Content uploaded by Sung Woo Yoo
Author content
All content in this area was uploaded by Sung Woo Yoo on Feb 09, 2019
Content may be subject to copyright.
The role of heterogeneous political discussion and partisanship on the effects of
incidental news exposure online
Sung Woo Yoo and Homero Gil De Zúñiga
ABSTRACT
Scholars have for some time shed light on the effects incidental news exposure have for the
democratic process. However, limited work has explored how the ease and openness of discus-
sion online interplay with unintentional encounters with news in explaining citizens’political
engagement. Using a U.S. national survey, this study seeks to contribute to the literature by
testing the mediating role of heterogeneous and homogeneous political discussion in predicting
the relationship between incidental news exposure and political participation. Findings show that
heterogeneous discussion fully mediates the relationship with offline participation. The relation-
ship with online participation was partially mediated. Mediation of homogeneous discussion to
political participation did not occur. Moreover, a moderated mediation analysis finds that the
mediation of heterogeneous discussion is more likely to occur among partisans than nonpartisans.
Limitations and further suggestions to advance this line of research are provided in this study.
KEYWORDS
Incidental news exposure;
political participation;
political discussion;
partisanship; mediation;
moderated mediation;
digital media
Digital technology has empowered people to have
more control over news than ever before.
Information is now abundant and widely accessi-
ble online. People selectively consume media con-
tent that agrees with their needs. However, for the
very reason that people have more choice, it is
argued they are less likely to engage in civic activ-
ities (Prior, 2007; Sunstein, 2007).
This paradox has long been explained by the
argument that the new media environment drasti-
cally reduced chances for unintended encounters
with news about public affairs. In the world of
traditional media, people used to stumble upon
news while going about other daily business
(Fiorina, 1990; Tewksbury, Weaver, & Maddex,
2001). Information acquired by chance was called
“a free bonus”(Baum, 2002, p. 96), because it
spurred meaningful participatory activities of the
less sophisticated people. What is paradoxical is
that when people have other options than viewing
news, this path to engagement is not free anymore
as they pay the opportunity cost of foregoing con-
tents such as entertainment (Downs, 1957; Prior,
2007, Rittenberg, Tewksbury, & Casey, 2012).
Scholars argue the loss of this bonus is detrimental
to democracy as a whole.
However, recent research shows evidence the
digital media use is positively related to political
engagement overall (e.g., Boulianne, 2009,2015).
Thus, the prediction that the increased choice of
information will decrease encounter with political
news, and thereby weaken the political engage-
ment of the people did not come true. And yet,
few research adequately addressed why the premo-
nition did not turn into reality.
We note the recent evolution of media technology
has focused on expressive forms of communication
and to have conversations. Now, people almost simul-
taneously exchange thoughts about news they stumble
upon via social media or mobile messaging appliances
(Bae, Kwak, & Campbell, 2013;Vraga,Anderson,
Kotcher, & Maibach, 2015). The ease of discussion
may have crucially changed the way people are
affected by the news they accidentally stumble upon.
Not only are people able to overcome the barrier of
time and space with less effort, but also the boundary
of primary network or even political views when hav-
ing political talks online. In short, incidental news
exposure positively relates to political engagement
because the opportunity cost of discussion decreased.
We test this argument by examining the mediation of
homogenous and heterogeneous political discussion
CONTACT Sung Woo Yoo swyoo@utexas.edu P.O. Box 2000, Cortland, NY 13405
Color versions of one or more of the figures in the article can be found online at www.tandfonline.com/witp.
JOURNAL OF INFORMATION TECHNOLOGY & POLITICS
https://doi.org/10.1080/19331681.2018.1561346
© 2019 Taylor & Francis
on the relationship of incidental news exposure online
and participatory activities.
Additionally, this study tests the effect of poli-
tical antecedent that has been thought to restrict
the scope of news reception and discussion
online: Individuals’level of political partisanship.
To reevaluate its impact, the paper tests the
moderation of partisanship on the mediating
role of political discussion. This study is an
advancement to the previous studies that exam-
ined the influence of partisanship on online
communication (e.g., Brundidge, 2010), in that
now it investigates how partisanship conditions
the specific path of news –discussion –partici-
pation. In brief, this study attempts to depict
a modified picture of the effect of incidental
news exposure in accordance with the recent
evolution of digital media, and show why it is
still a viable path to political participation.
Incidental news exposure in the digital age
One of the persistent concerns in political commu-
nication is that the public has neither capability nor
intention to be sufficiently informed about public
affairs to make democracy work (Converse, 1964;
Delli Carpini & Keeter, 1996). Nevertheless, people
are constantly provided with a certain amount of
public affairs information, because they are acciden-
tally exposed to standardized news packages of the
mass media (Tewksbury et al., 2001). Considered as
an indispensable conduit of balanced information,
scholars have incessantly studied ‘news discovery by
chance.’This category of news encounter has been
given various names: “incidental news exposure”
(Tewksbury et al., 2001), “inadvertent exposure”
(Brundidge, 2010), “by-product”(Downs, 1957), or
more descriptively, “information acquired while
doing other things”(Atkin, 1973). Called by many
terms, a safe definition would be “encounter with
news without intention.”That is, that portion of total
news exposure from which purposeful news seeking
is excluded.
Thus, when digital technology liberated people
from standardized patterns of news consumption
(Negroponte, 1995), many scholars foresaw detrimen-
tal consequences to democracy. Some predicted
increased partisan polarization of the public (Iyengar
&Hahn,2009;Stroud,2010); others warned of
avoidance of public affairs information and decline
of political engagement as a whole (Prior, 2007;
Sunstein, 2003). Both arguments are based on deeply
rooted distrust of the public’scompetencethatwith-
out incidental exposure, a balanced diet of informa-
tion acquisition is lost.
As a theoretical explanation why the digital media
block incidental news exposure, Prior (2007)pro-
posed the relative entertainment use hypothesis.
That is, when people have more choices in the selec-
tion of media content, they will abandon news for
entertainment. Prior’s theoretical framework com-
bines Downs’concept of information cost (1957)
and information gain model (Delli Carpini &
Keeter, 1996; Luskin, 1990), which posits that
increased choice will create a condition that
increases the opportunity cost of news consumption.
According to these theoretical models, increased
information cost will interact with people’s motiva-
tion to learn, and dampen their knowledge gain on
public affairs. All in all, the implication is that the
new communication environment of social media
and digital technology to civic decline and to
a systematic abandonment of political engagement.
Digital media and political participation
However, the current state of the literature is that
the relationship of digital media use with political
participation is positive overall, although the cau-
sal direction of such relationship is not firmly
established (Bakker & de Vreese, 2011,
Boulianne, 2009,2015). In a meta-analysis of 38
studies, Boulianne (2009) found positive effects of
the Internet use on participation, while some stu-
dies found a stronger association with the use of
the Internet for news as an independent variable
or without the control of variables like political
interest. In a more recent meta-analysis,
Boulianne (2015) found social media use is posi-
tively associated with political and civic participa-
tion in 80% of 36 studies examined. After
reviewing related literature, Stromer-Galley and
Wichowski (2011) concluded, “the Internet, as
a channeler-of-channels, offer a number of char-
acteristics that invite the possibility of increased
participation generally in the U.S.”(p. 170).
Meanwhile, studies indicate that citizen’s self-
filtering behavior of information is more complex
2S. W. YOO AND H. GIL DE ZÚÑIGA
than previously conceived. People may seek infor-
mation agreeable to them online, but that does not
mean they shut out information they disagree
(Garrett, 2009; Garrett, Carnahan, & Lynch, 2011;
Webster, 2011). Users of partisan online media are
occasionally exposed to general news of main-
stream media that they did not seek (Trilling &
Schoenbach, 2014). While social media use has
been often characterized by selective news expo-
sure, recent studies show that as much incidental
news exposure occurs on social media, leading to
more news use, or increased political knowledge
(Feezell, 2018; Fletcher & Nielsen, 2018; Lee &
Kim, 2017: Oeldorf-Hirsch, 2018). The continued
chances of incidental news exposure in the digital
media warrant reassessment of its effect on politi-
cal participation.
Homogeneous and heterogeneous political
discussions as mediators
News-use facilitates political participation in
a number of ways: providing mobilizing information
(Lemert, 1984) and increasing the level of knowledge
(Delli Carpini & Keeter, 1996) and mental elaboration
(Eveland, 2004, Shah et al., 2007), as well as energizing
partisan involvement (Dilliplane, 2011). Political dis-
cussion among citizens has been deemed as the soul of
democracy (De Tocqueville, 1863) and decades of
scholarly work has shown its importance in the poli-
tical process. In a dense communication environment
such as during campaign cycles, various forms of
political discussion work in tandem with news use
(Beck, 1991). Especially, digital media not only func-
tion as sources of information but also as platforms of
conversation.Therefore,increasingattentionhas
been drawn to the role of political discussion in the
relationship between news use and political participa-
tion (Bae et al., 2013;Vragaetal.,2015).
However, the relationship among news use, dis-
cussion and participation are not always clear.
From the perspective of the differential gains
model, scholars argue the level of political discus-
sion conditions the relationship of news use with
political participation, in such a way that the effect
of news use on political participation are stronger
among those who talk frequently about politics
than who talks less often (Hardy & Scheufele,
2005; Scheufele, 2002). Thus, some studies
examined individual’s frequency of political dis-
cussion as a moderator of the digital media use
effects (e.g., Hyun & Kim, 2015).
On the other hand, effects of news use have
long been considered to occur indirectly, political
discussion being one of the key mediators inter-
vening the outcome.
1
In the tradition of two-step
flow theory (Lazarsfeld, Berelsen, & Gaudet, 1948;
Katz & Lazarsfeld, 1955), interpersonal political
discussion is the central determinant of political
behavior, and news use work through opinion
leaders and their discussion network.
Subsequently, findings of the communication
mediation model (McLeod, Scheufele, & Moy,
1999; Sotirovic & McLeod, 2001)confirmedthe
role of political discussion as a mediator between
news use and participation. In various settings,
described as the citizen communication media-
tion model (Shah, Cho, Eveland, & Kwak, 2005)
or the campaign mediation model (Shah et al.,
2007), news consumption online and offline
encouraged political discussion, which created
channeled paths to participation. Although
many studies relied on cross-sectional data,
some studies (e.g. Shah et al., 2005) used long-
itudinal data to claim causalities.
The mediating role of discussion is particularly
essentialifincidentalnewsexposureweretohave
democratic outcomes. Motivation of an individual
influences the pattern of news reception and its
outcomes (Price & Zaller, 1993). Passive news expo-
sure is less likely to lead to political participation
than active information seeking. When it does, it is
probable that a person goes through additional
processing of the information acquired, and politi-
cal discussion functions as a cognitive elaboration
process (Yoo, Kim, & Gil de Zúñiga, 2017 ;
Eveland, 2004;Eveland,Morey,&Hutchens,
2011). In this context, it is revealing that studies
found late-night comedy viewing, a form of passive
news exposure like incidental news exposure online,
led to increased political participation with the
mediation of political discussion (Lee, 2012;Moy,
Xenos, & Hess, 2005).
Based on the reasoning that accidental encoun-
ter with news fosters political talk, which in turn
contributes to increase political participation, this
study predicts significant and positive mediation
of political discussion in the process of incidental
JOURNAL OF INFORMATION TECHNOLOGY & POLITICS 3
news exposure effects. To further contribute to the
literature, we differentiate outcome and mediating
variables adapting to the new practice of commu-
nication. As for outcome variable, this study tests
each path to offline and online political participa-
tion separately. Prior research suggests that the
input of time, money and skills required in offline
participation are less important for online partici-
pation (Best & Krueger, 2005; Quintelier &
Vissers, 2008). Given that incidental news expo-
sure effects is expected to be cost-sensitive, it is
probable relationships with online and offline par-
ticipation are different. Also, the ease of discussion
(in other words, the lowered cost) may enhance
the information gain that the previous high cost of
exposure had suppressed. Thus, previous studies
found that the relationship of social media use
with online political participation is different
from the relationship with offline political partici-
pation (e.g., Kim, Russo, & Amna, 2017; Yoo &
Gil-de-Zúñiga, 2014).
Next, to investigate what types of discussion is
conducive to democratic outcomes online, this
study tests mediation of heterogeneous and
homogeneous political discussion separately.
The literature shows scholars have developed
two types of measurement for discussion network
heterogeneity. The first type focused on political
disagreement. Variables were created by asking
how often the respondents talk with partners
having different political views or favoring differ-
ent candidates (e.g., Huckfeldt & Sprague, 1995;
Jang, 2009; Lee, Kwak, & Campbell, 2015;Mutz,
2002; Wojcieszak & Price, 2012). The second type
is total-scale-of-discussion-network heterogene-
ity, which combines demographic items (race,
social and economic status, gender) and network
ties (friends, family) with congruity of political
views (e.g., Brundidge, 2010; McLeod, Sotirovic,
&Holbert,1998;Scheufele,Nisbet,Brossard,&
Nisbet, 2004). We opted for the use of the latter
conceptualization, since the distinctive character-
istics of the digital media are found in demo-
graphic, socioeconomic and political aspects.
Early works on the role of face-to-face discussion
were built on the assumption that the sphere of
influence is confined to the homogeneous discus-
sion network (e.g., Katz & Lazarsfeld, 1955). In
the digital environment, it is much easier to
engage in conversation with people outside nor-
mal discussion network. Using Twitter, Facebook,
Reddit or LinkedIn means you are often exposed
to, or exchange views with people of different
race, gender, socioeconomic status as well as dif-
ferent political views. Digital communication
easily transcend strong-tie relationships such as
family and close friends. And the proportion of
weak-tie relationship on one’snetworkisfound
to increases the chance of incidental news expo-
sureonline(Lee&Kim,2017). Previous studies
confirm that homogenous and heterogeneous
political discussion are not mutually exclusive,
and may occur simultaneously as well as sequen-
tially to have independent outcomes (Eveland &
Hively, 2009; Lee, 2012). Thus, this study each
examined the role of homogeneous political dis-
cussion and the role of heterogeneous political
discussion as a mediator in the incidental news
exposure context.
First, when people accidentally encounter
public affair information, they may choose to
talk about it with like-minded groups, or
trusted family or friends to evaluate informa-
tion on a supportive network before starting
participatory activities. A number of studies
(Eveland & Hively, 2009;Mutz,2002)arein
support of the expectation that the harmony
of homogeneous discussion encourages, while
the ambivalence of heterogeneous discussion
discourages, participation. Also, people may
take this path because accidentally acquired
information can be perceived as less trust-
worthy than their regular source of informa-
tion. Thus, the following hypotheses.
H1a: Homogenous political discussion positively
mediates the relationship between incidental
news exposure on the Internet and offline political
participation.
H1b: Homogenous political discussion positively
mediates the relationship between incidental news
exposure on the Internet and online political
participation.
Next, new information online may open up
a conversation outside the primary discussion net-
work, or people with opposing political views. We
4S. W. YOO AND H. GIL DE ZÚÑIGA
argue digital communication fosters this mediation
path. It may be true that from the receptive view of
communication, people tend to selectively expose
themselves to information that fit their existing opi-
nions (Chaffee and Metzger, 2001). However, from
theexpressiveview,digitaltechnology enables indivi-
dualstoreachouttootherswithlesseffortthanwith
efforts needed for face-to-face contact. On the
Internet, a motivated person easily seeks opportunity
to persuade others with a different opinion, or get to
know new people based on new information.
As opposed to the argument that heterogeneous
political discussion depresses political participa-
tion, a substantial number of studies advocate
such positive role of heterogeneous political dis-
cussion encouraging political participation. Studies
in this line of argument posit that interactions with
people outside their usual discussion network give
rise to the mental upscaling process necessary for
political engagement (Guidetti, Cavazza, &
Graziani, 2016; Kwak, Williams, Wang, & Lee,
2005; Leighley, 1990; Scheuefele, Nisbet, Brossard,
& Nisbet, 2004). Unsurprisingly, these studies pro-
pose more complex models of multivariate rela-
tionship in which heterogeneous political
discussion operates as mediator (e.g., Scheufele
et al., 2004) or moderator (e.g., Kim, Scheufele, &
Han, 2011). For example, Guidetti and colleagues
(2016) described a process wherein heterogeneous
political discussion increases political interest and
knowledge, and in turn, political participation (see
p.237).
One of the reasons for predicting negative relation-
ship of heterogeneous political discussion with parti-
cipation is because of social accountability, that is, fear
of isolation as result of expressing opinions outside
the norms of the reference group (Mutz, 2006).
However, in online discussion network, the pressure
to conform to reference group may not be as strong as
in interpersonal communication network (Kim et al.,
2011).Thus,wepredictthefollowing:
H2a: Heterogeneous political discussion positively
mediates the relationship between incidental news
exposure on the Internet and offline political
participation.
H2b: Heterogeneous political discussion positively
mediates the relationship between Incidental news
exposure on the Internet and online political
participation.
Impact of partisanship: conditioning the
mediation process
Although social change has reshaped the nature of
party identification in modern democracies
(Dalton, 2007), the strength of partisanship
remains a central predictor of a wide range of
political behaviors (Converse & Pierce, 1992).
Communicative activities such as news consump-
tion, information processing, and information giv-
ing have been found highly partisan in early
studies of political communication (e.g., Berelson,
Lazarsfeld, & McPhee, 1954).
Partisanship is no less relevant in digital commu-
nication. Echo chamber, filter bubble, and other
arguments of cyber-pessimism are intertwined
with individual levels of partisanship. In these argu-
ments, partisanship has deemed a suppressant of
information flow, accelerating the homogeneity of
networks. Selective-exposure research tests parti-
sanship as an information filter that fuels further
inclination to partisanship (Iyengar & Hahn, 2009;
Stroud, 2010). Brundidge examined moderation of
partisanship in the relationship between online dis-
cussion and network heterogeneity, and maintained
that partisanship contributes to “rigid communica-
tion”that sorts partisans into “ideological enclave-
[s]”(p. 695). Studies of interpersonal discussion are
in support of this view. Huckfeldt and Sprague
(1995) observed that Reagan supporters were
more likely to discuss politics with Reagan suppor-
ters in 1984 presidential elections. Thus, partisan
political discussion was considered “protective
cocoons for an individual’s political preferences”
(Beck, 1991, p. 379).
However, partisans may be restrictive in what
news they seek, but more active than nonpartisans
in discussing the news as they stumble upon it.
Two aspects of partisanship revealed in the litera-
ture lend support to this argument. First, partisans
have a cognitive base that allows them to assign
meanings to information they encounter
(Valentino, Beckman, & Buhr, 2001). Second,
because partisans are more engaged in overall
political activities (Bartels, 2000), they are also
likely to exchange ideas, often to persuade others
JOURNAL OF INFORMATION TECHNOLOGY & POLITICS 5
about information encountered. A recent finding
showed that on social media, partisans are more
likely to talk to others (disseminate) about news
than nonpartisans (Weeks & Holbert, 2013).
Partisanship has long been conceived as
a“perceptual screen”(Campbell, Converse, Miller,
& Stokes, 1960, p. 133) that shapes the effect of
political information; that plays the role of cue-
giving or cost-saving device (Borre & Katz, 1973).
This view conceptualized partisanship as inherent in
the character of an individual. It might be stable over
a lifetime or even inherited in the form of psycholo-
gical attachment (Campbell et al., 1960;Miller,1991)
or social identity (Green, Palmquist & Schickler,
2002). However, with the cost of discussion and
participatory activities drastically lowered, the influ-
ence of partisanship may be better interpreted as
a utilitarian tool to realize personal or group interest
(Achen, 1992; Fiorina, 1990; Page & Jones, 1979).
This conceptualization will lend support to the pre-
diction that partisans will actively seek to talk to
people outside of their ordinary discussion network.
Thus, this study hypothesizes partisanship
strengthen not only the role of homogenous discus-
sion but also that of heterogeneous discussion:
H3a: The strength of partisanship moderates the
mediated relationship between incidental news expo-
sure and political participation with homogeneous
political discussion, such that the mediated relation-
ship is stronger for partisans than non-partisans.
H3b: The strength of partisanship moderates the
mediated relationship between incidental news expo-
sure and political participation with heterogeneous
political discussion, such that the mediated relation-
ship is stronger for partisans than non-partisans.
Methods
Data
To test the proposed hypotheses, a national online
survey was administered by the XXX Research Unit at
the [author’s university] between 15 December 2013
and 05 January 2014. Respondents were selected from
those registered to participate in an online panel
administered by the media-polling group AS
Nielsen. In order to assure national representative-
ness, Nielsen employed a stratified quota sampling
to recruit respondents from among 200,000 people.
A quota was established so that the demographic
distributions of the data would match as closely as
possible, that reported by the U.S. Census. From an
initial sample of 5,000 individuals, there were 2,060
respondents; later, 247 responses were deleted due to
incomplete or invalid data. Employing the American
Association of Public Opinion Research response-rate
calculator(RR3),theresponseratewas34.6%
(American Association of Public Opinion Research,
2008,p.45).
This sample had slightly younger, more edu-
cated respondents and fewer Hispanics. However,
the overall sample was comparable to other sur-
veys (Pew Research Center for the People and the
Press, 2008) and to the national population as
whole (For variable to variable comparison with
U.S. Census, see Authors, in press).
Measures
Incidental news exposure online
This variable measures the level of unintended
news encounters on seven types of web services.
Respondents were asked how often they encoun-
ter news about current events, public issues, or
politics while using the following devices for
a purpose other than news: search engines,
online portals, email, blogs, social networking
sites, microblogging sites, and mobile devices.
The 10-point scale included choices ranging
from never to all the time (7 items;
α= .86, M = 3.45, SD = 1.9). Self-reported
measurement of incidental exposure has been
used previously (Ardèvol-Abreu, Diehl, & Gil
de Zúñiga, 2017;Brundidge,2010;Lee&Kim,
2017; Pew Research Center, 1998, 1996). This
study, although drawing in these prior works,
expanded the items to better capture the current
digital environment.
Offline political participation
While strands of the literature treat political
discussion as a form of participatory political
activities (e.g., Ekman& Amnå, 2012), the pre-
sent study defines political participation as beha-
viors designed to directly affect the choice of
6S. W. YOO AND H. GIL DE ZÚÑIGA
government personnel and policies (Verba,
Schlozman, & Brady, 1995). Thus, respondents
were asked how often they had engaged in the
following seven activities in the past 12 months:
“Attended a political rally,’’ ‘‘Participated in any
demonstrations, protests, or marches,”“Donated
money to a campaign or political cause,’’
“Participated in groups that took any local
actionforsocialorpoliticalreform,’’ ‘‘Been
involved in public interest groups, political
action groups, political clubs, political cam-
paigns,”and how often they had voted for
“local and statewide elections”,“federal and pre-
sidential elections.”Responses were recorded on
a 10-point scale with choices ranging from
“never”to “all the time.”An index was created
by averaging scores across measures (7 items;
α=.79,M=3.6,SD=1.56).
Online political participation
This variable measured how often the respondents
engaged in the following activities online, and on
social media in the past 12 months. The items
were: “Participated in an online question and
answer session with a politician or public official,”
“Created an online petition,”“Signed up online to
volunteer to help with a political cause,”“Joined
a political or cause-related group on a social media
site,”“Started a political or cause-related group on
a social media site.”An index was created by
averaging scores from the 10-point scale (5 items;
α= .88, M = 1.65, SD = 1.41).
Heterogeneous political discussion
The variable was constructed by averaging four
items on a 10-point scale based on how often
respondents talked about politics or public affairs
online and offline with: (a) people who disagree,
(b) people with different political views, (c) people
from different race or ethnicity, and (d) people
from different social class (4 items;
α= .93, M = 3.59, SD = 2.45). Similarly, the
variable homogenous political discussion is a mean
of four items on a 10-point scale based on discus-
sions with: (a) people who agree, (b) people with
similar political views, (c) family and relatives, (d)
friends (4 items; α= .90, M = 4.58, SD = 2.52).
Strength of partisanship
Respondents were asked to rate their party identi-
fication using an 11-point scale ranging from
strong Republican (coded ‘0ʹ; 6.4% of respondents)
to strong Democrat (coded ‘10ʹ; 9.9% of respon-
dents), with the midpoint (coded ‘5ʹ) being
Independent (34.3% of respondents). This item
was folded into a 6-point scale ranging from no
partisanship to strong partisanship (M= 3.07,
SD = 1.42).
Political knowledge. This variable was created
by adding scores of multiple-choice questions
measuring the knowledge of respondents regard-
ing civic affairs and current issues and used as
a control variable. For civic knowledge, ques-
tions included, “What job or political office
does Joe Biden hold?”“For how many years
are there in one full term of office for a U.S.
Senator?”“What job or political office does John
Roberts hold?”“On which of the following does
the U.S. government currently spend the least?”
For current issues knowledge, questions
included, “Do you know which party introduced
theimmigrationbillbeforeCongress?”“Do you
knowtherulingoftheSupremeCourtabout
Obamacare was?”“Which organization’sdocu-
ments were released by Edward Snowden?”
“Recently, the UN and US were in negotiations
with the Syrian government over the removal
of?”The number of correct answers were
summed up to construct a variable (8 items;
α=.75,M=4.44,SD=2.19).
Internal political efficacy
This control variable tapped people’s self-confidence
in their ability to engage in politics. The items were:
“People like me can influence the government,”“I
consider myself qualified to participate in politics,”
“I have a good understanding of the important poli-
tical issues facing our country.”(3 items; 10-point
scale, α=.79,M=5.16,SD = 2.25).
Political interest
This control variable measured respondents’inter-
est and attention in information regarding what is
going on in politics and public affairs (2 items; 10-
point scale, r = .93, M = 6.66, SD = 2.71).
JOURNAL OF INFORMATION TECHNOLOGY & POLITICS 7
News media use
In this study, to control purposive news exposure in
the analysis, a composite scale of news-media-use
from 11 sources ranging from traditional to digital
outlets was created. Respondents were asked how
often they used media to get information about
current events. The media included network TV
news, local TV news, national newspapers, local
newspapers, online news sites (e.g., Gawker,
Politico), cable TV news, radio news, social media,
tablet applications or browsers, smartphone appli-
cations or browsers, and news aggregators (10-
point scale, α= .70, M = 4.03, SD = 1.43).
Socio-demographic variables
Five variables were used as control variables. Age
was measured with an open-ended question
(M = 50.71, SD = 14.77). Income was measured
with an 8-point scale that tapped family’stotal
household income in the previous year, ranging
from “under $10,000”to “over $200,000”
(Mdn = $50,000 to $99,999). Education was mea-
sured with an 8-point scale ranging from “less than
high school”to “doctoral degree”(Mdn = some
college education). The respondent’sgender(50%
females), and race (76.2% whites) were also
included in the model as controls. Overall, respon-
dents were slightly older, more educated and
included fewer Hispanics than the U.S. population
(see Appendix 1). However, the sample demo-
graphics were comparable to other surveys employ-
ing random collection methods (Pew Research
Center for the People and the Press, 2008).
Results
This study first employed two hierarchical ordin-
ary least squares (OLS) regressions predicting off-
line and online political participation. The
variables were entered in separate blocks: demo-
graphics, political orientation, media use (Model
1), and last, discussion variables (Model 2). Table 1
shows the results. Incidental news exposure online
was a predictor of offline political participation
(β= .082, p < .01) in Model 1. In Model 2, it
was non-significant. It was positively associated
with online political participation both in Model
1(β= .15, p < .001) and in Model 2 (β= .12,
p < .001). This result implies stronger relation of
incidental news exposure with online, than with
offline participation. News media use, which taps
purposeful use of online and offline media, was
a much stronger predictor of political participation
than incidental news exposure online, and
remained significant after controlling for political
discussion variables (offline participation model 2,
β= 0.16, p < .001; online participation model 2,
β= .17, p < .001).
Homogenous political discussion did not predict
political participation. In contrast, heterogeneous
political discussion was positively associated with
both offline (β= 0.13, p < .001) and online participa-
tion (β= .13, p < .001). The socio-demographic vari-
ables accounted for 9.9% of the variance in the offline-
participation model and 4.3% in the online-
participation model, while the discussion variables
accounted for 2.7% in the offline-participation
model, and 3.3% for the online-participation model.
Political orientation mattered more for offline
Table 1. OLS hierarchical regression predicting offline and
online political participation.
Offline participation
Online
participation
Model 1 Model 2 Model 1 Model 2
Block 1 –
Demographics
Age .007* .007* −.014*** −.014***
Gender (male) .083 .06 .324*** .297***
Education .098** .092** .008 .002
Race (white) .173 .171 −.080 −.075
Income .052 .056 −.063* −.058*
ΔR
2
9.9% 9.9% 4.3% 4.3%
Block 2 –Political
Orientation
Political Efficacy .188*** .167*** .132*** .111**
Strength of
Partisanship
.145*** .148*** .069* .075**
Political Knowledge .01 .007 −.096 −.099
Political Interest .093*** .054* .026* .009
ΔR
2
24.4% 24.4% 12.4% 12.4%
Block 3 –Media
News Media Use .169*** .156*** .186*** .174***
Incidental News
Exposure
.082** .052 .146*** .117***
ΔR
2
4.1% 4.1% 10.3% 10.3%
Block 4 –Mediation
Homogenous
Political
Discussion
–.016 –−.016
Heterogeneous
Political
Discussion
–.126*** –.133***
ΔR
2
–2.7% –3.3%
Total R
2
38.7% 41.4% 27% 30.3%
Note. Entries are standardized final regression coefficients
* = p < .05, ** = p < .01, *** = p < .001
8S. W. YOO AND H. GIL DE ZÚÑIGA
participation (ΔR2 = 24.4%) than online participation
(ΔR2 = 12.4%). The overall predictive values were
larger for offline participation (R2 = 41.4%) than
online participation (R2 = 30.3%).
Next, for a more precise examination of the
mediating role of homogenous (H1a, b) and het-
erogeneous political discussion (H2a, b), a parallel
multiple-mediation model using SPSS macro
PROCESS Model 4 as in Hayes (2013) was
employed. A statistical diagram of this model is
illustrated in Figure 1.
Bootstrapping method is found to be more power-
fulthantheSobeltestandthecausalstepapproachof
Barron and Kenny to test mediating effects (Hayes,
2009; Williams & MacKinnon, 2008). Also, this model
allows for a statistical comparison between the indir-
ect effects of different mediators (see Hayes, 2013,
pp.124–125) and is more parsimonious than
a sequential multiple-mediation model (Model 6) in
Hayes (2013).
The results of mediation analysis are presented
in Table 2. Two paths of indirect effects are com-
pared: (1) Incidental news exposure online –
homogeneous political discussion –offline/online
political participation, (2) Incidental news expo-
sure online –heterogeneous political discussion –
offline/online political participation.
2
The role of
homogeneous discussion as mediator between
incidental exposure and offline and online partici-
pation was not significant. H1a, H1b are rejected.
To the contrary, the second path was significant
for both offline (β= .029, 95% CI = 0.013–056)
and online participation (β= .032, 95%
CI = .0011 –.022). H2a, H2b are supported.
The total effect of incidental exposure on online
participation (β= .15, 95% CI = .94 –.2) was
much larger than offline participation (β= .082,
95% CI = .028–.14). The direct effect of incidental
news exposure online on offline participation was
insignificant. This means heterogeneous political
discussion fully mediates the relationship of inci-
dental news exposure with offline participation,
while partially mediating the relationship with
online participation.
3
To test H3a and H3b, which posit that the
strength of partisanship moderates the mediation
Incidental
Exposure
Incidental
Exposure
Homogeneous
Discussion
Homogeneous
Discussion
Heterogeneous
Discussion
Online
Participation
0.21*** 0.005
0.24*** 0.12***
0.052 ← (0.082**)
0.12***← (0.15***)
Heterogeneous
Discussion
0.21***
0.24***
-0.016
0.13***
Offline
Participation
Figure 1. A parallel multiple mediator model between Incidental news exposure, Homogenous political discussion, Heterogeneous
political discussion, Offline, and Online political participation.
Note. Estimates are calculated using SPSS macro PROCESS (Hayes, 2013) Model 4 applying 5,000 bootstrapped bias corrected
resample is. The number in parenthesis is the total effect without adding the mediators. * = p < .05, ** = p < .01, *** = p < .001
JOURNAL OF INFORMATION TECHNOLOGY & POLITICS 9
of political discussion, this study undertook
a conditional process analysis using Hayes (2013)
PROCESS Model 59. This model was selected
because partisanship moderates each path of indir-
ect and direct effects.
4
Figure 2 shows a conceptual
and statistical diagram of this model.
Table 3 presents the estimates, standard errors
and significance value of moderated mediation.
The effects were estimated at values of one stan-
dard deviation below (low) and above (high) the
sample mean of partisanship.
We can speak of moderated mediation when the
strength of an indirect effect depends on the level of
moderators (Preacher & Hayes, 2007). For homo-
geneous discussion, moderated mediation was not
observed. The indirect effects were not significant
regardless of the level of partisanship. H3a is not
supported. For heterogeneous discussion, an indir-
ect effect significantly increased as the level of par-
tisanship increased. It was stronger for strong
partisans (offline β= .037, 95% CI = .014–.072;
online β= .042, 95% CI = .017–.081), than average
(offline β= .027, 95% CI = 0.12–053; online β=.03,
95% CI = .013–.05). The indirect effect of non-
partisans was not significant. This implies that par-
tisans tend to derive more participatory activities
from incidental news exposure by talking to people
with different political views, or outsiders. The
Table 2. Mediation of political discussion between incidental
news exposure and political participation.
Offline participation Online participation
Effects βSE 95% CI βSE 95% CI
Total .082 .0275 .028–.14 .15 .026 .094–.2
Direct .052 .027 −.0014–.11 .12 .026 .066–.17
Indirect .03 .0081 .016–.048 .028 .0078 .013–.058
IE→HO→OP .001 .007 −.012–.016 −.0034 .0072 −.018–.011
IE→HE→OP .029 .01 .013–.056 .032 .011 .0011–.022
Notes 1. SPSS macro PROCESS (Hayes, 2013) Model 4 applying 5,000
bootstrapped bias corrected resample is used.
2. IE = Incidental news exposure online, HO = Homogeneous political
discussion, HE = Heterogeneous political discussion, OP = Offline
political participation, online political participation.
3. Statistical controls: age, gender, education, race, income, internal
political efficacy, strength of party identification, political knowledge,
political interest, news media use.
Incidental
Exposure
Partisanship
Political
Participation
Political
Discussion (M)
Incidental
Exposure (X)
Political
Participation
Partisanship (W)
X * W M * W
Political
Discussion
(a) Conceptual diagram
(b) Statistical diagram
Figure 2. Conceptual and statistical diagram and statistical model of a moderated mediation model of incidental news exposure,
political discussion and partisanship on political participation. PROCESS Model 59 described in Hayes (2013).
10 S. W. YOO AND H. GIL DE ZÚÑIGA
difference between the effects is statistically signifi-
cant. H3b is supported.
Discussion
Cyber-pessimists have long conceived that
empowered, and yet politically less engaged audi-
ence will emerge, as they no longer encounter
news by chance. The present study finds tangible
support, however, for somewhat hopeful prospect
of the relationship between digital media use and
political engagement. Using a measurement that
taps various ways people stumble upon public
affairs news online, and controlling for purposeful
news seeking on a wide range of media, incidental
news exposure is shown to be positively associated
with offline and online political participation.
The result of mediation test offers a key to under-
standing this relationship. Heterogeneous political
discussion exerts a full mediating role in the relation-
ship of incidental exposure with offline participation.
The finding supports the argument that political
talks offline and on digital media enhanced the effect
of passive news exposure. A significant portion of the
relationship with online participation is also
mediated by heterogeneous political discussion.
The difference between the results of offline and
online participation reflects the cost of activities.
The mediating effects of discussion over partici-
pation are not new. Yet, the fact that this mechanism
takes place between incidental exposure and hetero-
geneous discussion is one of the main contributions
this study has made to this strand of the literature.
What is shown here is that accidental encounter with
news is conducive to both heterogeneous discussion
and homogeneous discussion. However, it is hetero-
geneous discussion that is linked with participatory
outcomes. This mediating process can be interpreted
as citizens engaging in a deliberative process of
speaking with diverse others. Alternatively, this
could also be action by politically motivated people
to persuade people of the other side. In either case,
we consider it promising for democracy. Diversity in
the communication network makes the public less
susceptible to cognitive errors and biases (Bohman,
2007; Mill, 1989), an aspect that is certainly beneficial
for passive learning of information.
The strength of partisanship, the usual suspect
curbing the effectiveness of the communication
environment, is found to be a facilitator of the med-
iation process. Strong partisans are more likely to
engage in cross-cutting conversation, which in turn
relates to strengthened political activity. This result is
comparable to the previous findings that attitude
certainty (Matthes, 2012), and ideological strength
(Wojcieszak, Baek, & Delli Carpini, 2010)condition
the relationship between disagreement discussion
and political engagement. Also comparable are find-
ings that suggest it is not partisanship, but congruity
with the opinion environment that suppresses cross-
cutting conversation (e.g., Noelle-Neumann, 1993).
More importantly, this result calls for
a reassessment of the role of partisanship in
media effects. A long-standing approach of dosage
(exposure) and resistance (pre-existing attitude)
considers partisanship as repressing media effects,
because partisans tend to have crystallized political
attitudes (Converse, 1964; Zaller, 1992). Agenda
setting, priming, and framing theory have all
assumed partisanship mitigates media effects
(Hoffman & Young, 2011; lyengar & Kinder,
1987: McCombs & Shaw, 1972). However, with
the advancement of expressive communication
technology in digital media, speaking to others
Table 3. Moderated mediation results for political participation through homogeneous and heterogeneous political discussion.
Offline participation Online participation
Mediator Partisanship Indirect effects SE 95% CI Indirect effects SE 95% CI
Homo-geneous
Discussion
1.67 .015 .012 −.004–.041 .083 .011 −.012–.04
3.1 .003 .007 −.009–.019 .001 .007 −.014–.014
4.53 −.006 .008 −.022–.009 −.009 .009 −.032–.004
Hetero-geneous
Discussion
1.67 .018 .013 −.001–.051 .017 .014 −.005–.051
3.1 .027 .01 .012–.053 .03 .013 .013–.057
4.53 .037 .015 .014–.072 .042 .017 .017–.081
Notes 1. SPSS macro PROCESS (Hayes, 2013) Model 59 applying 5,000 bootstrapped bias corrected resample is used.
2. CI = Confidence Interval
3. Statistical controls: age, gender, education, race, income, internal political efficacy, strength of party identification, political knowledge, news media use.
JOURNAL OF INFORMATION TECHNOLOGY & POLITICS 11
has become as important as listening is for com-
munication, to support democratic outcomes.
Thus, as we go beyond the unidirectional sender-
receiver paradigm, to a paradigm of networked
news flow where the distinction of sender and
receiver is blurred, strong partisanship indicates a
willingness to communicate across boundaries (a
facilitating trait of a deliberative process).
Conclusively, this finding provides grounds for
the strengthening of political parties in
democracies.
There remain a number of important limita-
tions to be addressed. The first involves the self-
reported measure of incidental exposure.
Although this study significantly expanded the
constructs as compared to items in previous stu-
dies (Pew Research Center, 2008, 1998,
Tewksbury et al., 2001), authors note that this
measure cannot capture much of the acquisition
of incidental information occurs unconsciously.
Despite its drawbacks, this measurement strategy
is used to analyze multivariate correlates. An
experimental design that makes it possible to
control the amount of exposure should help clar-
ify this predicament. Second, the measurement of
political discussion variables needs to be
improved. The simple binary distinction of
agree/disagree is relevant for well-debated issues,
but less so for less prominent issues, because the
attitudes of the less involved people largely
remain in the gray area (Delli Carpini & Keeter,
1996). Also in the online environment, indivi-
duals may not recognize political discussion
even when they engage in it (Walsh, 2004).
Although this study included many dimensions
to generate the most valid measurement of het-
erogeneous discussion, further research would
serve well to reflect the complex nature of what
it means for people to maintain diverse discus-
sions network. Third, the cross-sectional nature
of our data obliges us to interpret the findings
with due caution, as there might be alternative
causal orders to our interpretation. Also, studies
indicate that a mediation process requires certain
time to unfold, which makes any hasty conclusion
of the result problematic (Preacher, 2015). The
results of this study should be interpreted with
caution. The result does not show that those who
use digital media accidentally acquire free infor-
mation gain.
All in all, this article makes an important contri-
bution to the query about the effect of media on the
public’s capability to political engagement.
Overcoming the usual framework of effects vs. non-
effects, this study provided the conditions and paths
by which today’s incidental exposure to information
online, may contribute to political engagement. In
addition, this study reveals the necessity of
a paradigm that takes into account the full function-
ality of digital communication. As digital media
evolve, the weakening boundaries of one form of
communication (news exposure) seem to be condu-
cive to weakening the boundaries of another form of
communication (heterogeneous political discussion
networks). In the earlier paradigm focused on news-
reception, the increased choice for people envisioned
the paradoxical result of a detriment to democracy.
In the new paradigm of digital communication, it is
part of a seamless cycle, opening up the possibility of
deliberative exchange of thoughts.
Notes
1. Whereas a moderator is defined as the third variable
that affects the direction and/or strength of the rela-
tionship between two variables, a mediator is the
variable that intervenes, and is a part of the causal
sequence (Barron & Kenny, 1986; Hayes, 2013). In
the study of news effects, mediation analysis is an
attempt to answer the question, “How does this
occur?”(Hayes, 2013, p. 7).
2. This study also tested the serial multiple-mediation
model (Model 6) that Hayes (2013) recommends
when mediators are highly correlated. The result
was the same in that the indirect effect of the path
of incidental news exposure online –homogeneous
discussion –participation is not significant.
3. Hayes (2013) argued the distinction between full
(complete) and partial mediation should be aban-
doned (pp. 170–172), but this study considers it
meaningful as a point of comparison between effects.
4. This model was chosen over other PROCESS models
such as Model 58 which assumes moderations of
indirect paths (X→M, M →Y), not direct path (X
→Y), or Model 14 which only assumes moderation
of X→M based on separate regression analyses. In
the analyses, the interaction term of partisanship and
incidental news exposure was a significant predictor
of offline participation, online participation, homo-
geneous discussion, and heterogeneous discussion.
12 S. W. YOO AND H. GIL DE ZÚÑIGA
Notes on contributors
Sung Woo Yoo (PhD, The University of Texas at Austin) is
an assistant professor in the Department of Communication
and Media Studies at SUNY Cortland. A former journalist
working for Hankook Ilbo in South Korea, his research cen-
ters around the effect of digital media use on political engage-
ment, sophistication and nationalism.
Homero Gil de Zúñiga (PhD, Universidad Europea de
Madrid; PhD, University of Wisconsin –Madison) holds
the Medienwandel Professorship at University of Vienna,
where he directs the Media Innovation Lab (MiLab). He
also serves or has served as Research Fellow at Universidad
Diego Portales, Chile, at the Center for Information
Technology Policy at Princeton University, and the Nieman
Journalism Lab at Harvard University. His research addresses
the influence of new technologies and digital media over
people’s daily lives, as well as the effect of such use on the
overall democratic process.
ORCID
Homero Gil De Zúñiga http://orcid.org/0000-0002-4187-
3604
References
Achen, C. H. (1992). Social psychology, demographic vari-
ables, and linear regression: Breaking the iron triangle in
voting research. Political Behavior,14(3), 195–211.
doi:10.1007/BF00991978
American Association of Public Opinion Research. (2008).
Standard definitions: Final dispositions of case codes and out-
come rates for surveys. Retrieved July 8, 2009, from http://www.
aapor.org/uploads/Standard_Definitions_04_08_Final.pdf.
Ardèvol-Abreu, A., Diehl, T., & Gil de Zúñiga, H. (2017).
Antecedents of internal political efficacy incidental news
exposure online and the mediating role of political discus-
sion. Politics,https:// doi.org/10.1177/0263395717693251.
Atkin, C. (1973). Instrumental utilities and information seek-
ing. In P. Clarke (Ed.), New models for mass communica-
tion research (Vol. 2, pp. 205–242). Beverly Hills, CA: Sage.
Bae, S. Y., Kwak, N., & Campbell, S. W. (2013). Who will
cross the border? The transition of political discussion into
the newly emerged venues. Computers in Human Behavior,
29(5), 2081–2089. doi:10.1016/j.chb.2013.04.006
Bakker, T. P., & de Vreese, C. H. (2011). Good news for the
future? Young people, internet use, and political
participation. Communication Research,38(4), 451–470.
doi:10.1177/0093650210381738
Bartels, L. M. (2000). Partisanship and voting behavior, 1952-
1996. American Journal of Political Science,44(1), 35–50.
doi:10.2307/2669291
Baum,M.A.(2002). Sex, lies, and war: How soft news brings
foreign policy to the inattentive public. American Political
Science Review,96(1), 91–109. doi:10.1017/
S0003055402004252
Beck, P. A. (1991). Voters’intermediation environments in
the 1988 presidential contest. Public Opinion Quarterly,55
(3), 371. doi:10.1086/269269
Berelson, B. R., Lazarsfeld, P. F., & McPhee, W. N. (1954).
Voting: A study of opinion formation in a presidential elec-
tion. Chicago, IL: Univ. of Chicago Press.
Best, S. J., & Krueger, B. S. (2005). Analyzing the representa-
tiveness of internet political participation. Political
Behavior,27(2), 183–216. doi:10.1007/s11109-005-3242-y
Bohman, J. (2007). Political communication and the episte-
mic value of diversity: Deliberation and legitimation in
media societies. Communication Theory,17(4), 348–355.
doi:10.1111/comt.2007.17.issue-4
Borre, O., & Katz, D. (1973). Party identification and its
motivational base in a multiparty system: A study of the
Danish general election of 1971. Scandinavian Political
Studies,8(A8), 69–111. doi:10.1111/scps.1973.8.issue-A8
Boulianne, S. (2009). Does Internet use affect engagement? A
meta-analysis of research. Political Communication,26,
193–211. doi:10.1080/10584600902854363
Boulianne, S. (2015). Social media use and participation: A
meta-analysis of current research. Information,
Communication & Society,18(5), 524–538. doi:10.1080/
1369118X.2015.1008542
Brundidge, J. (2010). Encountering “difference”in the contem-
porary public sphere: The contribution of the Internet to the
heterogeneity of political discussion networks. Journal of
Communication,60, 680–700. doi:10.1111/jcom.2010.60.
issue-4
Campbell, A., Converse, P. E., Miller, W. E., & Stokes, D. E.
(1960). The American Voter. New York, NY: Wiley.
Chaffee, S. H., & Metzger, M. J. (2001). The end of mass
communication?. Mass Communication & Society,4(4),
365–379. doi:10.1207/S15327825MCS0404_3
Converse, P. E. (1964). The nature of belief systems in mass
politics. Critical Review,18(1–3), 1–74. doi:10.1080/
08913810608443650
Converse, P. E., & Pierce, R. (1992). Partisanship and the
party system. Political Behavior,14(3), 239–259.
doi:10.1007/BF00991980
Dalton, R. J. (2007). Partisan mobilization, cognitive mobili-
zation and the changing American electorate. Electoral
Studies,26, 274–286. doi:10.1016/j.electstud.2006.04.009
De Tocqueville, A. (1863). Democracy in America.
Cambridge, MA: Sever and Francis.
Delli Carpini, M., & Keeter, S. (1996). What Americans know
about politics and why it matters.NewHaven,CT:Yale
University Press.
Dilliplane, S. (2011). All the news you want to hear: The impact of
partisan news exposure on political participation. Public
Opinion Quarterly,75(2), 287–316. doi:10.1093/poq/nfr006
JOURNAL OF INFORMATION TECHNOLOGY & POLITICS 13
Downs, A. (1957). An economic theory of political action in a
democracy. The Journal of Political Economy,65(2),
135–150. doi:10.1086/257897
Ekman, J., & Amnå, E. (2012). Political participation and
civic engagement: Towards a new typology. Human
Affairs,22(3), 283–300. doi:10.2478/s13374-012-0024-1
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(2), 205–224. doi:10.1111/jcom.2009.59.
issue-2
Eveland, W. P., Morey, A. C., & Hutchens, M. J. (2011).
Beyond deliberation: New directions for the study of infor-
mal political conversation from a communication
perspective. Journal of Communication,61(6), 1082–1103.
doi:10.1111/j.1460-2466.2011.01598.x
Eveland, W. P. (2004). The effect of political discussion in produ-
cing informed citizens: The roles of information, motivation,
and elaboration. Political Communication,21(2), 177–193.
doi:10.1080/10584600490443877
Feezell, J. T. (2018). Agenda setting through social media:
The importance of incidental news exposure and social
filtering in the digital era. Political Research Quarterly,71
(2), 482–494. doi:10.1177/1065912917744895
Fiorina, M. (1990). Information and rationality in elections.
In J. A. Ferejohn & J. H. Kuklinski (Eds.), Information and
democratic processes (pp. 329–342). Urbana, IL: University
of Illinois Press.
Fletcher, R., & Nielsen, R. K. (2018). Are people incidentally
exposed to news on social media? A comparative analysis.
New Media & Society,20(7), 2450–2468. doi:10.1177/
1461444817724170
Garrett,R.K.(2009). Politically motivated reinforcement seeking:
Reframing the selective exposure debate. Journal of
Communication,59(4), 676–699. doi:10.1111/jcom.2009.59.
issue-4
Garrett, R. K., Carnahan, D., & Lynch, E. K. (2011). A turn
toward avoidance? Selective exposure to online political
information, 2004–2008. Political Behavior,35(1),
113–134. doi:10.1007/s11109-011-9185-6
Guidetti, M., Cavazza, N., & Graziani, A. R. (2016). Perceived
disagreement and heterogeneity in social networks:
Distinct effects on political participation. The Journal of
Social Psychology,156(2), 222–242. doi:10.1080/
00224545.2015.1095707
Hardy, B. W., & Scheufele, D. A. (2005). Examining differen-
tial gains from Internet use: Comparing the moderating role
of talk and online interactions. Journal of Communication,
55(1), 71–84. doi:10.1111/j.1460-2466.2005.tb02659.x
Hayes, A. F. (2009). Beyond Baron and Kenny: Statistical
mediation analysis in the new millennium.
Communication Monographs,76(4), 408–420.
doi:10.1080/03637750903310360
Hayes, A. F. (2013). Introduction to mediation, moderation,
and conditional process analysis. New York, NY: Guilford.
Hoffman, L. H., & Young, D. G. (2011). Satire, punch lines,
and the nightly news: Untangling media effects on political
participation. Communication Research Reports,28(2),
159–168. doi:10.1080/08824096.2011.565278
Huckfeldt, R., & Sprague, J. (1995). Citizens, politics, and social
communication: Information and influence in an election cam-
paign. Cambridge, UK: Cambridge University Press.
Hyun, K. D., & Kim, J. (2015). Differential and interactive
influences on political participation by different types of
news activities and political conversation through social
media. Computers in Human Behavior,45, 328–334.
doi:10.1016/j.chb.2014.12.031
I
yengar,S.,&Hahn,K.S.(2009).Redmedia,bluemedia:
Evidence of ideological selectivity in media use. Journal of
Communication,59(1), 19–39. doi:10.1111/jcom.2009.59.
issue-1
Iyengar, S., & Kinder, D. (1987). News that matters: Television
and American opinion. Chicago, IL: University of Chicago
Press.
Jang, S. J. (2009). Are diverse political networks always bad
for participatory democracy? Indifference, alienation, and
political disagreement. American Politics Research,37(5),
879–898. doi:10.1177/1532673X09332790
Katz, E., & Lazarsfeld, P. (1955). Personal influence. The part
played by people in the flow of mass communication.
New York, NY: Transaction Publishers.
Kim, E., Scheufele, D., & Han, J. Y. (2011). Structure or
predisposition? Exploring the interaction effect of discus-
sion orientation and discussion heterogeneity on political
participation. Mass Communication and Society,14(4),
502–526. doi:10.1080/15205436.2010.513469
Kim, Y., Russo, S., & Amnå, E. (2017). The longitudinal relation
between online and offline political participation among
youth at two different developmental stages. New Media &
Society,19(6), 899–917. doi:10.1177/1461444815624181
Kwak, N., Williams, A. E., Wang, X., & Lee, H. (2005). Talking
politics and engaging politics: An examination of the inter-
active relationships between structural features of political talk
and discussion engagement. Communication Research,32(1),
87–111. doi:10.1177/0093650204271400
Lazarsfeld, P. E., Berelson, B., & Gaudet, H. (1948). The people's
choice: How the voter makes up his mind in a presidential
campaign. New York, NY: Columbia University Press.
Lee, H. (2012). Communication mediation model of late-night
comedy: The mediating role of structural features of interper-
sonal talk between comedy viewing and political participation.
Mass Communication and Society,15(5), 647–671.
doi:10.1080/15205436.2012.664239
Lee, H., Kwak, N., & Campbell, S. W. (2015). Hearing the other
side revisited: The joint workings of cross-cutting discussion
and strong tie homogeneity in facilitating deliberative and
participatory democracy. Communication Research,42(4),
569–596. doi: 0093650213483824.
Lee, J. K., & Kim, E. (2017). Incidental exposure to news:
Predictors in the social media setting and effects on
14 S. W. YOO AND H. GIL DE ZÚÑIGA
information gain online. Computers in Human Behavior,
75, 1008–1015. doi:10.1016/j.chb.2017.02.018
Leighley, J. E. (1990). Social interaction and contextual influ-
ences on political participation. American Politics Research,
18(4), 459–475. doi:10.1177/1532673X9001800404
Lemert,J.B.(1984). News context and the elimination of mobi-
lizing information: An experiment. Journalism Quarterly,61,
243–249. doi:10.1177/107769908406100201
Luskin, R. C. (1990). Explaining political sophistication.
Political Behavior,12(4), 331–361. doi:10.1007/
BF00992793
Matthes, J. (2012). Exposure to counterattitudinal news cov-
erage and the timing of voting decisions. Communication
Research,39, 147–169. doi:10.1177/0093650211402322
McCombs, M. E., & Shaw, D. L. (1972). The agenda-setting
function of mass media. Public Opinion Quarterly,36(2),
176–187. doi:10.1086/267990
McLeod, J. M., Scheufele, D. A., & Moy, P. (1999).
Community, communication, and participation: The role
of mass media and interpersonal discussion in local poli-
tical participation. Political Communication,16(3),
315–336. doi:10.1080/105846099198659
McLeod, J. M., Sotirovic, M., & Holbert, R. L. (1998). Values
as sociotropic judgments influencing communication
patterns. Communication Research,25(5), 453–485.
doi:10.1177/009365098025005001
Mill, J. S. (1989). On Liberty and other writings, Ed. Stefan
Collini.Cambridge,UK.Cambridge University Press,8, 15-22.
Miller, W. E. (1991). Party identification, realignment, and
party voting: Back to the basics. The American Political
Science Review,85(2), 557–568.
Moy, P., Xenos, M. A., & Hess, V. K. (2005). Communication
and citizenship: Mapping the political effects of
infotainment. Mass Communication & Society,8(2),
111–131. doi:10.1207/s15327825mcs0802_3
Mutz, D. C. (2002). The consequences of cross-cutting net-
works for political participation. American Journal of
Political Science, 838–855. doi:10.2307/3088437
Mutz, D. C. (2006). Hearing the other side: Deliberative versus
participatory democracy. Cambridge, UK: Cambridge
University Press.
Negroponte, N. (1995). Being digital (pp. 65–67). London,
UK: Hoder and Stoughton.
Noelle-Neumann, E. (1993). The spiral of silence: Public opinion–
Our social skin.Chicago,IL:UniversityofChicagoPress.
Oeldorf-Hirsch, A. (2018). The role of engagement in learn-
ing from active and incidental news exposure on social
media. Mass Communication and Society,21(2), 225–247.
doi:10.1080/15205436.2017.1384022
Page, B. I., & Jones, C. C. (1979). Reciprocal effects of policy
preferences, party loyalties and the vote. The American
political science review,73(4)1071–1089.
Palmquist,B., G. D., & Schickler, E. (2002). Partisan hearts
and minds: Political parties and the social identities of
voters. New Haven, CT: Yale University Press.
Pew Research Center for the People and the Press, (2008).
Audience Segments in a Changing News Environment: Key
News Audiences Now Blend Online and Traditional Sources.
Preacher, K. J. (2015). Advances in mediation analysis: A survey
and synthesis of new developments. Annual Review of
Psychology,66, 825–852. doi:10.1146/annurev-psych-010814-
015258
Preacher, K. J., & Hayes, A. F. (2007). Asymptotic and
resampling strategies for assessing and comparing indirect
effects in multiple mediator models. Behavior Research
Methods,40, 879–891. doi:10.3758/BRM.40.3.879
Price, V., & Zaller, J. (1993). Who gets the news? Alternative
measures of news reception and their implications for
research. Public Opinion Quarterly,57(2), 133. doi:10.1086/
269363
Prior, M. (2007). Post-broadcast democracy: How media choice
increases inequality in political involvement and polarizes elec-
tions. Cambridge, UK: Cambridge University Press.
Quintelier, E., & Vissers, S. (2008). The effect of Internet use
on political participation: An analysis of survey results for
16-year-olds in Belgium. Social Science Computer Review,
26(4), 411–427. doi:10.1177/0894439307312631
Rittenberg, J., Tewksbury, D., & Casey, S. (2012). Media pre-
ferences and democracy: Refining the “relative entertain-
ment preference”hypothesis. Mass Communication and
Society,15(6), 921–942. doi:10.1080/15205436.2011.622065
Scheufele, D. A. (2002). Examining differential gains from
mass media and their implication for participatory
behavior. Communication Research,29,46–65.
doi:10.1177/009365020202900103
Scheufele, D. A., Nisbet, M. C., Brossard, D., & Nisbet, E. C.
(2004). Social structure and citizenship: Examining the impacts
of social setting, network heterogeneity, and informational
variables on political participation. Political Communication,
21, 315–338. doi:10.1080/10584600490481389
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
Shah, D. V., Cho, J., Nah, S., Gotlieb, M. R., Hwang, H., Lee, N.
J
.,Scholl,R.M.,&McLeod,D.M.(2007). Campaign ads,
online messaging, and participation: Extending the commu-
nication mediation model. Journal of Communication,57,
676–703. doi:10.1111/j.1460-2466.2007.00363.x
Sotirovic,M.,&McLeod,J.M.(2001). Values, communication
behavior, and political participation. Political Communication,
18, 273–300. doi:10.1080/10584600152400347
Stromer-Galley, J., & Wichowski, A. (2011). Political discus-
sion online. The Handbook of Internet Studies,11, 168.
Stroud, N. J. (2010). Polarization and partisan selective
exposure. Journal of Communication,60(3), 556–576.
doi:10.1111/jcom.2010.60.issue-3
Sunstein, C. (2003). . In Republic.com 2.0. Princeton, NJ:
Princeton University Press.
JOURNAL OF INFORMATION TECHNOLOGY & POLITICS 15
Sunstein, C. R. (2007). Republic.com 2.0. Princeton, NJ:
Princeton University Press.
Tewksbury, D., Weaver, A. J., & Maddex, B. D. (2001).
Accidentally informed: Incidental news exposure on the
World Wide Web. Journalism & Mass Communication
Quarterly,78(3), 533–554. doi:10.1177/107769900107800309
Trilling, D., & Schoenbach, K. (2015). Challenging selective
exposure: Do online news users choose sites that match
their interests and preferences? Digital Journalism,3(2),
140–157. doi:10.1080/21670811.2014.899749
Valentino, N. A., Beckman, M., . A., & Buhr, T. A. (2001).
A spiral of cynicism for some: The contingent effects of
campaign news frames on participation and confidence in
government. Political Communication,18(4), 347–367.
doi:10.1080/10584600152647083
Verba, S., Schlozman, K. L., & Brady, H. E. (1995). Voice and
equality: Civic volunteerism in American politics.
Cambridge, MA: Harvard University Press.
Vraga, E. K., Anderson, A. A., Kotcher, J. E., & Maibach, E. W.
(2015). Issue-specific engagement: How Facebook contributes
to opinion leadership and efficacy on energy and climate
issues. Journal of Information Technology & Politics,in
press.12(2): 200–218. doi:10.1080/19331681.2015.1034910.
Walsh, K. C., & Cramer, K. J. (2004). Talking about politics:
Informal groups and social identity in American life.
Chicago, IL: University of Chicago Press.
Webster, J. G. (2011). The duality of media: A structurational
theory of public attention. Communication Theory,21(1),
43–66. doi:10.1111/comt.2011.21.issue-1
Weeks, B. E., & Holbert, R. L. (2013). Predicting dissemination of
news content in social media: A focus on reception, friending,
and partisanship. Journalism & Mass Communication
Quarterly,90(2), 212–232. doi:10.1177/1077699013482906
Williams, J., & MacKinnon, D. P. (2008). Resampling and
distribution of the product methods for testing indirect
effects in complex models. Structural Equation Modeling,
15(1), 23–51. doi:10.1080/10705510701758166
Wojcieszak,M.E.,Baek,Y.M.,&Carpini,M.X.D.
(2010). Deliberative and participatory democracy?
Ideological strength and the processes leading from
deliberation to political engagement. International
Journal of Public Opinion Research,22(2), 1–27.
doi:10.1093/ijpor/edp050
Wojcieszak,M.E.,&Price,V.(2012). Perceived versus actual
disagreement: Which influences deliberative experiences?
Journal of Communication,62(3), 418–436. doi:10.1111/
jcom.2012.62.issue-3
Yoo, S., & Gil-de-Zúñiga, H. (2014). Connecting blog,
Twitter and Facebook use with gaps in knowledge and
participation. Communication & Society,27(4), 33–48.
doi:10.15581/003.27.4.33-48
Yoo, S., Kim, J., & Gil de Zúñiga. (2017). Cognitive benefits
for senders: Antecedents and effects of political expression
on social media. Journalism & Mass Communication
Quarterly,94(1), 17–37.
Zaller,J.(Ed.).(1992). The nature and origins of mass
opinion.Cambridge,UK:Cambridgeuniversitypress.
Appendix 1: Demographic profile of study survey and other comparable surveys
XXX Study Survey Dec. 2013–Jan.
2014
(%)
Pew Research Center for the People & the
Press
Political Survey July 2013
(%)
U.S. Census American Community
Survey
2012 (1-Year Estimates)
(%)
Age:
18–24 5.0 10.1 10.0
25–34 13.5 11.3 13.4
35–44 15.7 11.9 13.0
45–64 43.0 38.8 26.4
65 or more 22.8 28.6 13.7
Gender:
Male 50.0 49.9 49.2
Female 50.0 50.1 50.8
Race/Ethnicity:
White 76.2 72.2 73.9
Hispanic 7.5 11.2 16.9
African American 10.5 10.3 12.6
Asian 2.9 2.5 5.0
Education:
High school or less 19.3 32.5 41.6
Some college 34.5 27.6 29.2
Bachelor’s degree 30.5 22.6 18.2
Graduate degree 8.8 14.9 10.9
Household Income:
Less than $49,999 46.0 45.9 51.9
$50,000 to $99,999 36.5 26.1 32.7
$100,000 or more 17.4 17.2 15.4
16 S. W. YOO AND H. GIL DE ZÚÑIGA
