How individual sensitivities to disagreement shape youth political expression on Facebook

Full text

How individual sensitivities to disagreement shape youth political expression on Facebook
Emily K. Vragaa
evraga@gmu.edu
Kjerstin Thorsonb
kjerstin.thorson@usc.edu
Neta Kligler-Vilenchikb
kliglerv@usc.edu
Emily Geeb
emilygee88@gmail.com
AUTHOR ACCEPTED VERSION
a George Mason University, 4400 University Drive, MS 3D6, Fairfax, VA 22030, United States.
b University of Southern California, 3502 Watt Way, Los Angeles, CA, 90089, United States.
Please address correspondence to Emily Vraga at the Department of Communication, MS 3D6,
George Mason University, via phone at 703-993-1090, or via email at evraga@gmu.edu
Full article citation: Vraga, E. K., Thorson, K., Kligler-Vilenchik, N., & Gee, E. (2015). How individual
sensitivities to disagreement shape youth political expression on Facebook. Computers in Human
Behavior, 45, 281-289. https://www.sciencedirect.com/science/article/pii/S0747563214007419

ABSTRACT
Social networking sites like Facebook increasingly shape youth engagement with politics, but
less is known about the factors that shape willingness to engage in political interaction on the
site. This study combines twenty in-depth interviews with a survey of young adults to examine
how individual predispositions, perceptions of the Facebook political climate, and network
characteristics shape attitudes and behaviors toward posting political content on Facebook. Our
results suggest that predispositions like political interest and conflict avoidance create distinct
“sensitivities” to Facebook’s political climate and the potential for contentious political
disagreement that condition willingness to post about politics.
Keywords: social media, political communication, discussion norms, conflict avoidance,
heterogeneous discussion

Highlights
• Mixed-methods study to examine youth perceptions of political content on Facebook
• Complicated social norms govern appropriate youth political expression on Facebook
• Qualitative interviews suggest politics on Facebook associated with rants and drama
• Survey finds effects of disagreement depend on interest and conflict avoidance
• Disagreement on Facebook only uncomfortable and demobilizing for some individuals

How individual sensitivities to disagreement shape youth political expression on Facebook
ABSTRACT
Social networking sites like Facebook increasingly shape youth engagement with politics, but
less is known about the factors that shape willingness to engage in political interaction on the
site. This study combines twenty in-depth interviews with a survey of young adults to examine
how individual predispositions, perceptions of the Facebook political climate, and network
characteristics shape attitudes and behaviors toward posting political content on Facebook. Our
results suggest that predispositions like political interest and conflict avoidance create distinct
“sensitivities” to Facebook’s political climate and the potential for contentious political
disagreement that condition willingness to post about politics.
Keywords: social media, political communication, discussion norms, conflict avoidance,
heterogeneous discussion

How individual sensitivities to disagreement shape youth political expression on Facebook
1. Introduction
On Election Day 2012, more than nine million Facebook users clicked an “I voted”
button, telling their friends they had participated in the election (Bakshy, 2012a). More than a
third of Americans who use social networking sites (SNS) report using them to like or promote
materials related to politics and social issues, with 34 percent posting or commenting on these
topics (Rainie, Smith, Schlozman, Brady, & Verba, 2012). Facebook use is linked to increased
social capital and political participation (Gil de Zúñiga, Jung, & Valenzuela, 2012; Valenzuela,
Park, & Kee, 2009; Vitak, Zube, Smock, Carr, Ellison, & Lampe, 2011), the potential for
incidental exposure to news and political information (Bode, 2012; Kim, 2011; Valenzuela,
2013), and may even serve as a space for political conversation and deliberation (Authors,
2013a; Fernandes, Giurcanu, Bowers, & Neely, 2010). In addition, there is growing evidence for
normatively positive effects stemming from online political expression itself (Shah, Cho,
Eveland, & Kwak, 2005; Östman & Ekström, 2013).
Scholarly attention has focused more on the outcomes of political uses of SNS rather than
exploring what motivates such uses (though see Baek et al., 2011; Bumgarner, 2007 on
motivations for SNS use in general). These are important questions to ask because Facebook is a
distinct social setting for political talk. On the one hand, the size and relative diversity of
Facebook information flows could produce exposure to new ideas, spurring engagement and
interest (Bakshy, 2012b; Bode, 2012; Vitak et al., 2011). On the other hand, heterogeneity in
offline discussion networks often limits political conversation, primarily because of social
pressures to preserve harmony (Eliasoph, 1998; Festinger & Thibaut, 1951; Mutz, 2006).
Facebook takes discussion heterogeneity to a new level, as compared to most offline social

settings. The audience for a political post on Facebook combines people from multiple social
spheres of one’s life and, because content posted to SNS can be copied, shared, and spread
widely or could be hidden from some friends by a news feed display algorithm, the “real”
audience on Facebook is indeterminate (Litt, 2012; Marwick & boyd, 2011).
The purpose of this article is to explore why some young citizens post about politics on
Facebook and others do not. In doing so, we consider but also push beyond the most obvious
answer—political interest—to show how individual characteristics interact with perceptions of
the Facebook political climate to shape willingness to engage on the site. We use a mixed
method approach, heeding calls for wider utilization of “methodological pluralism” (Johnson &
Onwuegbuzie, 2004, p. 15) to achieve a richer understanding of our research questions through
the integration of qualitative and quantitative approaches (Cresswell & Clark, 2006). We draw
first on twenty in-depth interviews with 18-29 year olds during the 2012 election cycle, probing
variation among youth in their perceptions of the political climate on Facebook and exploring
how these perceptions shape willingness to talk about politics on the site. In study two, we draw
on a convenience sample survey of young adults to test a series of hypotheses emergent from the
qualitative analysis about the role of individual sensitivities in predicting engagement.
1.1 The social settings for political talk
A great deal of attention has been paid to the conditions that facilitate the emergence of
political talk, and, in particular, to the conditions that enable the “cross-cutting” talk that permits
exposure to multiple viewpoints on political issues (Mutz, 2006). Research produces mixed
results about the extent to which citizens talk about politics and how different social settings
shape the content of those conversations (McClurg, 2006; Kwak et al., 2005; Wyatt, Kim, &
Katz, 2000). Some ethnographic studies find that political talk is relatively rare and often

difficult (Eliasoph, 1998), while others find that rates of casual political talk are quite high
(Huckfeldt & Sprague, 1987; Kim, Wyatt, & Katz, 1999).
Part of the reason for this discrepancy is that different kinds of social settings are more or
less favorable to political talk. Social pressures to preserve harmony and norms of politeness that
urge people to just “get along” mean that bringing up politics can seem like a risky act,
especially under conditions of opinion or partisan heterogeneity (Eveland & Hively, 2009;
Klofstad, Sokhey, & McClurg, 2013; Mutz, 2006). Despite the fact that many theories of
democracy identify exposure to multiple viewpoints as central to democratic practice,
heterogeneity in discussion networks is frequently identified as a negative predictor of political
talk (Klofstad et al., 2013), and, in some forms, can suppress political participation and create
higher levels of opinion ambivalence (Knoke, 1990; Mutz, 2006; Nir, 2011).
Most of what we know about heterogeneity and discussion networks comes from studies
of face-to-face social groups and interest driven online communities (e.g., Mutz, 2006; Eliasoph,
1998; Wojcieszak & Mutz, 2009; although see Kim, 2011). Despite a great deal of scholarship
concerning activist uses of SNS (Authors, 2013b; van Laer & van Alelest, 2010), there has been
little research to explore how SNS function as settings for everyday political talk. In what
follows, we offer a brief review of how sites like Facebook compare to more well-understood
contexts for political conversation.
1.2 Heterogeneity, disagreement, and political interaction on Facebook
Scholars studying social network sites (SNS) have identified several ways in which SNS
differ from most everyday forms of offline social interaction. First, expression on SNS involves
addressing a “networked public” (boyd, 2010), characterized by persistence (expressions are
archived), replicability (content can be duplicated), scalability (content can be made visible

beyond its initial audience) and searchability (archived content can be found via search). The
networked audience on Facebook combines friends and acquaintances from different social
spheres, and this “context collapse” creates potentially uncomfortable tension from the
intersection of these competing contexts (Marwick & boyd, 2011). Although Facebook friend
networks tend to map closely to offline social networks, individuals from different parts of that
network—friends, family, work colleagues, distant acquaintances—may all be part of the
audience for any post to the site.
Second, political content flows through Facebook are more heterogeneous than those that
characterize most offline social networks (Kim, 2011; Pew, 2012). As Huckfeldt and Sprague
(1987) observed in studying flows of political content in offline social networks, the supply of
political discussion in one’s environment is not fully under individual control. Heterogeneity on
Facebook is related to two features of the site: (1) individuals are unlikely to select Facebook
friends on the basis of shared political opinions and (2) the proliferation of loose ties (e.g.,
people we are not particularly close to who populate our Facebook friends), both of which limit
the possibility for de facto selective exposure based on shared background and ideals (Bakshy,
Rosenn, Marlow, & Adamic, 2012; Haythornthwaite, 2005; Ellison, Steinfeld, & Lampe, 2007).
Third, the above-mentioned affordances of Facebook mean that the imagined audience
for any given post is not only potentially heterogeneous and context collapsed, but the extent of
this heterogeneity is impossible to determine (Litt, 2012). The networked audience combines
features of the unknown “broadcast” audience with features of more personal audiences like a
group of friends at dinner or work colleagues out for a drink (Marwick & boyd, 2011). This in
turn creates pressures on strategies for impression management (Rui & Stefanone, 2013).
1.3 Young citizens and perceptions of the Facebook political climate

The brief review above offers reasons to wonder how young citizens will navigate
political expression in the Facebook context. Although survey-based studies suggest that during
a highly politicized time, most young citizens will be exposed to at least some political content
through the site (Rainie et al., 2012), we know much less about how imaginings of a
heterogeneous audience, the potential for risking exposure to those who may disagree with your
political opinions, and the thought of speaking to a context collapsed audience shape decisions to
post political content and what kinds of political content to post.
We explore these questions first through in-depth interviews with young citizens. In
addition to the questions about heterogeneity, disagreement, and audience raised above, our
research questions were also developed with an eye to bridging two distinct traditions in the
study of political talk. First, we consider how perceptions of political climate—not simply the
presence of disagreement, but also the amount and tone of political discussion—shape
willingness to talk about politics on the site (Kim et al., 1999; Noelle-Neumann, 1984). Second,
we explore how perceptions of political interaction on Facebook are shaped by individual
differences. Previous studies have shown that, as Hayes et al. note, “nonparticipation is, at least
in part, a social psychological phenomenon” (p. 261; also Matthes, 2013). Our second goal is
therefore to explore whether and how perceptions of politics on Facebook depend on individual
motivational and personality characteristics.
RQ1: How did young citizens perceive the Facebook political climate (the amount and
tone of political discussion) during the 2012 elections?
RQ2: How do perceptions of the political climate shape young people’s willingness to
express themselves politically on the site?
RQ3: How are individual differences among respondents related to their willingness to
express themselves politically on Facebook?
2. Study 1

2.1 Study 1 materials and methods
We developed a semi-structured, Facebook-aided interview protocol in which questions
about life history, media use, and political engagement were combined with a review of the
participant’s Facebook profile and news feed. We conducted interviews with twenty 18-29 year
olds (11 females, 9 males, mean age=22.3). We used purposive sampling in two locations on the
East and West coast, targeting participants who were college students or recent graduates, and
who were Facebook users.
1
We chose respondents that could give us insight into exposure and
reception of political content, thus the sample included a mix of more and less politically
interested participants. The interviews were conducted between September and November 2012,
covering the period of the three presidential debates, as well as the election itself.
We viewed and talked through several days’ worth of content on each respondent’s
Facebook feed, which served as conversation starters for their wider perceptions around posting
and engaging with political content on Facebook. The interviews lasted approximately 60-90
minutes, and were audio recorded with participants’ permission and transcribed. For the
interview guide and additional methodological details, see Authors, forthcoming.
2
We used the qualitative data analysis program Dedoose (www.dedoose.com) to aid us in
identifying and coding patterns in the ways interviewees related to political content on Facebook,
both in terms of seeing other’s content, their wider perception of the Facebook opinion climate,
and their own posting behavior. Through an inductive analysis, we first identified an initial large
set of themes and coded those in interview transcripts. Through iterative in-depth analysis, we
!
1
We used a snowball sampling approach to develop our sample. Initial participants were recruited through classes
and via fliers posted on a college campus. These initial participants were then encouraged to recommend other
participants, which increased the number of college graduates, in addition to college students.
2
For review purposes, the interview guide and methodological appendix are attached as a supplemental file.

then narrowed down to the emergent themes most relevant to our research questions, and
identified patterns in these themes across respondents. Our analysis of these themes is presented
by research question, paying attention both to common patterns among most participants, and to
the idiosyncrasies of specific respondents. In the following analysis, respondents are identified
by pseudonym and age.
2.2 Study 1 results
When we asked interview participants about their perceptions of politics on Facebook,
many opened with a blanket statement of dislike for “political content” on the site. We quickly
discovered that by “political content,” respondents were referencing their most salient
experiences of Facebook politics: “rants” and the social drama they produce. Jake (24)
explained,
On Facebook, obviously you have friends that you aren't friends with. That you
just have, that you might have and didn't really talk to them. You see their
political affiliations happening in front of your eyes. You almost can't wonder
how you are friends with them or how this even happened. There are many people
who will just spout out the most craziest, homophobic rants after any major news
story relating to Obama or Romney. It is difficult to bear.
For Mary (20), people seeking drama were often to blame for the nastiness around politics:
I think people are more trying to be politically neutral on Facebook, because unless
they're [neutral] they want controversy. I see people like my friends fighting over things.
I remember one of my friends posted up a very controversial status. It just blew up. And
then the mob. It just became nasty. Just like people unfriending each other. And just like,
‘I unfriended you now.’
Rants were viewed as “obnoxious,” (Mary, 20), “offensive” (Laura, 24), or “annoy[ing]” (Harry,
19), and, as Jake pointed out, they were tied to the vastness and heterogeneity of friend networks
on Facebook. The interviewees had on average more than 750 Facebook friends and nearly all of
the respondents noted at least one, and sometimes several among their Facebook acquaintances,
whose political views are different from their own.

I have one friend who is... and it's fine, many of my friends are Atheists, but it's
like he would deliberately find articles about the Catholic church just to post and
piss people off. It was like, I already know this is how you are. I already accept
you for this, but you don't have to post so much. (Katie, 22)
There is a complicated relationship between the “rants” described to us in interviews and
what scholars of political talk normally mean by exposure to disagreement. In most studies,
disagreement is operationalized as either an identified difference of opinion within a dyad
(supporting a different candidate or taking opposing positions on an issue) or as heterogeneity of
partisanship within a network (Eveland & Hively, 2009; Klofstad et al., 2013). In our interviews,
these forms of disagreement were less salient than the tone in which an opinion was expressed.
When Jake tells his story, it matters that he is seeing opinions different than his own—he does
not share his Facebook friend’s “homophobic” views—but this cognitive disagreement is less
important than his revulsion at the tone. Similarly, Katie does not mind that her friend has
politicized views on religion (even though she disagrees), what bothers her is that her friend is
trying to “piss people off.” Rants are a form of emotionally charged disagreement, even more
salient in the minds of these youth than disagreement on issues.
It is no surprise that these dramatic memories of rants and drama would be top of mind
for interview respondents. Their rant stories are both vivid and negative, two characteristics of
media messages that promote accessibility in memory (Shrum, 2008). However, that these
observations of charged disagreement and their uncomfortable consequences (Facebook fights,
potential unfriending) are so salient has important ramifications for how the young citizens we
talked to perceive the political climate on Facebook.
First, the accessibility of these rants shaped the initial negative assessment of political
content on Facebook. As the interviews proceeded and we looked through individuals’ Facebook
news feeds, we saw and heard about forms of politics on Facebook that were widely seen as

more palatable. The distinctions most of our respondents made can be characterized as the
difference between controversial rants about electoral politics (perceived negatively) and
“consensual” posts about civic engagement (perceived positively). For example, interviewees
tended to think that encouraging voting via Facebook was fine: “Tell people to vote. It’s
awesome,” said Erin (21). Similarly, supporting social causes was deemed appropriate: “I think
cause stuff is OK. It's all about the way it's portrayed for me” (Lisa, 24). Likewise, many of our
respondents said they enjoyed or didn’t mind seeing political news stories on Facebook, at least
when the content was “unbiased.” Yet mentions of these better-liked forms of political content
appeared slowly in the interviews—and were markedly less salient than stories about rants.
The second way in which disagreement and “drama” shape perceptions of politics on
Facebook is by triggering a belief in political “neutrality” as the primary safe strategy for posting
about politics on the site. We saw above how Mary directly contrasts neutrality and controversy:
“I think people are more trying to be politically neutral on Facebook, because unless they're
[neutral] they want controversy.” In another example, Julie (19) explained:
Yeah, I think generally with the things that I say and stuff, like I don't really like
saying things that I know will make waves. So that's why I don't post anything
about politics or post any rivalry things. I feel like if I have really strong opinions
about something, Facebook's not necessarily the place that I'd want to put it on.
(Julie, 19)
When we dig deeper, it becomes clear that neutrality is preferred not because respondents do not
have opinions. Most (though not all) do have policy and candidate preferences, and many report
discussing these with their friends in offline settings. It is not even simply that Facebook is seen
as a heterogeneous environment, a characteristic of social settings that is known to make some
political interactions uncomfortable (Mutz, 2006). Instead, it is these factors combined with the
unknowability of the imagined audience that raises the stakes for political expression (Litt,

2012). Bay, 22, explicitly links her reluctance to post about politics on Facebook to concerns
about conflict, heightened by context collapse: “I don't post a lot of political stuff. Definitely not,
because I don't want to alienate certain friends. I don't know everyone's political opinion. I don't
want to offend anyone so I stay away from that.”
Among our respondents, about one third reported they occasionally posted political
content. But of these posters, nearly all said they post content that is “more neutral” (Carrie, 22)
or, as Harry (19) says:
I'm more likely to put something that's not opinionated, something that's ‘look at this’
type thing versus this is my opinion about what's said. A lot of people [during the
debates] were putting up the fact checker website. I looked at those because that's the
main thing you should look at after a debate is the fact check from lots of websites
because even fact check websites are biased.
Ben, age 23, explicitly linked his decision to avoid opinions in his post with the desire to
manage a contentious political audience on Facebook: “I'm getting to be a little bit more active in
posting political stuff, but I still would never [post opinionated content]. I don't want to get into
those comment wars, and I don't want to offend anyone.” Concerns linked to the unknown
audience combined with fear of “comment wars” and rants to guide not only choices about
whether to post but also what to post.
Another method used to manage the heterogeneous audience was humor. Many of those
who posted about politics noted that they try to keep their posts relatively light-hearted:
I mean, I usually don't post anything on politics because I don't want anyone to know
what I'm thinking […] If I do post anything that's kind of newsish, I'll just post it to
someone I think is interested and then the people that are also interested, that's fine.
Generally, if I post stuff on Facebook, it's mostly funny stuff and sometimes I get likes
from people that I have no idea ever...I don't even remember them, so I'm like oh, this is
nice. (James, 21)
In fact, humorous political content allows people who wouldn’t otherwise post about
politics to share their views. Katie, 22, was not willing to post about her opinions on who won

the first presidential debate, but she was willing to post about Romney’s Big Bird comment: “I
think both sides would just think that that was randomly funny, not making fun of Mitt Romney,
not saying Obama won, but something in the middle and funny about it.”
In these findings we see how perceptions of the networked audience as big, unknowably
diverse, and containing individuals who have no qualms about creating drama and controversy,
act to shape political posting behavior on Facebook. And yet we still saw a great deal of variation
even among our rather homogeneous group of respondents. Some were willing to post, some
didn’t care enough to do so, and others actively engaged in self-censorship. Although none of
our respondents identified as the kind of person who provokes comment wars on Facebook, we
know from their stories that these provocateurs are out there as well. Our initial analysis left us
with a question: if these negative perceptions of the political climate on Facebook are widely
shared, why is there so much variation in the way such perceptions shape behavior? For answers,
we turned to the life history portion of our interviews.
2.3 Factors accounting for individual differences
Research on the evolution of norms and practices on SNS has shown that because social
settings on SNS are highly ambiguous, offline practices are especially likely to shape online
behaviors (McLaughlin & Vitak, 2012). This is exactly what we found when we compared the
political socialization histories of our respondents who did post about politics on Facebook with
those that did not. We found that not only are certain types of politics more valued than others,
but that some people are more open toward political expression than others, and that the
distinction is related to pre-existing individual differences. Non-posters like Elisa explained,
Politics never was interesting to me because I feel like you go through one issue
and you go to another issue. I hate never-ending arguments. I just really don't like
it and I feel like politics is a huge one. When you get to religion and stuff, all that
stuff, I just don't like conflict and arguments and it's really centered around that so

I didn't like to get into it that much. There was one golden rule that I followed in
high school and it was don't talk about politics and don't talk about religion with
people unless you want to get into a huge debate. (Elisa, 21)
Sandi equated the way she feels about political discussions on her Facebook to her
reaction to heated political arguments that happen in her extended family gatherings:
I don't like feeling forced to express my opinion in terms of politics, because I feel
like it's kind of private. I know there's things that I think that probably most of the
people in my life disagree with, or vice versa. I don't like being forced to have to
stand up for one side or the other, to make an argument on one side or the other,
just because I don't feel the need to argue for the sake of arguing. (Sandi, 24)
We saw for both of these respondents that their feelings about Facebook were shaped by conflict
avoidant attitudes that pre-date their usage of SNS. References to general conflict avoidance
appear as well among respondents when they are talking about their preference for political
neutrality. Previous studies have shown that conflict avoidance is related to lower levels of
political participation as well as willingness to self-censor (Ulbig & Funk, 1999; Hayes,
Scheufele, & Huge, 2006). Further, experimental research suggests that conflict avoidance
increases sensitivity to uncivil political messages. Mutz and Reeves (2005) found that watching
uncivil political debate on television depressed levels of trust in government only among those
who are conflict avoidant. Among those who enjoy political conflict, the pattern was reversed.
3. Building hypotheses from the interview findings
We found that respondents initially reported a dislike of political content on Facebook,
primarily because their most salient memories of politics on the site are of witnessing charged
displays of controversy. These in turn heightened uncertainties about the imagined audience for
Facebook posts, creating a tight association between opinionated posts and the fear of drama.
That is, respondents were not sure who would read political posts or how they would react, but
the salience of “ranters” means that negative consequences like comment wars appeared as a

likely outcome. Ranters serve as vivid, if not representative, exemplars of political interaction on
Facebook. As such, the political climate on Facebook is seen as high risk. In our study, these
memories of rants and fights are akin to the uncivil television spots that Mutz and Reeves (2005)
developed to test the effects of incivility on political attitudes—and we postulate they have
similar effects, further heightened by the context collapsed social environment of Facebook.
Not all of our respondents allowed these perceptions to prevent them from posting, with a
third reporting they did post about politics. In our interviews, these posters generally attempted
to avoid drama and conflict by sticking to posts that were neutral or expressing opinions only
under cover of humorous content. However, given Mutz and Reeves’ findings, we might imagine
that a subset of the population exists for whom the possibility of stirring controversy would
actually motivate opinionated political posting on Facebook. Such a mechanism would explain
the continued presence of the opinionated “ranters,” at least one (or more) of whom were
identified in the networks of nearly all our respondents.
Both our interviews and extant literature therefore suggest two likely suspects to explain
why young citizens react differently to the possibility of political drama on Facebook: political
interest (as a proxy for general political involvement), which is one of the most important
predictors of nearly all forms of political participation (Verba, Schlozman & Brady, 1995), and
conflict avoidance, a predisposition that is an important predictor of self-censorship in the
context of political talk (Hayes et al., 2006; Ulbig & Funk, 1999). Our findings leave us with a
clear set of propositions about the relationships among conflict avoidance and political interest,
the salience of political disagreement on Facebook, and attitudes toward the appropriateness of
the site as a place to engage in political interaction.
Our first three hypotheses predict direct effects of individual differences and perceptions

of disagreement on attitudes toward Facebook as a place for political interaction, and on
willingness to post political content:
H1: Young citizens with higher levels of political interest will be more likely to (a)
believe Facebook is an appropriate place for politics and (b) post political content.
H2: Young citizens who are more conflict avoidant will be less likely to (a) believe
Facebook is an appropriate place for politics and (b) post political content.
H3: Young citizens who perceive more political disagreement on their Facebook news
feed will be less likely to (a) believe Facebook is an appropriate place for politics and (b)
post political content.
We then consider the interaction between these individual differences and the salience of
political disagreement on one’s own Facebook feed. We propose that being conflict avoidant and
less politically interested creates a sensitivity to disagreement, such that both perceptions of the
Facebook opinion climate and willingness to post will be lower among these groups.
H4: The impact of political disagreement on (a) believing Facebook is an appropriate
place for politics and (b) posting political content will be conditioned by individual
predispositions, such that the effects of political disagreement will be stronger among
those who are less interested in politics.
H5: The impact of political disagreement on (a) believing Facebook is an appropriate
place for politics and (b) posting political content will be conditioned by individual
predispositions, such that the effects of political disagreement will be stronger among
those who are more conflict avoidant.
4. Study 2
4.1 Study 2 material and methods
To test our hypotheses, we conducted a survey using Amazon’s Mechanical Turk, a
crowdsourcing service in which participants receive money for completing tasks. Previous
research has suggested that the Mechanical Turk population is similar to other Internet samples,
reliable, and substantially more diverse than populations of American college students
(Buhrmester, Kwang, & Gosling, 2011; Casler, Bickel, & Hackett, 2013; Mason & Suri, 2012).

That said, we make no claims as to the generalizability of these findings to all young Facebook
users. Instead, they are meant to extend our findings from the qualitative interviews by exploring
the underlying mechanics linking individuals’ predispositions, attitudes about the Facebook
opinion climate, and posting behaviors, among a similar, larger sample.
3
The survey was fielded on November 5, 2012, the day before the presidential election.
Four hundred participants were paid $0.75 for taking part in a 10-minute survey. In the analyses
below, we include only respondents who reported being between the ages of 18-29, to match our
qualitative interviews, leaving 231 respondents (Mage=24.15, S.D.=3.07).
4.2 Measures
Political Orientations. Several indicators measured political orientations. A single item
asked respondents on a six-point scale which response best described their party affiliation,
which was recoded to compare Democrats to Republicans (80.8% Democrat).
4
This scale also
was folded to a 0-2 point scale to measure partisan strength (M=.76, S.D.=.75). A single item
asked on a seven-point scale how interested participants were in politics (M=4.99, S.D.=1.64;
see Appendix A for question wording).
Conflict Avoidance. Four questions measured on seven-point scales tapped into
respondents’ feelings towards conflict around politics (Mutz & Reeves, 2005). These items were
!
3
Our Mechanical Turk sample is different from young adults overall in several ways. Compared to Pew data from
February of 2012 (examining SNS users ages 18-29), our sample is more educated and more Democratic in their
party affiliation than young social media users in general. Roughly 80% of our sample classified themselves as
Democrats, compared to 67% of young adults in the Pew sample when excluding Independents. In our sample, 89%
have at least some college education, compared to 54% of the Pew sample. However, these differences generally
reflect the educated, largely Democratic population that we drew from for our qualitative interviews. Further,
research suggests that the Internet population overall – including Facebook users – tends to be more skewed towards
the young, the educated, and the liberal, much like our sample populations (Pew, 2013).
!
4
We forced people to choose an affiliation with a party by eliminating “true Independent” as an option from the six-
point scale. Instead, Independents were forced to select which party they leaned towards. 17 people refused to
answer the question and were excluded from the analysis.

combined into an index (alpha=.84, M=3.71, S.D.=1.33).
Network Characteristics. A single item measured reliance on Facebook on a seven-
point scale (M=4.94, S.D.=1.94). A second measure gauged network size, asking respondents
about their number of friends, with a median response of 201-250 friends.
Respondents also reported on seven-point scales how much political and news content
they saw on their Facebook feed (alpha=.78, M=4.96, S.D.=1.21). We measured perceptions of
political agreement and disagreement on Facebook using two indicators which asked respondents
on four-point scales how often they agree and disagree with political opinions and content posted
by friends on social networking sites (Magree=2.26, S.D.=.61, Mdisagree=2.27, S.D.=.62).
Perceptions of the Facebook political climate. Eight items gauged people’s beliefs
about political content on Facebook. These items were adapted from the common themes that
people discussed in the qualitative interviews, including the suitability of Facebook as a place for
politics, their enjoyment of such content, and the potential negative consequences of politics on
the site, including drama, anger, and discomfort. Factor analysis revealed two factors: 1) beliefs
about the appropriateness of Facebook as a place for politics, and 2) the outcomes of such
political expression – outcomes that are largely negative (Appendix D). Based on these analyses,
we created two variables indicating the belief that Facebook is an appropriate place for politics
(alpha=.83, M=3.70, S.D.=1.42) and the belief that politics on Facebook produces negativity
(alpha=.81, M=4.54, S.D.=1.25). These two factors are negatively correlated (r=-.49, p<.001).
Posting about politics. We examined how much people enjoy sharing political content
and images and how often they post political content on Facebook using a three item index, with
each question answered on a seven-point scale (alpha=.91, M=2.98, S.D.=1.70).
4.3 Study 2 results

4.3.1 Predicting perceptions of the Facebook opinion climate
To test these relationships, we used Hayes (2013) PROCESS macro, using model 2 and
applying mean centering for the independent variables. The results confirm that individual
predispositions play a large role in explaining attitudes toward the Facebook political climate,
supporting H1a and H2a. Political interest increases the belief that Facebook is an appropriate
place for politics (β=.16, S.E.=.06, p<.01), while depressing perceptions that it will produce
negative outcomes (β=-.10, S.E.=.06, p<.10). Conflict avoidance produces the opposite pattern:
heightening perceptions that political expression on Facebook produces negativity (β=.29,
S.E.=.07, p<.001) and inhibiting beliefs that Facebook is an appropriate space for politics (β=-
.39, S.E.=.07, p<.001).
Table 1: Predicting Facebook Attitudes and Behaviors
Facebook is an appropriate
place for politics
Politics on Facebook
produces negativity
Posting political content
Beta
S.E.
Beta
S.E.
Beta
S.E.
Age
.04
.03
-.02
.03
-.00
.03
Gender
.17
.17
-.16
.16
.10
.18
Education
-.11
.11
.20
.11
-.08
.12
Democrat
-.12
.20
-.03
.19
.18
.21
Partisan Strength
.16
.11
-.11
.11
.14
.12
Adjusted R2
.040*
.041*
.048**
Conflict avoidant
-.39***
.07
.29***
.07
-.14†
.08
Political interest
.16**
.06
-.10†
.06
.20**
.06
Adjusted R2
.320***
.155***
.339***
Facebook is part of
daily routine
.07
.04
-.09*
.04
.00
.05
Number of Facebook
friends
-.05
.04
.10**
.03
-.02
.04
See political content
.17*
.07
.14*
.07
-.06
.08
Agreement on
Facebook
.22
.15
-.09
.14
.16
.16

Disagreement on
Facebook
-.21
.14
.46**
.13
.23
.15
Adjusted R2
.344*
.271***
.336
Conflict X
Disagreement
-.27*
.11
.13
.10
-.28*
.12
Interest X
Disagreement
-.10
.09
.06
.09
-.21*
.10
Adjusted R2
.358*
.270
.368**
Facebook is an
appropriate place for
politics
---
---
---
---
.57***
.08
Politics on Facebook
produces negativity
---
---
---
---
-.06
.08
Adjusted R2
---
---
---
---
.521***
Adjusted R-squares for blocks calculated using multiple regression models
† p≤.10, * p<.05, ** p<.01, *** p<.001
Perceptions of disagreement within one’s network are also important in explaining
attitudes towards Facebook as a political space. We find mixed support for H3a: the presence of
political disagreement heightens beliefs that political expression will create negativity (β=.46,
S.E.=.13, p<.01), but is not related to beliefs about its appropriateness (β=-.21, S.E.=.14, p=n.s.).
Finally, we test whether the effect of political disagreement is conditioned by individual
predispositions. We find no support for H4a: political interest does not moderate the impact of
disagreement on attitudes about political expression on Facebook. We do, however, find mixed
support for H5a: there is a significant interaction between conflict avoidance and disagreement
(β=-.27, S.E.=.11, p<.05) on beliefs about the appropriateness for Facebook for political
discussion. Deconstructing this interaction supports our hypothesis: disagreement is particularly
influential among those who are highly conflict avoidant, with relatively high levels of
disagreement depressing beliefs that Facebook is an appropriate place for politics (see Figure 1).
Meanwhile, among the less conflict avoidant, seeing disagreement does not impact their attitudes
towards Facebook as an appropriate site for politics.

Figure 1: Effects of conflict avoidance and exposure to disagreement on beliefs that Facebook is
an appropriate place for politics
4.3.2 Predicting Posting of Political Content
Next, we examine how these individual predispositions, perceptions of the political
climate, and attitudes about political content on Facebook are linked to posting behavior.
Individual predispositions are also powerful in predicting posting behavior (see Table 1). Both
H1b and H2b are supported: conflict avoidance inhibits posting (β=-.14, S.E.=.08, p<.10), while
political interest promotes posting (β=.20, S.E.=.06, p<.01). But, in contrast with our earlier
models and H3b, characteristics of the Facebook network are not significant as a block, nor does
seeing political disagreement predict posting behaviors (β=.23, S.E.=.15, p=n.s.).
However, the effect of seeing political disagreement is conditioned by predispositions
towards politics and conflict. As predicted by H4b and H5b, significant interactions emerge for
both political interest (β=-.21, S.E.=.10, p<.05) and conflict avoidance (β=-.28, S.E.=.12, p<.05).
For the highly conflict avoidant, seeing political disagreement does not impact posting behaviors.
The conflict avoidant remain relatively unlikely to post political content (see Figure 2). But
0
1
2
3
4
5
6
7
Low!disagreement Average
disagreement
High!disagreement
Low!conflict!avoidance
Average!conflict!avoidance
High!conflict!avoidance

among the low conflict-avoidant – in other words, those who enjoy conflict – seeing
disagreement encourages more posting behaviors. This finding matches the pattern of effects
found in Mutz et al.’s studies of the effects of incivility.
Figure 2: Effects of conflict avoidance and exposure to disagreement on political posting
behaviors
The pattern is reversed for political interest: those with high levels of political interest are
likely to post about politics, regardless of the amount of political disagreement they see (see
Figure 3). But in contrast to our expectations, political disagreement can actually spur those
generally uninterested in politics to join the conversation by sharing their own political posts.
This suggests that at least under some circumstances, political disagreement may serve to
mobilize those who are generally uninterested in politics or who enjoy political conflict.
Figure 3: Effects of political interest and exposure to disagreement on political posting behaviors
0
1
2
3
4
5
6
7
Low!disagreement Average
disagreement
High!disagreement
Low!conflict!avoidance
Average!conflict
avoidance
High!conflict!avoidance

5. General Discussion
Facebook is often lauded in the scholarly literature for its potential to serve as a source of
political information and discussion for young adults, and the resulting possibility to enhance
knowledge and engagement (Bode et al., 2012; Gil de Zúñiga et al., 2012; Valenzuela, 2013).
Yet not enough is known about how young citizens navigate political content in this complex
social space, nor how attitudes about political expression are related to the characteristics of the
individual and the egocentric networks they inhabit on Facebook. This study attempts to fill this
void through a mixed-methods design, combining qualitative interviews with young adults about
political uses of social media with a quantitative survey of a similar group, testing the
relationships between individual sensitivities to disagreement and posting behaviors.
Our interviews suggested that for many young citizens, politics on Facebook is most
often associated with rants, drama, and virulent disagreement. When asked about political
content on Facebook, their first response – and for many, their most salient experience – is
describing the rogue political provocateur, deliberately stirring up trouble. Thus, for many, the
political climate associated with Facebook created the same discomfort that limits the voicing of
0
1
2
3
4
5
6
7
Low!disagreement Average
disagreement
High!disagreement
Low!political!interest
Average!political!interest
High!political!interest

political opinions in public settings that ethnographic research has previously uncovered
(Eliasoph, 1998). Further, it was not just the existence of disagreement, but the acrimony of such
disagreement, paired with a context-collapsed, networked audience that heightened tendencies
toward self-censorship (Marwick & boyd, 2011).
However, it is worth considering whether the identification of “rants” and the distaste
they engender results from truly uncivil exchanges, or whether classifying many political
expression as “rants” is a way to minimize a viewpoint participants disagree with. This question
is one this study cannot answer, yet merits future inquiry. At the same time, even if these “rants”
are largely respondents’ perceptions and the content itself is civil and measured, that these are
the perceptions of Facebook disagreement that many youth carry with them have consequences
for attitudes and posting behaviors that we begin to uncover here.
Yet avoidance of politics on Facebook was by no means absolute. Not only could our
interview participants almost always name individuals in their network who posted political
content, but also a third of the young adults we talked to occasionally posted about politics. The
posters identified by others were often distinguished not just by their ability to withstand the
conflict that often resulted, but also by an enjoyment of such disagreement. For them, the
heterogeneous and potentially hostile political climate, which served as a deterrent for their more
conflict avoidant peers, whetted their appetite for political engagement in this social space.
The quantitative survey allowed us to more precisely test and validate these interactions.
Political interest and conflict avoidance not only emerged as two of the most important
predictors of attitudes about political expression and posting behaviors on Facebook, but also
conditioned responses to political disagreement. For the conflict avoidant, higher levels of
disagreement further depressed an already dubious sense that Facebook could be an appropriate

place for political discussion. Conversely, disagreement encouraged those who enjoy conflict
(e.g., those low in conflict avoidance) to post more frequently. This result confirms many of our
young adults’ suspicions: some individuals posting about politics enjoy being provocateurs,
posting not in spite of the potential for disagreement but because of it. Similarly – and in contrast
to our expectations – political disagreement had the potential to boost posting behaviors among
those relatively uninterested in politics. Much like research suggesting that negative advertising
and incivility have the potential to encourage political participation (Brooks & Geer, 2007; Fung
et al., 2012), it may be that seeing political disagreement in their social network, often in the
forms of rants, reminds the uninterested of the importance of the campaign. In other words, the
passionate partisan advocacy of friends, family, and peers may provide another way that social
influence concerning political engagement is spread on Facebook (Bakshy, 2012b).
What the quantitative survey cannot tell us is the types of political posts that are most
likely to occur. Turning back to the qualitative interviews, even the subset of our participants
who posted about politics were wary of the potential risks. In response, they often chose to frame
their posts in terms of neutral information sharing and humorous observations about the political
process. These tactics mirror those adopted by other news professionals: news organizations rely
on the mantel of objectivity to avoid accusations of partisan bias (Cunningham, 2003; Kovach &
Rosensteil, 2007), while humor is valued for its potential to mitigate anger or aggression (Miron,
Brummett, Ruggles, & Brehm, 2008; Authors, 2012). In the end, while there may be a minority
of provocateurs posting opinionated political content on Facebook, many others are posting
about politics in more tentative ways, designed to minimize the risks of inspiring the ire of
others.
Of course, this study is limited in its ability to generalize its findings. The qualitative

interviews gave us a rich sense of how some young adults are thinking about the possibilities for
sharing political content on Facebook, but these findings may not apply to other populations, or
to all young adults. Our survey quantitatively tests and models the relationships uncovered
during the qualitative interviews, yet our online sample is biased towards educated and
Democratic respondents (a population comparable to those people we interviewed). Future
research should continue to explore the connections between individual dispositions and the
practices of Facebook politics among the population more broadly.
Our study is also bounded by the context of the 2012 presidential election. We
deliberately tested our expectations during a period when posting about politics on Facebook was
likely at its height: towards the end of a long, competitive campaign. Our same respondents
might feel very differently about the appropriateness of political content being shared via
Facebook and its potential to produce negativity during a period when passions are running less
high and the competition between the parties is not so extreme.
Overall, our research provides another step in explaining why research on the effects of
disagreement on political interaction have so often been mixed (Huckfeldt et al., 2004; Mutz,
2006). Researchers are beginning to parse the variety of meanings that heterogeneity can have
and their differing implications, including diverse versus dangerous discussion (Eveland &
Hively, 2009) and general versus partisan disagreement (Klofstad et al., 2013). Our qualitative
interviews suggest it is not just the existence of disagreement, but also the tone of that
disagreement that is particularly memorable and off-putting in the Facebook context. Our
quantitative survey allowed us to better test exactly what contributes to perceptions of the
political climate on Facebook. While personal predispositions matter, the Facebook network that
individuals were embedded in was also consequential in shaping these beliefs. People with a

larger network of Facebook friends and those who see more disagreement were more likely to
believe that political discussion on Facebook was fraught with drama and discomfort. Our
interviews suggest that the images of disagreement that were likely triggered by our survey were
of rants and comment wars. Thus, our results suggest that heterogeneity will be most detrimental
for political participation and expression when that disagreement becomes synonymous with
negativity, discord, and discomfort.
Further, our quantitative survey provided evidence that disagreement’s effects are more
nuanced than originally thought; seeing more political disagreement did not have a direct impact
on posting behavior, but one that was channeled through personal orientations towards politics
and conflict. Future research needs to continue to test these assumptions.
Ultimately, this study speaks to the promise and the pitfalls of political discussion on
social networking sites. Many of the same normative and practical concerns that apply to offline
political discussion (Eliasoph, 1998; Mutz, 2006) are reflected and even reinforced online. Yet
while the young adults we talked to and surveyed were aware of and concerned with these risks,
they did not disdain political talk on Facebook altogether. Instead, they developed complicated
attitudes regarding political expression: valuing information exchange (rather than opinion),
humor (rather than rants), and consensual civic engagement (rather than contested political
advocacy). And ultimately, those who were more motivated and able—those who cared deeply
about politics and who were less conflict avoidant—were more likely to see the benefits of
political expression on Facebook and to contribute to such expression themselves, in spite of (or
sometimes because of) the potential risks of disagreement.

References
Baek, K., Holton, A., Harp, D., & Yaschur, C. (2011). The links that bind: Uncovering novel
motivations for linking on Facebook. Computers in Human Behavior, 27(6), 2243-2248.
doi:10.1016/j.chb.2011.07.003
Bakshy, E. (2012a). The 2012 Election day through the Facebook lens. Last accessed 12.8.2014.
Retrieved from: https://www.facebook.com/notes/facebook-data-science/the-2012-
election-day-through-the-facebook-lens/10151181043778859
Bakshy, E. (2012b). Rethinking information diversity in networks. Last accessed 12.8.2014.
Retrieved from: https://www.facebook.com/notes/facebook-data-team/rethinking-
information-diversity-in-networks/10150503499618859
Bakshy, E., Rosenn, I., Marlow, C., & Adamic, L. (2012). The role of social networks in
information diffusion. Proceedings of the 21st international conference on World Wide
Web, 519-528. doi: 10.1145/2187836.2187907!
Baumgartner, J. C., & Morris, J. S. (2010). My FaceTube politics: Social networking web sites
and political engagement of young adults. Social Science Computer Review, 28, 24-44.
doi:10.1177/0894439309334325
Bode, L. (2012). Political information 2.0: A study in political learning via social media
(Doctoral Dissertation).
boyd, d. (2010). Social network sites as networked publics: Affordances, dynamics, and
implications. In Z. Papacharissi (Ed.) Networked Self: Identity, Community, and Culture
on Social Networking Sites (pp. 39-58). New York: Routledge.
Brooks, D. J., & Geer, J. G. (2007). Beyond negativity: The effects of incivility on the electorate.
American Journal of Political Science, 51, 1-16. doi: 10.1111/j.1540-5907.2007.00233.x
Buhrmester, M., Kwang, T., & Gosling, S.D. (2011). Amazon's Mechanical Turk: A New Source
of Inexpensive, Yet High-Quality, Data? Perspectives on Psychological Science, 6, 3-5.
doi: 10.1177/1745691610393980
Bumgarner, B. A. (2007). You have been poked: Exploring the uses and gratifications of
Facebook among emerging adults. First Monday,12(11).
Casler, K., Bickel, L., & Hackett, E. (2013). Separate but equal? A comparison of participants
and data gathered via Amazon’s MTurk, social media, and face-to-face behavioral
testing. Computers in Human Behaviors, 29, 2156-2160. doi:10.1016/j.chb.2013.05.009
Cresswell, J.W., & Clark, V.L (2006). Designing and conducting mixed methods research. Los
Angeles, CA: Sage.
Cunningham, B. (2003). Rethinking objectivity. Columbia Journalism Review, 42, 24-32.
Eliasoph, N. (1998). Avoiding politics: How Americans produce apathy in everyday life.
Cambridge, MA: Cambridge University Press.
Ellison, N. B., Steinfield, C., & Lampe, C. (2007). The benefits of Facebook “friends:” Social
capital and college students’ use of online social network sites. Journal of Computer-
Mediated Communication, 12(4), 1143-1168. doi: 10.1111/j.1083-6101.2007.00367.x
Eveland, W. P., & 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
Fernandes, J., Giurcanu, M., Bowers, K. W., & Neely, J. C. (2010). The writing on the wall: A
content analysis of college students’ Facebook groups for the 2008 presidential election.
Mass Communication and Society, 13, 653-675. doi:10.1080/15205436.2010.516865

Festinger, L., & Thibaut, J. (1951). Interpersonal communication in small groups. Journal of
Abnormal and Social Psychology, 46, 92-99.
Gil de Zuniga, H., Jung, N., & Valenzuela, S. (2012). Social media use for news and individuals’
social capital, civic engagement, and political participation. Journal of Computer-
Mediated Communication, 17, 319-336. doi: 10.1111/j.1083-6101.2012.01574.x
Hayes, A. F., Scheufele, D. A., & Huge, M. E. (2006). Nonparticipation as self-censorship:
Publicly observable political activity in a polarized opinion climate. Political Behavior,
28(3), 259-283. doi: 10.1007/s11109-006-9008-3
Hayes (2013). Introduction to mediation, moderation, and conditional process analysis. New
York: Guilford Press.
Huckfeldt, R. R., Johnson, P. E., & Sprague, J. (2004). Political disagreement: The survival of
diverse opinions within communication networks. Cambridge: Cambridge University
Press.
Huckfeldt, R., & Sprague, J. (1987). Networks in context: The social flow of political
information. American Political Science Review, 81, 1197-1216.
Johnson, R.B., & Onwuegbuzie, A.J., (2004). Mixed methods research: A research paradigm
whose time has come. Educational Researcher, 33(7), 14-26.
doi: 10.3102/0013189X033007014
Kim, Y. (2011). The contribution of social networking sites to exposure to political difference:
The relationship among SNSs, online political messaging, and exposure to cross-cutting
perspectives. Computers in Human Behaviors, 27, 971-977. doi:
10.1016/j.chb.2010.12.001
Kim, J., Wyatt, R. O., & Katz, E. (1999). News, talk, opinion, and participation: The part played
by conversation in deliberative democracy. Political Communication 16, 361-385. doi:
10.1080/105846099198541
Klofstad, C. A., Sokhey, A. E., & McClurg, S. D. (2013). Disagreeing about disagreement: How
conflict in social networks affects political behavior. American Journal of Political
Science, 57, 120-134. doi: 10.1111/j.1540-5907.2012.00620.x
Knoke, D. (1990). Networks of political activism: Toward theory construction. Social Forces,
68, 4, 1041-1063.
Kovach, B. & Rosensteil, T. (2007). The elements of journalism: What newspeople should know
and the public should expect. New York: Crown Publishers.
Litt, E. (2012). Knock, knock. Who’s there? The imagined audience. Journal of Broadcasting
and Electronic Media, 56, 330-345. doi:10.1080/08838151.2012.705195
Marwick, A. E., & boyd, d. (2011). I tweet honestly, I tweet passionately: Twitter users, context
collapse, and the imagined audience. New Media & Society, 13(1), 114-133.
doi: 10.1177/1461444810365313
Mason, W., & Suri, S. (2012). Conducting behavioral research on Amazon’s Mechanical Turk.
Behavior research methods, 44(1), 1-23. doi: 10.3758/s13428-011-0124-6
Matthes, J. (2013). Do hostile opinion environments harm political participation? The
moderating role of generalized social trust. International Journal of Public Opinion
Research, 25, 23-42. doi: 10.1093/ijpor/eds006
McClurg, S.D. (2006). Political disagreement in context: The conditional effect of neighborhood
context, disagreement, and political talk on electoral participation. Political Behavior, 28,
349-366. doi: 10.1007/s11109-006-9015-4
McLaughlin, C., & Vitak, J. (2012). Norm evolution and violation on Facebook. New Media &

Society, 14, 299-315. doi:10.1177/1461444811412712
Miron, A. M., Brummett, B., Ruggles, B. & Brehm, J. W. (2008). Deterring anger and anger-
motivated behaviors. Basic and Applied Social Psychology, 30, 326-338. doi:
10.1080/01973530802502259
Mutz, D. C., & Reeves, B. (2005). The new videomalaise: Effects of televised incivility on
political trust. American Political Science Review, 99, 1-15.
doi: http://dx.doi.org/10.1017.S0003055405051452
Mutz, D. C. (2006). Hearing the other side: Deliberative versus participatory democracy.
Cambridge, MA: Cambridge University Press.
Nir, L. (2011). Disagreement and opposition in social networks: Does disagreement discourage
turnout? Political Studies, 59, 674-692. doi: 10.1111/j.1467-9248.2010.00873.x
Noelle-Neumann, E. (1984). The Spiral of Silence: Public Opinion-Our Social Skin. Chicago, IL:
University of Chicago Press.
Östman, J., & Ekström, M. (2013). Information, interaction and creative production: The effects
of three forms of Internet use on youth democratic engagement. Advance online
publication. Communication Research. doi:10.1177/0093650213476295
Pew (2012, September). Politics on Social Networking Sites. Last accessed 12.8.2014. Pew
Internet and American Life Project. Retrieved from
http://pewinternet.org/~/media//Files/Reports/2012/PIP_PoliticalLifeonSocialNetworking
Sites.pdf
Rainie, L., Smith, A., Schlozman, K. L., Brady, H., & Verba, S. (2012, Oct. 19). Social media
and political engagement. Last accessed 12.8.2014. Pew Internet and American Life
Project. Retrieved from
http://pewinternet.org/~/media//Files/Reports/2012/PIP_SocialMediaAndPoliticalEngag
ement_PDF.pdf
Rui, J. R., & Stefanone, M. A. (2013). Strategic image management online: Self-presentation,
self-esteem, and social network perspectives. Information, Communication, and Society,
16, 1286-1305.
Shah, D. V., Cho, J., Eveland, W. P., & Kwak, N. (2005). Information and expression in a digital
age: modeling Internet effects on civic participation. Communication research, 32(5),
531-565. doi: 10.1177/0093650205279209
Shrum, L. J. (2008). Effects and underlying processes. Media Effects: Advances in Theory and
Research, 50.
Ulbig, S. G., & Funk, C. L. (1999). Conflict avoidance and political participation. Political
Behavior, 21(3), 265-282. doi: 10.1023/A:1022087617514
Valenzuela, S. (2013). Unpacking the use of social media for protest behavior: The roles of
information, opinion expression, and activism. American Behavioral Scientist, 57, 920-
942. doi:10.1177/0002764213479375
Valenzuela, S., Park, N., & Kee, K. F. (2009). Is there social capital in a social networking site?:
Facebook use and college students’ life satisfaction, trust, and participation. Journal of
Computer-Mediated Communication, 14, 875-901. doi: 10.1111/j.1083-
6101.2009.01474.x
Van Laer, J., & van Aelst, P. (2010). Internet and social movement action repertoires:
Opportunities and limitations. Information, Communication, and Society, 2010, 1146-
1171. doi:10.1080/13691181003628307
Verba, S., Schlozman, K. L., & Brady, H. F. (1995). Voice and equality: Civic voluntarism in

American politics. Cambridge, MA: Harvard University Press.
Vitak, J., Zube, P., Smock, A., Carr, C. T., Ellison, N., & Lampe, C. (2011). It’s complicated:
Facebook users’ political participation in the 2008 election. Cyberpsychology, Behavior,
and Social Networking, 14, 107-114. doi:10.1089/cyber.2009.0226
Wyatt, R. O., Kim, J., & Katz, E. (2000). How feeling free to talk affects ordinary political
conversation, purposeful argumentation, and civic participation. Journalism and Mass
Communication Quarterly, 77, 99-114. doi: 10.1177/107769900007700108
Wojcieszak, M. E., & Mutz, D. C. (2009). Online groups and political discourse: do online
discussion spaces facilitate exposure to political disagreement? Journal of
Communication, 59(1), 40-56. doi: 10.1111/j.1460-2466.2008.01403.x