussion with political in-group members (see Table 3). The results suggest that male political opinion leaders who used political websites and nonsatirical political commentary programming for political information TABLE 3 Summary of Regression Analysis for Predicting Political Discussion with Political In-Group and Out-Group Members Variable b(In-Group) b(Out-Group) Step 1: Audience characteristics Age .03 .09 Gender .12 .15 Political social identity .00 .06 Political opinion leadership .22 .23 Political content affinity .06.05 Step 2: Motives for using media Idiomatic-use .08 .10 Political utility .05.11 Political in-group achievement .11.01 Step 3: Media use Home pages .03 .01 Internet news sites .03.01 Blogs =Microblogs .02 .20 Political websites .17 .02 Video-sharing websites .05 .11 Social networking sites .05 .06 Television news .05 .03 Nonsatirical political commentary .15 .20 Programming Satirical late-night programming .02 .02 Television advertisements .04 .04 Movies =Documentaries .03 .04 Political talk radio .00 .12 Radio news .09 .03 Radio advertisements .01 .02 Newspapers .07 .06 Magazines .13 .08 Books .16 .10 Print advertisements .01 .07 Step 4: Cognitive elaboration Elaboration .09 .03 Note. All betas are final betas on the last step of the regression. N¼ 279.

p< .05. p< .01. p< .001. 294 PONDER AND HARIDAKIS discussed politics with people in their political party more than did their counterparts.

Political Discussion with Political Out-Group Members Personal factors entered in the first step accounted for 23.6% of the variance in political discussion with out-group members. Male gender (b¼ .17, p < .01), political opinion leadership ( b¼ .29, p< .001), and political content affinity ( b¼ .19, p< .01) were significant predictors of discussion with political out-group members.

Motives for using media for political information, entered on Step 2, con- tributed an additional 5.4% of variance in political discussion with political in-group members. The Fchange was significant ( p< .001). The political utility motive was a significant predictor of discussion frequency with polit- ical out-group members ( b¼ .19, p< .05). Political opinion leadership and gender remained significant predictors on this step, and political content affinity ceased to be a predictor. Entering media use on Step 3 resulted in an additional 8.3% of the vari- ance explained. The Fchange was significant ( p< .01). Use of nonsatirical political commentary programming ( b¼ .20, p< .05) was a positive predic- tor and use of blogs =microblogs ( b¼ .20, p< .05) was a significant nega- tive predictor of out-group discussion. Political opinion leadership and gender remained significant predictors. Elaboration, entered on Step 4, accounted for less than 1% change in variance explained. The Fchange was not significant ( p¼ .72). Political opi- nion leadership, gender, use of nonsatirical political commentary program- ming, and blogs =microblogs remained significant predictors of political discussion with political out-group members. The final equation accounted for 37.4% of the variance in political dis- cussion with political out-group members (see Table 3). The results suggest that male political opinion leaders, who tended to use nonsatirical political commentary programming but not blogs or microblogs for political infor- mation, tended to discuss politics with people outside their political party more than did their counterparts.

DISCUSSION In this study, we argued that in order to understand political discussion better, scholars need to consider the role of personal characteristics, motives for using media for political information, media use, and elaboration on political issues and examine how these work in concert to influence political discussion with political in-group and out-group members. Guided by uses IN-GROUP AND OUT-GROUP 295 and gratifications theory, we developed and tested a model designed to examine these relationships. The results of our analyses provide a starting point for future investigations into the media system as a whole, and how certain background characteristics, media-use motives, and the use of parti- cular media play a role in facilitating in-group and out-group discussion.

Personal Characteristics As suggested by previous researchers, we found that male participants and political opinion leaders engaged in political conversations with in-group and out-group members. Perhaps even more interesting is that these relation- ships remained consistent, regardless of those with whom they discussed politics. However, because the tenor of political discussions in which the participants engaged was not examined, it is possible that the conversations occurring between in-group members and out-group members were funda- mentally different. For example, these conversations could be arguments or discussions of politics as a whole. However, because we did not examine the tenor of these conversations, we are unable to postulate the nature of these interactions. Future researcher should look at the tenor of these conver- sations (e.g., discussions or arguments) with political out-group members.

Media Use An interesting finding of this study was the role that media use played in predicting political discussion. We looked at 18 different media sources in this study, examining how frequently each media source was used for political information (see Table2) and, ultimately, examining how these sources related to in-group political discussion. Of interest, none of the five most com- monly used sources for political information was related to political discussion when controlling for personal characteristics and motives for using media for political information. This suggests that although political discussion may be a common outcome of using media for political information, some media sources were more influential than others. In addition, the media sources that did seem to influence political discussion were not the most widely used. As identified in the regression equations, only nonsatirical political commentary programming and political website use predicted political discussion with in-group members, and nonsatirical political commentary programming and blog =microblog use predicted discussion with out-group members.

Of interest, nonsatirical political commentary programming was the only type of programming that was positively related to both in-group and out-group discussion. Nonsatirical political commentary programming hosts take time to break down political issues and provide commentary 296 PONDER AND HARIDAKIS on the potential ramifications of a piece of legislation, or even provide an examination of the motives of certain political agents. In doing so, these hosts can serve as opinion leaders for the specific groups in the public.

Previous scholars have found political commentary programs such asThe O’Reilly Factor andThe Rachel Maddow Show are popular programs because viewers use these programs to develop, influence, and inform their own opinions on politics (e.g., Conway, Grabe, & Grieves, 2007; McCombs, Holbert, Kiousis, & Wanta, 2011). Furthermore, Norton ( 2011) suggested that the stylistic approaches used by hosts of these types of programs provide interpretive frames for their viewers to understand the world around them and readily recall that information in later conversations, essentially training them for the political discourse boxing ring. Moreover, these programs provide users with the opportunity to learn about their own party’s views on specific issues, as well as the views of oppos- ing parties. Many of these programs commonly bring on guest commentators from a variety of political stances that allow viewers to be exposed to the opinions of others. In addition, many of these programs’ hosts dissect the arguments of their opponents on the show, allowing their viewers to learn more about politics and arming these viewers with the tools to defend their own opinions from politically different stances. Furthermore, this format allows viewers to understand how an argument with a person from a differ- ent political party may progress from a relatively safe vantage point. Of interest, political blogs and microblogs were a negative predictor of dis- cussion with political out-group members. This may suggest that the people who use blogs and microblogs are a unique cross-section of society. Specifi- cally, scholars have found that people who regularly use political blogs and microblogs for their political information are the most likely to immerse themselves in partisan echo chambers (Kim & Johnson, 2012; Saleton, 2012). Previous scholars have suggested that political blog users—a small section of society—are strongly partisan people who seek out only those political messages that are in-line with their already existing political views while avoiding information that contradicts their notions of the political arena (e.g., Brundidge & Rice, 2009). Our results suggest that it would seem that the people who use political blogs and microblogs for their political information fit into this cross-section of society—at least to the extent that they may not use these particular social media to discuss politics with politi- cally disparate others. In addition, although we did not specifically measure for political polarization, other scholars have suggested that consistent avoidance of differing political opinions can lead people to become more polarized in the political attitudes, which can lead to the election of more par- tisan officials and, ultimately, more partisan gridlock (Jamieson & Cappella, 2008 ; Hetherington, 2001). IN-GROUP AND OUT-GROUP 297 This finding may provide a fruitful venue for future research in examining the role of blogs and other online information sources to determine the impact of these sources on political extremism, political polarization, and attitudes toward out-groups. The fact that certain social media—blogs and microblogs—were negative predictors of outgroup discussion may temper claims of others that these are robust tools for political dialogue. For instance, Vis ( 2014) explained that Twitter and other blogs have a history of quashing large-scale misinformation campaigns, specifically citing the 2011 UK riots as examples where people took to Twitter to debunk the rumor well before any traditional media channel. The fact that use of these tools did not predict in-group discussion, but detracted from out-group discussion suggests that these particular media sources were not used widely for these purposes in our sample.

Limitations There were many different limitations to this study. One such limitation was the response rate to our survey. Although 327 participants were sufficient for our analysis, it is important to recognize that this is a small number of people from which to reach definitive conclusions about what influences the entire U.S. population to discuss politics. Although previous researchers have reported similar response rates, and indeed have found that response rates as low as 20% do not impact the results of a survey, ideally a higher response rate would have allowed for more generalizable findings. Next, we did not account for who initiated the discussions. Presumably, people who initiate discussions are different from those who happen to join a conversation because of proximity. Previous investigations have found that people who initiate conversations are more active in their media use and are more likely to control the topic of conversation (Himielboim, Gleve, & Smith, 2009). Understanding these differences could be useful for future investigations. In addition, we did not account for the type of discussions that occurred.

It would be useful to understand the nature of these discussions with in-group and out-group members. As discussed earlier, a variety of different types of discussion can occur. Moreover, Korostelina (2007) suggested that in situations where people’s views differ on politics, they are more likely to engage in arguments about politics; however, when people have a similar view, they are more likely to engage in a casual conversation. Presumably, there are differences between casual conversations about politics and argu- ments about politics. The nature of the questions we asked did not explore this area. The differences, at least in terms of frequency, between casual conversations and arguments could have influenced the results of this study. 298 PONDER AND HARIDAKIS Finally, where there were several different background characteristics we did include, there are numerous psychographic factors in addition to these that may provide interesting insight into political discussion and should be considered in future research. For instance, we did not include a measure of introversion. Previous scholars have found that introverts, or those who are described as more reserved, quiet, or shy, are less likely to socialize than extroverts (McCrae & Costa, 1987). In this light, it would be relevant to know if the participants who engaged in political discussion were more likely to be introverted or extroverted. In addition, we could have measured the participants’ political self-efficacy or confidence in their political beliefs.

Presumably, those who are more confident in their beliefs may engage in political discussion more than those with less confidence.

REFERENCES Bennett, S. E., Flickinger, R. S., & Rhine, S. L. (2000). Political talk over here, over there, overtime. British Journal of Political Science ,30 , 99–119.

Borgida, E., Federico, C. M., & Sullivan, J. L. (2009). The political psychology of democratic citizenship . New York, NY: Oxford University Press.

Brundidge, J., & Rice, R. E. (2009). Political engagement and exposure to heterogeneous political discussion: Do the (information) rich get richer and the similar get more similar?

In A. Chadwick & P. N. Howard (Eds.), The handbook of internet politics(pp. 134–156).

New York, NY: Routledge.

Conway, M., Grabe, M. E., & Grieves, K. (2007). Villain s, victims and the virtuous in Bill O’Reilly’s ‘‘No-Spin Zone’’: Revisiting wor ld war propaganda techniques.Journalism Studies,8, 197–223.

Curtin, R., Presser, S., & Singer, E. (2000). The effects of response rate changes on the index of consumer sentiment. Public Opinion Quarterly ,64 , 413–428.

Delli-Carpini, M. X., & Keeter, S. (1996). What Americans know about politics and why it matters . New Haven, CT: Yale University Press.

Denton, R. E., & Woodward, G. C. (1998). Political communication in America. New York, NY: Praeger.

Eveland, W. P., Jr. (2004). The effect of political discussion in producing informed citizens: The roles of information, motivation, and elaboration. Political Communication,21 , 177–193.

doi:10.1080 =10584600490443877.

Eveland, W. P., Jr., & Dunwoody, S. (2000). Examining information processing on the World Wide Web using think aloud protocols. Media Psychology,2, 219–244.

Eveland, W. P., Jr., & Hively, M. H. (2009). Political discussion frequency, network size, and ‘‘heterogeneity’’ of discussion as predictors of political knowledge and participation. Journal of Communication ,59 , 205–224. doi:10.1111 =j.1460-2466.2009.01412.x Eveland, W. P., Jr., Shah, D., & Kwak, N. (2003). Assessing casuality in the Cognitive Mediation Model: A panel study of motivations, information processing, and learning during Campaign 2000. Communication Research ,30 , 359–386. doi:10.1177/0093650203253369 Flanigan, A. J., & Metzger, M. J. (2001). Internet use in the contemporary media environment. Human Communication Research ,27 , 153–181. doi:10.1093/hcr/27.1.153 Flynn, L. R., Goldsmith, R. E., & Eastman, J. K. (1996). Opinion leaders and opinion seekers: Two new measurement scales. Journal of the Academy of Marketing Science ,24 , 137–147.

IN-GROUP AND OUT-GROUP 299 Greene, S. (1999). Understanding party identification: A social identity approach.Political Psychology ,20 , 393–403. doi:10.1111 =0162-895X.00150 Gue ´ rin, D., Petry, F., & Cre ˆ te, J. (2004). Tolerance, protest and democratic transition: Survey evidence from 13 post-communist countries. European Journal of Political Research,43 , 371–395. doi:10.1111/j.1475-6765.2004.00158.x Hanson, G., Haridakis, P., Wagstaff, A., Sharma, R., & Ponder, J. (2011). The 2008 Presidential Campaign: Political cynicism in the age of Facebook, MySpace and YouTube. Mass Communication & Society ,13 , 584–507. doi:10.1080 =15205436.2010.513470 Harwood, J. (1999). Age identification, social identity gratifications and television viewing. Journal of Broadcasting & Electronic Media ,43 , 123–136. doi:10.1080 =08838159909364479 Hetherington, M. J. (2001). Resurgent mass partisanship: The role of elite polarization. American Political Science Review ,95 , 619–631.

Himelboim, I., Gleave, E., & Smith, M. (2009). Discussion catalysts in online political discussions: Content importers and conversation starters. Journal of Computer Mediated Communication ,14 (4), 771–789. doi:10.1111/j.1083-6101.2009.01470.x Hmielowski, J. D., Holbert, R. L., & Lee, J. (2011). Predicting the consumption of political TV satire: Affinity for political humor, The Daily Show, andThe Colbert Report .Communication Monographs ,78 , 96–114. doi:10.1080 =03637751.2010.54257 Hogg, M. A., & Reid, S. A. (2006). Social identity, self-categorization, and the communication of group norms. Communication Theory ,16 , 7–30.

Holbrook, R. A., & Hill, T. (2005). Agenda-setting and priming in prime time television: Crime dramas as political cues. Political Communication,22 , 277–295. doi:10.1080=1058460059 1006519 Huddy, L. (2001). From social to political identity: A critical examination of social identity theory. Political Psychology ,22 , 127–156.

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

Kaid, L. L., Fernandes, J., & Painter, D. (2011). Effects of political advertising in the 2008 Presidential Campaign. American Behavioral Scientist ,55 , 437–456. doi:10.1177=00027642 11398071 Kaye, B., & Johnson, T. (2004). A web for all reasons: Uses and gratifications of internet components for political information. Telematics and Informatics,21 , 197–223. doi:10.1016 = S0736-5853(03)00037-6 Keeter, S., Kennedy, C., Dimock, M., Best, J., & Craighill, P. (2006). Gauging the impact of growing nonresponse on estimates from a national RDD telephone survey. Public Opinion Quarterly ,70 , 759–779.

Kim, D., & Johnson, T. J. (2012). Political blog readers: Predictors of motivations for accessing political blogs. Telematics and Informatics ,29 ,doi:10.1016/j.tele.2011.04.003 Kim, J. K., Wyatt, R. O., & Katz, E. (1999). News, talk, opinion, participation: The part played by conversation in deliberative democracy. Political Communication,16 , 361–385.

Korostelina, K. V. (2007). Social identity and conflict: structures, dynamics, and implications .

New York: Palgrave-MacMillan.

Kwak, N., Williams, A., Wang, X., & Lee, H. (2005). Talking politics and engaging politics: An examination of the interactive relationships between structural features of political talk and discussion engagement. Communication Research ,32 ,87–111. doi:10.1177 =0093650204271400 Lazarsfeld, P. F., Berelson, B., & Gaudet, H. (1948). The people’s choice,second edition. New York, NY: Columbia University Press.

Levy, M. R., & Windahl, S. (1984). Audience activity and gratifications: A conceptual clarification and exploration. Communication Research,11 , 51–78. doi:10.1177 =0093650840 11001003 300 PONDER AND HARIDAKIS McCombs, M., Holbert, L., Kiousis, S., & Wanta, W. (2011).The news and public opinion:

Media effects on civic life . Malden, MA: Wiley.

McCrae, R. R., & Costa, P. T., Jr. (1987). Validation of the Five-Factor Model of personality across instruments and observers. Journal of Personality and Social Psychology, 52 , 81–90.

McLeod, J. M., & Becker, L. B. (1974). Testing the validity of gratification measures through political effects analysis. In J. G. Blumler & E. Katz (Eds.), The uses of mass communications:

Current perspectives of gratifications research (pp. 137–164). Beverly Hills, CA: Sage.

Mendez, J. M., & Osborn, T. (2010). Gender and the perception of knowledge in political discussion. Political Research Quarterly ,63 , 269–279. doi:10.1177 =1065912908328860 Moy, P., & Gastil, J. (2006). Predicting deliberative conversation: The impact of discussion networks, media use, and political cognitions. Political Communication,23 , 443–460.

doi:10.1080 =10584600600977003 Mutz, D. C. (2002). The consequences of cross-cutting networks for political participation.

American Journal of Political Science ,46 , 838–855.

Mutz, D. C., & Martin, P. S. (2001). Facilitating communication across lines of political difference: The role of mass media. American Political Review,95 , 97–114.

Norton, M. (2011). A structural hermeneutics of The O’Reilly Factor.Theory and Society , 40 , 315–346.

Papacharissi, Z., & Rubin, A. M. (2000). Predictors of Internet use. Journal of Broadcasting & Electronic Media ,44 , 175–196. doi:10.1207 =s15506878jobem4402_2 Perse, E. M. (1990). Media involvement and local news effects. Journal of Broadcasting & Electronic Media ,34 , 17–36.

Pew Research Center for the People and Press. (2009). Public evaluations of the news media:

1985–2009 press accuracy rating hits two decade low . Retrieved fromhttp://people-press.

org/2009/09/13/press-accuracy-rating-hits-two-decade-low/ Pfau, M., Houston, J. B., & Semmler, S. M. (2005). Presidential election campaigns and American democracy: The relationship between communication use and normative outcomes. American Behavioral Scientist ,49 , 48–62.

Putnam, R. D. (2000). Bowling alone: America’s declining social capital. New York, NY: Simon and Schuster.

Rill, L. A., & McKinney, M. S. (2011). Talking politics: Young citizens’ interpersonal interaction during the 2008 presidential campaign. In M. S. McKinney & M. C. Banwart (Eds.), Communication in the 2008 U.S. Election: Digital natives elect a president (pp. 51–70).

New York, NY: Peter Lang.

Rubin, A. M. (1983). Television uses and gratifications: The interactions of viewing patterns and motivations. Journal of Broadcasting ,27 , 27–51.

Rubin, A. M. (2009). The uses-and-gratifications perspective of media effects. In J. Bryant & M. B. Olliver (Eds.), Media effects: Advances in theory and research (3rd ed., pp. 165–184).

Mahwah, NJ: Erlbaum.

Rubin, A. M., & Perse, E. M. (1987). Audience activity and television news gratifications. Communication Research ,14 , 58–84.

Saletan, W. (2010, March 3). Bubble think: How to escape a partisan echo chamber. Retrieved from: http://www.slate.com/id/2252685/ Scheufele, D. A. (2002). Examining differential gains from mass media and their implications for participatory behavior. Communication Research,29 , 46–65. doi:10.1177 =009365020202 900103 Shah, D. V., Kwak, N., & Holbert, R. L. (2001). ‘‘Connecting’’ and ‘‘disconnecting’’ with civic life: The effects of Internet use on the production of social capital. Political Communication, 18 , 141–162. IN-GROUP AND OUT-GROUP 301 Shah, D. V., & Scheufele, D. A. (2006). Explicating opinion leadership: Nonpolitical disposi-tions, information, consumption, and civic participation. Political Communication,23 , 1–22.

Stroud, N. J. (2008). Media use and political predispositions: Revisiting the concept of selective exposure. Political Behavior ,30 , 341–366. doi:10.1007 =s11109-007-9050-9 Trail, G. T., & James, J. D. (2001). The motivation scale for sport consumption: Assessment of the scale’s psychometric properties. Journal of Sport Behavior,24 , 108–127.

U.S. Census Bureau. (2013). Voting and registration in the Election of November 2012 . Retrieved from http://www.census.gov/hhes/www/socdemo/voting/publica tions/p20/2012/tables.html Vis, F. (2014). The rapid spread of misinformation online . Retrieved fromhttp://www.

huffingtonpost.com/farida-vis/the-rapid-spread-of-misinformation-online_b_4665678.html Wann, D. L. (1995). Preliminary validation of the sport fan motivation scale. Journal of Sport & Social Issues ,19 , 377–396.

302 PONDER AND HARIDAKIS Copyright ofMass Communication &Society isthe property ofTaylor &Francis Ltdand its content maynotbecopied oremailed tomultiple sitesorposted toalistserv without the copyright holder'sexpresswrittenpermission. However,usersmayprint, download, oremail articles forindividual use.