Content uploaded by Jayeon (Janey) Lee
Author content
All content in this area was uploaded by Jayeon (Janey) Lee on Dec 05, 2017
Content may be subject to copyright.
Article
Why People Post News on
Social Networking Sites:
A Focus on Technology
Adoption, Media Bias,
and Partisanship Strength
Jayeon Lee
1
and Hyunjin Song
2
Abstract
While social networking sites (SNSs) have become popular news platforms where
people receive and post news, little is known about why some people are relatively
more active in receiving and posting news on SNSs. Based on the uses and gratification
perspective, this study investigates (a) the direct and indirect effects of online news
seeking on SNS news posting, through SNS news exposure, (b) the role of technology
adoption as a moderator between online news seeking and SNS news exposure, and (c)
the role of media bias perceptions and partisanship strength as moderators between SNS
news exposure and SNS news posting. Analysis of 2010 Pew Research Center Media
Consumption Survey data (N¼2,259) reveal that online news seeking significantly
predicts SNS news posting, both directly and indirectly. While technology adoption was
not a significant moderator, media bias perception and partisanship strength significantly
interacted with SNS news exposure in predicting SNS news posting.
Keywords
social networking site, social media, news, posting, technology adoption, media bias,
partisanship, moderated mediation
1
Lehigh University, Bethlehem, PA, USA
2
University of Vienna, Vienna, Austria
Corresponding Author:
Jayeon Lee, Lehigh University, 33 Coppee Dr., Bethlehem, PA 18015, USA.
Email: jayeon.lee@lehigh.edu
Electronic News
1-21
ªThe Author(s) 2017
Reprints and permission:
sagepub.com/journalsPermissions.nav
DOI: 10.1177/1931243117692084
journals.sagepub.com/home/enx
As digital media continues to develop, Americans are experiencing major changes in
their news environment. One of the most dramatic changes in the news environment
involves the role social networking sites (SNSs), such as Facebook and Twitter,
increasingly play as news outlets. Half of Facebook and Twitter users and 62%of
Reddit users in the United States get news on those sites, and 34%of Facebook news
consumers “like” (i.e., subscribe to) news organizations or individual journalists
(Matsa & Mitchell, 2014).
What makes SNSs distinct from other media outlets is that those sites enable users
to not only receive news but also to actively change the dynamics of the news flow by
redistributing news on a potentially massive scale (Lee & Ma, 2012; Weeks & Hol-
bert, 2013). According to a recent survey conducted in the United States, half of the
SNS users share or repost news stories, images, or videos, while nearly as many (46%)
discuss news issues or events on SNSs (Matsa & Mitchell, 2014). SNS users’ news
posting has become a phenomenon of growing political importance; it empowers
ordinary citizens to become discussion moderators, both redistributing news and
raising issues from news to their social contacts (Lee & Ma, 2012). Therefore, it is
important to examine what psychological factors predict or condition individuals’
news posting behaviors. Whereas the potentials of SNSs to serve democratically
useful functions are receiving a great deal of scholarly attention (Gil de Zu
´n
˜iga, Jung,
& Valenzuela, 2012; Hampton, Goulet, Ranie, & Purcell, 2011; Nermeen, 2011),
relatively little is known about what facilitates SNS users’ news posting behaviors.
Based on the uses and gratification perspective, the present study first examines the
basic associations among general online news seeking, news exposure on SNSs, and
news posting on SNSs. It also addresses the question of how—directly, indirectly, or
both—online news seeking predicts SNS news posting. Second, this study tests the
role of technology adoption as a moderator between online news seeking and SNS
news exposure to address the question of why some online news users are more likely
to utilize SNSs for receiving news. Finally, based on the previous research (e.g.,
Bartels, 2000; Conroy, Feezell, & Guerrero, 2012; Lim & Golan, 2011; Rojas,
2010), we attempt to empirically elaborate on the role of media bias perceptions and
partisanship strength in the relationship between SNS news exposure and SNS news
posting. Testing this model with the potential moderators will provide greater under-
standing of why some SNS news receivers are more prone to news posting and what
facilitates it.
Predictors of SNS News Exposure and Posting
According to the uses and gratification perspective, individuals select and use media
channels or content to fulfill perceived needs for information, personal guidance,
personal relationships, identity formation, diversion, and experience-related gratifica-
tions (Blumler, 1979; Katz, Blumler, & Gurevitch, 1973; Ruggiero, 2000). These
motivations are tied to media exposure and effects. For example, those who seek to
keep up with news tend to pay more attention to news, in turn learning more from
2Electronic News XX(X)
news than others (Eveland, 2001). Information seeking is also one of the main pre-
dictors of news exposure; individuals who are interested in news are likely to search
and pay attention to news to make that content available for initial processing in
working memory (Eveland, 2001; Lee & Ma, 2012).
Although connecting with family and friends is the primary reason Americans use
SNSs (Smith, 2011), some people utilize their SNS feeds to satisfy their information-
seeking needs. Recent studies found that information seeking is one of the main
motivations driving people to use social media (Dunne, Lawlor, & Rowley, 2010;
Lee & Ma, 2012). An increasing number of SNS users follow news organizations or
individual journalists on SNSs to receive news directly from them (Matsa & Mitchell,
2014; Purcell, Rainie, Mitchell, Rosenstiel, & Olmstead, 2010). Because of the
sharing-oriented nature of social networking services, SNS users are likely to be
exposed to at least some news content through their news feeds (Glynn, Huge, &
Hoffman, 2012). The more actively internet users seek news, the more likely they are
exposed to news on SNSs. Thus, the following hypothesis is posited:
Hypothesis 1: Online news seeking is positively related to news exposure
on SNSs.
Early theorists, such as Tarde (1899/1989), Katz (1992), and Bryce (1888/1973)
envisioned the influence of news as a multistep process that unfolds as follows: (a)
news in the media fuels conversation, (b) conversation shapes opinion, and (c) opinion
triggers action. According to this perspective, news consumption is the driving force
of discussion. That discussion among news audiences plays a crucial role in drawing
democratically meaningful outcomes out of news consumption behaviors (Kim,
Wyatt, & Kats, 1999). In the current interactive web media environment, the gaps
between the steps have narrowed, and people’s experience of news itself is becoming
similar to what Habermas (1984) called “communicative action” or “socially-
engaging activity” (Purcell et al., 2010, p. 4)—meaning that news consumers take
part in the production and distribution of the news by creating, highlighting, repurpos-
ing, sharing, and haggling over news with other consumers in no fixed order. As
Gillmor (2004) argued, journalism models of a Web 2.0 environment are moving
away from a conventional “big media’s lecture” type of unilateral communication
toward “a conversation or seminar,” as the lines blur between news producers and
consumers. Today, any user can start the mass communication process.
This “conversational” or “socially engaging” nature of news consumption seems to
be more apparent on SNSs, as those sites allow multidirectional information distribu-
tions and multiparty discussions, unconstrained by time and geographical boundaries.
SNS users tend to share various news items in their online social networks and react to
news with their personal thoughts and opinions. Particularly, by posting news items on
their profile and exchanging messages through comment sections, users often stimu-
late subsequent online discussion or subsequent participatory actions (Purcell et al.,
2010). We observed the power of news sharing in countries under a dictatorship in
Lee and Song 3
North Africa and the Middle East, such as Tunisia, Egypt, and Libya, over the months
of the “Arab Spring.” When Mohamed Bouazizi, a young Tunisian street vendor, was
tortured to death by the police, Tunisians everywhere turned to social media posts and
text messages to spread the news of what had happened. Their SNS usage stimulated
heated discussions and protests with an unprecedented speed and scale (Howard &
Hussain, 2011). Because spreading news on SNSs is often a starting point of discus-
sion in and out of the outlet, news posting can be understood as a communicative
behavior that initiates or facilitates further behavioral outcomes.
As we see SNS news posting as a communicative action, we expect that exposure
to news on SNSs positively predicts SNS news posting. An abundance of literature
supports the strong tie between news consumption and discussion (e.g., Eisenstein,
1979; Gouldner, 1976; Habermas, 1984; Kim et al., 1999; Koch, 1994). According to
Anderson, Dardenne, and Killenberg (1996, p. 37), “news is what people talk about,
and news makes people talk.” When people find a news item that they think is worth
sharing, it takes them only a couple of clicks to distribute the item through their social
networks. Given that news consumption triggers discussion and political action (Kim
et al., 1999), the more people are exposed to news on SNSs, the more likely they are to
pass it along to others. Thus, the following hypothesis is posited.
Hypothesis 2: SNS news exposure is positively related to SNS news posting.
As we hypothesize that information seeking predicts SNS news exposure and SNS
news exposure predicts SNS news posting, it is reasonable to test if information
seeking directly predicts SNS news posting. In the United States, three fourths
(75%) of online news consumers had news forwarded through email or posts on SNSs,
and 37%of internet users have shared news stories through SNSs at least once (Purcell
et al., 2010). A study found that information seeking was significantly related to
intending to share news through social media (Lee & Ma, 2012). SNS users can
receive the organizations’ news email they signed up for or can actively search for
the latest updates about specific topics using search engines. These online news-
seeking activities are likely to be related to news posting. When SNS users receive
breaking news or latest updates about issues of interest, they are likely to share those
on their SNS page. Thus, the following hypothesis is posited:
Hypothesis 3: Online news seeking is positively related to SNS news posting.
The percentage of people who receive news via SNS, compared to those who do
not, has doubled in the United States since 2010, but the current ratio still remains the
same: Only about half of the people seeking news use SNSs to find information
(Kohut, Doherty, Dimock, & Keeter, 2012). It means that the other half do not utilize
SNSs for news while they use other online sources such as news websites. This
phenomenon may represent what van Dijk (2000) identified as a “usage gap” among
internet users, where some socioeconomic groups make far greater use of advanced
web applications than other user groups who have access to the internet. The
4Electronic News XX(X)
possibility of a usage gap is proven by the fact that younger internet users are more
active in news reception on SNSs than older internet users. In 2012, for example,
while 35%of American adults ages 65 and older got news online or on mobile
devices, including email the previous day, only 2%of the same age-group got it
through SNSs (Kohut et al., 2012). For Americans ages 18–24, on the other hand,
34%got news via SNSs. So, among online news users, what individual characteristic
may facilitate or deter their SNS adoption or SNS news reception?
New technology or computer software diffusion processes have gained many
communication and information science researchers’ attention (e.g., Holton, 2011;
Vishwanath, 2009). In the well-known paradigm of diffusion of innovation, Rogers
(1962) explains when, how, and why a new idea, practice, or technique is adopted or
rejected in a given society. Rogers identified five major elements that predict inno-
vation adoption intent: relative advantage (how improved over the previous innova-
tion), compatibility (how easily assimilated into their life), complexity (how
difficult to learn), trialability (how easily tested), and observability (how visible
to their peers). Modifying the categorization, Venkatesh, Morris, Davis, and Davis
(2003) created the unified theory of acceptance and use of technology (UTAUT).
According to the UTAUT, four beliefs about a particular innovation are core deter-
minants of an individual’s technology adoption: performance expectancy, effort
expectancy, social influence, and facilitating conditions. Perceived complexity
(effort expectancy) was identified as a core element in both Rogers’ and the UTAUT
models and often received particular attention as a determinant of adoption (e.g.,
DeSanctis & Poole, 1994).
If individuals perceive SNSs as time-consuming, less visible, and complicated
innovations that are hard to test, use, or assimilate into individuals’ life, they are less
likely to make an entrance into the realm of SNSs. On the other hand, for those who
are generally active in new media and technology adoption (i.e., early adopters), the
entry barrier is not so high: They tend to value new innovation and enjoy observing,
trying, and adopting new technology. Thus, the following hypothesis is raised.
Hypothesis 4: Technology adoption moderates the relationship between online
news seeking and SNS news exposure, with the influence of online news
seeking being greater for early adopters.
Moderating Role of Media Bias Perception and Partisanship
Strength
The act of SNS news posting has a persuasive nature because by posting news, users
explicitly or implicitly reveal what they like or think is important. The fact that
political posts increase over the course of the election cycle (Smith, 2009) reflects
this persuasive nature of news posting behavior. Also, individuals whose friends post
political content on SNSs are much more likely to change their minds about the issue
or become more involved with the issue after reading or discussing it on SNSs (Judd,
Lee and Song 5
2012). However, not all SNS newsreaders utilize this persuasive tool. According to a
recent study, only 28%of social media users have used the tools to post links to
political stories or articles for others to read (Rainie, Smith, Schlozman, Brady, &
Verba, 2012). In other words, 73%of social media users have not posted political
news on their profiles. Stutzman (2008) categorized people who are politically
active in the social media space into the following four types: the window-
shoppers, the toe dippers, the communicators, and the mavens depending on the
extent to which the users disseminate political messages. According to Stutzman, it
is not consumption of information but production and communication of informa-
tion that differentiates these types. Then, what motivates some online newsreaders
to communicate about content, while some online newsreaders remain passive
window-shoppers?
One possible facilitator we suggest is media bias perception. According to a recent
report, 77%of Americans think that news organizations tend to favor one side, and
80%say powerful people and organizations often influence news organizations
(Kohut et al., 2012). Previous studies indicate that presumption of media influence
on people is positively associated with willingness to engage in “corrective” behaviors
(e.g., Lim & Golan, 2011; Rojas, 2010). This implies that the specific motivation to
offset, or counterbalance, media’s influence facilitates newsreaders’ subsequent beha-
vior. Supposing that most SNS users presume a certain extent of media influence,
those who strongly believe that the news media are biased are more likely than those
with less such perception to engage in corrective behaviors to offset or counterbalance
(choose one adjective) the influence of the media. For instance, they may seek alter-
native news sources and spread news they think is fair. As Rojas (2010, p. 350)
argued, “in order to correct for theses biases, they will seek to infuse public debate
with their own opinion.” Thus, media bias perception may work as a proxy for
measuring individuals’ motivation for news posting to persuade others. Based on the
mechanism, the following hypotheses are raised.
Hypothesis 5a: Media bias perception moderates the relationship between SNS
news exposure and SNS news posting, with the influence of SNS news exposure
being greater for those who believe that news media are biased.
Hypothesis 5b: Media bias perception moderates the relationship between
online news seeking and SNS news posting, with the influence of online news
seeking being greater for those who believe that news media are biased.
Another factor that is likely to lead individuals to actively distribute news in their
social networks is partisanship strength. Partisanship in the United States is defined
as the level of identification with either of the two major (Democratic vs. Repub-
lican) parties. People tend to use partisanship as a heuristic cue in interpreting the
political information they receive and determining how to respond to it (Campbell,
Converse, Miller, & Stokes, 1960; Downs, 1957; Druckman, 2001; Rahn, 1993;
Zaller, 1992). Because of this heuristic nature of partisanship as a shortcut that
6Electronic News XX(X)
reduces individuals’ cognitive process to a much simpler judgmental operation,
partisans tend to be more prompt in carrying thoughts into execution. Abundant
research in political communication has found that those who have strong partisan or
political ideology tend to participate in civic or political behaviors actively, online,
offline, or both (Bartels, 2000; Conroy et al., 2012; Finkel & Opp, 1991). Even
among SNS users, strong partisans join a political or social group to advance a cause
(Rainie et al., 2012). Given that SNS news posting is a persuasive action, we can use
partisanship strength as a proxy measure of motivation to engage in communicative
or persuasive behaviors, including distribution of news (i.e., news posting) on SNSs.
Thus, the following hypotheses are posited:
Hypothesis 6a: Partisanship strength moderates the relationship between SNS
news exposure and SNS news posting, with the influence of SNS news exposure
being greater for those who have strong partisanship.
Hypothesis 6b: Partisanship strength moderates the relationship between online
news seeking and SNS news posting, with the influence of online news seeking
being greater for those who have strong partisanship.
Method
To test our model suggested by the hypotheses (see Figure 1), we used national
telephone survey data collected by Pew Research Center for the People & the Press
in June 2010.
1
The original data combined landline random digit dialing (n¼2,005)
and cellular phone survey (n¼1,001) data (response rate: landline 16.9%and cellular
17.5%). The combined sample is weighted to achieve statistical representativeness of
the data, with the margin of sampling error +2.1 percentage points, based on the 2009
Figure 1. Hypothesized conditional process model predicting social networking site news
posting.
Lee and Song 7
Census Bureau’s demographic profiles. Among the 3,006 respondents, only those who
answered online news seeking questions—getting news online and using search
engine for news—were included in our final analysis (N¼2,259).
Measures
Online news seeking. Respondents were asked to identify (a) how often they receive
news online and (b) how often they use search engines such as Google to search
specific news topics that they are interested in on a 6-point scale ranging from never
(1) to every day (6). By averaging those 2 items, the online news use scale was created
(correcting passive voice; Spearman–Brown ¼.62, M¼4.36, SD ¼1.73). Initial
response options were reversed prior to summing the items, so that a higher value
would represent a higher value of online news seeking behaviors.
SNS news exposure and SNS news posting. Respondents were asked how often, if ever,
they get news or news headlines through SNSs (M¼1.16, SD ¼1.09) based on a 4-
point scale from never (0) to regularly (3). Respondents were also asked how often, if
ever, they post news or news headlines on SNSs (M¼0.62, SD ¼0.88) with a 4-point
scale from never (0) to regularly (3). As with online news seeking, all of the original
response options were reversed prior to deriving the measure.
Technology adoption. Respondents’ tendency of new communication technology adop-
tion was assessed with five dichotomous questions asking if respondent used a desktop
computer, laptop computer, cell phone, MP3 player, or tablet computer (1 ¼yes,0¼
no). Scores for all five questions were summed across a 6-point scale ranging from
none (0) to high (5), with higher values representing a wider range of new commu-
nication technology use (KR-20 ¼.58, M¼2.58, SD ¼1.26).
2
Media bias perception. The extent to which individuals perceived that news media were
biased was assessed by asking, “How much political bias do you see in news cover-
age?” with a 4-point scale (after reversing the initial options) from none at all (1) to a
lot (4), regardless of their partisan directionality. On average, respondents perceived
that the news coverage was significantly biased (M¼3.41, SD ¼0.82).
Partisanship strength. Following a standard treatment of 7-point scale partisan identity
measures, individuals’ partisanship strength was constructed by folding respondents’
self-identified political party affiliation and political leaning (see Hetherington, 2001;
Lavine, Johnston, & Steenbergen, 2012, for more details about the method; also see
Greene, 2002, for alternative operationalization of partisanship strength and their
correlation patterns with the standard 4-scale folded measure). First, respondents who
identified themselves as Republicans (n¼841) and Democrats (n¼961) were
assigned 3 points (Strong) in the partisanship strength scale. Independents (n¼
993) and others (n¼211) were additionally asked whether they lean more to the
8Electronic News XX(X)
Republican Party,the Democratic Party,ornone of the parties (n¼392). Those who
chose Republican Party (n¼433) or Democratic Party (n¼379) were assigned 2
points (weak) in the partisanship strength scale (n¼812). Lastly, those who leaned to
neither Republican nor Democratic Party were assigned 1 point (none) in the scale
(n¼392). On average, the majority of the respondents showed relatively strong
partisanship (M¼2.46, SD ¼0.71).
Control variables. Several socioeconomic variables were controlled in the analysis. This
includes respondents’ gender (56.92%female), age (M¼51.02, SD ¼18.40), race
(76.78%White), education (M¼4.72, SD ¼1.60), employment status (M¼0.98, SD
¼0.93), and annual household income (M¼5.20, SD ¼2.39). Education was
measured on a 7-point scale, from Grade 8 or lower (1) to postgraduate or higher
(7). Employment status was tapped on a 3-point scale from not employed (0) to full-
time job (2). Annual household income was measured on a 9-point scale, from less
than US$10,000 (1) to US$150,000 or more (9), with $40,000 to under $50,000 (5)
being a middle point of the scale.
We also controlled several media-related variables: attention to news media
(Cronbach’s a¼.77, M¼2.81, SD ¼0.86), network TV exposure (Cronbach’s
a¼.53, M¼2.30, SD ¼1.05), cable TV exposure (Cronbach’s a¼.61, M¼
2.60, SD ¼1.01), local TV exposure (M¼3.23, SD ¼1.01), newspaper exposure
(M¼2.98, SD ¼1.15), and news magazine exposure (M¼2.05, SD ¼1.01).
Attention to news media was measured with four questions asking how closely
respondents were following news about (a) current events in Afghanistan, (b) this
year’s congressional elections, (c) conditions of the national economy, and (d) news in
general. For network TV exposure, respondents indicated how regularly they had
watched national nightly news on CBS, ABC, or NBC. For cable news exposure,
respondents indicated how regularly they had watched CNN, MSNBC, or the Fox
News Cable. All media attention and exposure measures were based on a 4-point scale
from never (1) to regularly (4), with higher values representing greater attention or
exposure to respective media.
Analysis
Analysis progressed in three steps, all of which used a series of bootstrapped resam-
pling estimation method recommended by Hayes (2009, 2013; also see Preacher &
Hayes, 2004). Hayes (2009) discusses the major weaknesses of traditional methods of
testing mediation, such as Sobel test (1986), in estimating direct, indirect, and total
effects. Because the normality assumption of sampling distribution is often violated in
Sobel test, Hayes suggests the bootstrapping resampling method that offers more
powerful and reliable estimates of coefficients.
First, we assessed the effect of online news seeking on SNS news posting, both
directly and indirectly through SNS news exposure, using standard path-analytic
mediation analysis coupled with the bootstrapped resampling method. Next, we
Lee and Song 9
assessed if technology adoption interacted with online news seeking in predicting
SNS news exposure and if media bias perception and partisanship strength inter-
acted with SNS news exposure in predicting SNS news posting. In probing the
moderation effects, we used simple moderation analysis along with the moderated
mediation approach described by Preacher, Rucker, and Hayes (2007). The central
analytic foci of these analyses involve (1) the indirect effect of online news seeking
on SNS news posting through SNS news exposure and (2) moderating roles of
individuals’ levels of technology adoption, bias perception, and political identifica-
tion strength in the model.
The analyses were implemented using PROCESS macro (Hayes, 2013), which
incorporates bootstrapping-based inferences about indirect effects in ordinary least
squares (OLS)-based moderation-mediation models. Following the recommendation
suggested by Hayes (2009), a total of 5,000 bootstrapping resamples were generated
from 1,073 to 1,093 original samples (depending on specific models) and from 1,073
original samples for the final conditional process model. In the analyses,
heteroscedasticity-consistent SEs and bias-corrected maximum likelihood confidence
intervals (CIs) were estimated at .95 confidence level. All of the covariates were
controlled for both the focal mediator (i.e., SNS news exposure) and the dependent
variable (i.e., SNS news posting). Also, all of the moderators and their product terms
were mean centered in order to facilitate meaningful interpretations.
Results
Analysis of Simple Mediation
To test our proposed model, we began by analyzing the simple mediation from
online news seeking to SNS news posting through SNS news exposure. Our first
hypothesis posited that active online news users were more likely to be exposed to
news on SNSs, above and beyond the effects of demographic factors and other
media-related control variables. The result supported Hypothesis 1 (b¼.11, SE ¼
.02, t¼4.97, p< .001, CI ¼[.07, .16]). Our second hypothesis posited that those
whoreceivednewsonSNSswerealsolikelytopostonSNSs.Asexpected,SNS
news exposure had a positive and statistically significant association on SNS news
posting (b¼.36, SE ¼.02, t¼14.55, p< .001, CI ¼[.31, .41]) above and beyond the
effect of online news seeking (b¼.05, SE ¼.02, t¼3.08, p<.01,CI¼[.02, .09])
and other control variables, thereby supporting Hypothesis 2. The baseline media-
tion model explained 7.53%,r¼.27, R
2
¼.075, F(13, 1079) ¼7.59, p< .001, and
27.36%,r¼.52, R
2
¼.27, F(14, 1078) ¼27.45, p< .001, of total variance in SNS
news exposure and SNS news posting variable, respectively.
Next, the result of bootstrapping analysis indicated that the effect of online news
seeking on SNS news posting was significantly mediated through the respondent’s
SNS news exposure. Thus, Hypothesis 3 was confirmed. The magnitude of this indi-
rect effect (b¼.04, SE ¼.01, p< .001, CI ¼[.02, 06]) was approximately equivalent
10 Electronic News XX(X)
with that of the direct effect of online news seeking on SNS posting (b¼.05, SE ¼.02,
p< .01, CI ¼[.02, .09]).
Analyses of Simple Moderations (Two-Way Interactions)
Our fourth hypothesis posited that technology adoption would significantly moderate
the relationship between online news seeking and SNS news exposure, with the
influence of online news seeking being greater for early adopters. To test the roles
of technology adoption as a moderator of the effect of online news seeking, we
estimated an OLS regression model predicting SNS news exposure. Unlike our expec-
tation, we found no discernible moderating effect (b¼.01, SE ¼.02, p¼ns,CI¼
[.04, .05]) of technology adoption on respondents’ SNS news exposure. Thus,
Hypothesis 4 was not supported. The model, however, accounted for a significant
variance of SNS news exposure, r¼.28, R
2
¼.08, F(15, 1075) ¼6.71, p< .001.
Next, we hypothesized that media bias perception and partisanship strength would
moderate the influence of online news seeking and SNS news exposure on SNS news
posting, with the influence being greater for those who have a higher level of media
bias perception (Hypotheses 5a and 5b) and strong partisanship (Hypotheses 6a and
6b). Two separate OLS regression models predicting SNS news posting from media
bias perception and partisanship strength were estimated. The results revealed that
media bias perception (b¼.03, SE ¼.04, p¼ns,CI¼[.01, .06]) and partisanship
strength (b¼.02, SE ¼.02, p¼ns,CI¼[.07, .03]) did not significantly moderate
the direct effect of online news seeking on SNS news posting, although the model
accounted for significant variance of the dependent variable, r¼.52, R
2
¼.28, F(18,
1054) ¼21.49, p< .001. However, the effect of respondents’ SNS news exposure was
found to be significantly conditioned by both media bias perception (b¼.08, SE ¼
.03, p< .01, CI ¼[.02, .14]) and partisanship strength (b¼.09, SE ¼.04, p< .05, CI ¼
[01, .16]), controlling for demographic and other control variables. To further under-
stand the nature of these moderations, the conditional effect of SNS news exposure on
SNS news posting as a function of each of the moderators was probed with the sample
mean (being “moderate”) and plus and minus 1 standard deviation from the mean
(being “high” and “low,” respectively), controlling covariates in their mean level (see
Figures 2 and 3). Overall, our simple moderation model accounted for approximately
28.37%of variance in respondents’ SNS news posting behavior, r¼.53, R
2
¼.28,
F(18, 1054) ¼22.46, p< .001.
Analysis of Moderated Mediation
Finally, we turned to the analysis of the proposed moderated mediation. The simple
mediation analysis provided evidence of a positive effect of online news seeking
through SNS news exposure, which in turn was related to a higher level of news
posting on SNSs. The moderation analysis revealed that the effect of SNS news
exposure, which served as a mediator, was dependent on media bias perception and
Lee and Song 11
partisanship strength, with the effect of SNS news exposure being greater among those
who perceived the media as biased or had strong partisanship. Thus, the separate
mediation and moderation analyses jointly suggested the possibility of “moderated
mediation” (Preacher, Rucker, & Hayes, 2007); the indirect effect of online news
seeking through SNS news exposure was moderated by media bias perception and
partisanship strength, as depicted in our modified model (see Figure 4).
To test the proposed moderated mediation relationships, the bootstrapping resam-
pling analysis was conducted. The result, which is presented in Table 1, suggested that
both the direct effect of online news seeking (b¼.05, SE ¼.02, t¼2.78, p< .01, CI ¼
[.01, .08]) and its indirect effect through SNS news exposure were significant at .05
level, regardless of the level of moderators. As predicted, the indirect effect of online
news seeking on SNS news posting through SNS news exposure was stronger (b¼
.056, SE ¼.012, p< .05, CI ¼[.033, .080]) for those who had greater perception of
news media bias (.73 at þ1SD above the mean) and stronger partisanship (.64 at þ1
SD above the mean) than average. In contrast, for those who had less than average bias
perception (.73 at 1SD below the mean) and partisanship (.64 at 1SD below
Figure 2. Conditional effects of social networking site (SNS) news exposure on SNS news
posting as a function of media bias perception.
12 Electronic News XX(X)
Figure 3. Conditional effects of social networking site (SNS) news exposure on SNS news
posting as a function of partisanship strength.
Figure 4. Modified model of conditional process model predicting social networking site news
posting.
Lee and Song 13
the mean), the conditional indirect effect of online news seeking through SNS news
exposure was lower (b¼.028, SE ¼.007, p< .05, CI ¼[.014, .045]).
Discussion
The increasing number of people who consume and share news on SNSs awakens the
need for scholars to examine and understand the acts of news receiving and posting.
Communication literature has long demonstrated that discussion promotes democra-
tically desirable outcomes, such as knowledge, reasoned opinion, political sophistica-
tion, and civic participation (e.g., Delli Carpini, Cook, & Jacobs, 2004; Gastil &
Dillard, 1999; McLeod et al., 1999; Pattie & Johnston, 2009), and that online discus-
sion is not necessarily inferior to face-to-face discussion (Min, 2007; Price, 2006;
Price & Cappella, 2002; Stanley & Weare, 2004). As a unique act of online discussion
initiation, news posting behaviors of SNS users call for a closer investigation.
The present study sought to dissect the mechanism through which citizens engage
with news on SNSs and tested the influences of various individual characteristics
that might directly or indirectly encourage their news posting. In doing so, we paid
specific attention to the moderating role of technology adoption, media bias percep-
tion, and partisanship strength in various stages of testing using our basic mediation
model. The results indicate that both online news seeking and news exposure on
SNSsarehighlypredictiveofnewspostingonSNSs;themorenewspeopleare
exposed to, the more they share news content with others in their social networks.
Unlike most previous research, this study separately tested the influence of SNS
news exposure from other online news-seeking activities. As SNSs have unique
characteristics, such as the ease and the frequency of news sharing through net-
works, news exposure on SNSs needs to be examined separately from other online
Table 1. Conditional Indirect Effects of Online News Seeking on SNS News Posting, at Each
Value of Moderators.
Media Bias Perception Partisanship Strength Indirect Effect (b) Boot SE
Bootstrap CI
[Lower, Upper]
.73 (–1 SD).64 (1SD) .0286 .0075 [.0155, .0452]
.73 (–1 SD) .00 (Mean) .0351 .0080 [.0204, .0520]
.73 (–1 SD).64(þ1SD) .0416 .0096 [.0248, .0627]
.00 (Mean) .64 (1SD) .0358 .0085 [.0202, .0539]
.00 (Mean) .00 (Mean) .0423 .0090 [.0248, .0607]
.00 (Mean) .64 (þ1SD) .0488 .0106 [.0293, .0707]
.73 (þ1SD).64 (1SD) .0430 .0103 [.0243, .0642]
.73 (þ1SD) .00 (Mean) .0495 .0109 [.0290, .0713]
.73 (þ1SD).64(þ1SD) .0560 .0123 [.0330, .0809]
Note. All effects were significant at 95% level. CIs are estimated using 5,000 bootstrapped resamples with
bias-corrected, heteroscedasticity-consistent SEs. SNS ¼social networking site. CIs ¼confidence intervals.
14 Electronic News XX(X)
news activities. Indeed, SNS news exposure was found to be a much stronger pre-
dictor of SNS news posting than online news activities.
More importantly, we have found that the effect of SNS news exposure on SNS
news posting is contingent on media bias perception and partisanship strength. People
with a stronger belief of media bias prevalence and stronger partisanships were sig-
nificantly more active in news posting on SNSs. This is largely consistent with recent
evidence in hostile media perception research that the media bias and indignation
further promotes one’s willingness to engage in discursive activities (Hwang, Pan, &
Sun, 2008). Our findings support the idea that seeing mainstream media as biased has
tangible behavioral implications on discursive activities in social media—namely,
news consumption and sharing behaviors on SNSs; people with stronger media bias
perception may engage in news posting more actively to offset the harmful effect of
the biased media. Also, our findings suggest that people with stronger partisanships
tend to be more motivated to persuade those who have different opinions with them or
do not have strong opinions. Overall, our findings support motivation-boosting nature
of media bias perception and partisanship. The findings are also in line with the uses
and gratifications perspective, in that individuals’ motivations are the driving forces of
what they do with the media.
It is worth noting that the moderating roles of media bias perception and partisan-
ship strength were not significant between online news seeking (as opposed to SNS
news consumption) and SNS news posting. One possibility concerns with the differ-
ences between regular news sites and SNSs (e.g., presence or visibility of user-
generated comments). Research suggests that when encountered with comments
opposing their viewpoints, individuals perceive news articles to be more biased
against their own opinion (Lee, 2012). This may further amplify motivations to
counterbalance media’s influence. As active online news seeking is more likely to
result in attitude-consonant information, bias perceptions and partisanships may play a
limited role. Another possibility is that the main effect of online news seeking on SNS
news posting was so strong that it overpowers the influences of other motivational
factors. Frequent online news seekers may already be highly motivated to share news
on SNS, regardless of the level of bias perceptions or partisanship.
This motivational nature of online news seeking may also explain our failure of
finding the moderating role of technology adoption in predicting SNS news exposure.
It is possible that active online users adopt SNSs and receive news on sites regardless
of their general tendency toward technology adoption. Meanwhile, the technology
adoption tendency may explain whether people create and manage an account on
SNSs. Indeed, in our additional test, technology adoption was a strongly significant
predictor of SNS adoption (b¼.44, SE ¼.06, p< .001), with demographic and other
traditional and online news use variables controlled. This means that while early
adopters of technology tend to adopt SNSs early as well, once they begin consuming
news online, they do not need to be particularly technology savvy to utilize SNSs as
additional news outlets. For instance, among Rogers’ five factors that influence adop-
tion decision, relative advantage or disadvantage may be more crucial than
Lee and Song 15
complexity. That being said, those who think using SNS is not very productive will
not actively utilize the sites, no matter how active they are in adopting other new
technologies. Future research that looks into this relationship is encouraged.
Being a secondary analysis of existing survey data, the present study is not without
limitation. Principally, the cross-sectional nature of the data precludes testing for
causality. For example, other directions of causality (e.g., the more people post news,
the more they are engage in online news activities) or reciprocal relationships are
possible. Time series analysis or experimental studies are recommended for causality
establishment. Second, considering today’s fast changing media environments, the
2010 data set we used is not very young. We needed the data set to get all key variables
we wanted to use in this study. However, we believe the 2010 data set is still valid for
testing overall relationships between our key variables, even if the means of some
measures (e.g., SNS news exposure and posting) might have increased. We should
also acknowledge that many of the measures we used are not ideal (e.g., single- or two
question items). Future research, which employs more fine-tuned measurement,
would greatly add to this area of study. Also, while we focused on political motiva-
tions in the present study, caution should be taken not to assume that we are arguing
that political motivations are the primary reasons why people post news on SNSs.
Political postings make up a very small portion of information consumed on SNSs
when compared to news about entertainment, community, and sports (Anderson &
Caumont, 2014), and people’s hedonic motivation also plays a crucial role in shaping
their news consumption behaviors (Holbert, Hill, & Lee, 2014).
Despite these limitations, we believe that the present study contributes to expand-
ing our understanding of current news consumers’ news activities on SNSs. Our
findings add support to existing research that has shown motivation-boosting effects
of media bias perceptions and partisanship by presenting another example in the
context of social media. Having a partisanship and perceiving the media unfair were
the factors that have led some of SNS news consumers to take another step to actively
share news through SNSs. Considering the positive roles of informational SNS uses in
predicting civic and political engagement, communication scholars need to examine
factors that can facilitate SNS news posting. In this study, we attempted to test how
different behaviors within the sites relate to one another in order to develop a model
that predicts news posting behaviors. Future scholars are encouraged to investigate
other factors that motivate social media users’ active news engagement.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research,
authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publi-
cation of this article.
16 Electronic News XX(X)
Notes
1. A detailed description of this data set and survey methodologies can be found at http://www.
people-press.org/2010/09/12/about-the-survey-4
2. Considering the various range of communication technologies asked in the measure, this
measure is better thought of as indices than scales (see Streiner, 2003, for a detailed dis-
cussion of this distinction), which are not meant to be held to standards of internal
consistency.
References
Anderson, M., & Caumont, A. (2014). How social media is reshaping news. Retrieved from
http://www.pewresearch.org/fact-tank/2014/09/24/how-social-media-is-reshaping-news
Anderson, R., Dardenne, R., & Killenberg, G. (1996). The conversation of journalism: Com-
munication, community, and news (2nd ed.). Westport, CT: Praeger.
Bartels, L. (2000). Partisanship and voting behavior. American Journal of Political Science,40,
35–50.
Blumler, J. G. (1979). The role of theory in uses and gratifications studies. Communication
Research,6, 9–36.
Bryce, J. (1973). The American commonwealth (Vol. 3). London, England: Macmillan.
(Original work published 1888)
Campbell, A., Converse, P., Miller, W., & Stokes, D. (1960). The American voter. New York:
John Wiley and Sons; Chicago, IL: University of Chicago Press.
Conroy, M., Feezell, J., & Guerrero, M. (2012). Facebook and political engagement: A study of
online political group membership and offline political engagement. Computers in Human
Behavior,28, 1535–1546.
Delli Carpini, M., Cook, F., & Jacobs, L. (2004). Public deliberation, discursive participation,
and citizen engagement: A review of the empirical literature. Annual Review of Political
Science,7, 315–344.
DeSanctis, G., & Poole, M. (1994). Capturing the complexity in advanced technology use:
Adaptive structuration theory. Organization Science,5, 121–147.
Downs, A. (1957). An economic theory of democracy. Boston, MA: Addison Wesley.
Druckman, J. (2001). On the limits of framing effects: Who can frame? Journal of Politics,63,
1041–1066.
Dunne, A
´., Lawlor, M., & Rowley, J. (2010). Young people’s use of online social networking
sites—A uses and gratifications perspective. Journal of Research in Interactive Marketing,
4, 46–58.
Eisenstein, E. (1979). The printing press as an agent of change: Communications and cultural
transformations in early-modern Europe (Vol. 2). Cambridge, England: Cambridge.
Eveland, W. P. Jr. (2001). The cognitive mediation model of learning from the news evidence
from nonelection, off-year election, and presidential election contexts. Communication
Research,28, 571–601.
Finkel, S., & Opp, K. L. (1991). Party identification and participation in collective political
action. The Journal of Politics,53, 339–371.
Lee and Song 17
Gastil, J., & Dillard, J. (1999). Increasing political sophistication through public deliberation.
Political Communication,16, 3–23.
Gil de Zu
´n
˜iga, H., Jung, N., & Valenzuela, S. (2012). Social media use for news and individ-
uals’ social capital, civic engagement and political participation. Journal of Computer-
Mediated Communication,17, 331–336.
Gillmor, D. (2004). We the media: Grassroots journalism by the people, for the people. Sebas-
topol, CA: O’Reilly.
Glynn, C., Huge, M., & Hoffman, L. (2012). All the news that’s fit to post: A profile of news use
on social networking sites. Computers in Human Behavior,13, 113–119.
Gouldner, A. (1976). The dialectic of ideology and technology: The origins, grammar, and
future of ideology. London, England: Macmillan.
Greene, S. (2002). The social-psychological measurement of partisanship. Political Behavior,
24, 171–197.
Habermas, J. (1984). The theory of communicative action: Reason and the rationalization of
society (Vol. 1) (T. McCarthy, Trans.). Boston, MA: Beacon Press.
Hampton, K., Goulet, L., Ranie, L. S., & Purcell, K. (2011). Social networking sites and our
lives. Part 4: Trust, support, perspective taking, and democratic engagement. Retrieved
from http://www.pewinternet.org/Reports/2011/Technology-and-social-networks/Part-4/
Political-Engagement.aspx
Hayes, A. (2009). Beyond Baron and Kenny: Statistical mediation analysis in the new millen-
nium. Communication Monographs,76, 408–420.
Hayes, A. (2013). Introduction to mediation, moderation, and conditional process analysis: A
regression-based approach. New York, NY: The Guilford Press.
Hetherington, M. J. (2001). Resurgent mass partisanship: The role of elite polarization. Amer-
ican Political Science Review,95, 619–631.
Holbert, R. L., Hill, M., & Lee, J. (2014). The political relevance of entertainment media. In C.
Reinemann (Eds.), Handbooks of communication science, Vol. 18: Political communication
(pp. 427–446). Berlin, Germany: De Gruyter Mouton.
Holton, V. (2011). Negating nodes and liquid fragmentation: Extending conversations of diffu-
sion, social networks, and fragmentation. Communication Theory,22, 279–298.
Howard, P., & Hussain, M. (2011). The role of digital media. Journal of Democracy,22, 35–48.
Hwang, H., Pan, Z., & Sun, Y. (2008). Influence of hostile media perception on willingness to
engage in discursive activities: An examination of mediating role of media indignation.
Media Psychology,11, 76–97.
Judd, N. (2012). If your friend writes a political rant on Facebook, will it change your mind?
Pew Research Center, US, September. Retrieved from http://techpresident.com/news/
22809/if-your-friend-writes-political-rant-facebook-will-it-change-your-mind
Katz, E. (1992). On parenting a paradigm: Gabriel Tarde’s agenda for opinion and communi-
cation research. International Journal of Public Opinion Research,4, 80–85.
Katz, E., Blumler, J. G., & Gurevitch, M. (1973). Uses and gratifications research. Public
Opinion Quarterly,37, 509–523.
Kim, J., Wyatt, R., & Kats, E. (1999). News, talk, opinion, participation: The part played by
conversation in deliberative democracy. Political Communication,16, 361–385.
18 Electronic News XX(X)
Koch, N. (1994). Changing times? The effect of The New York Times on college students’
political information and behavior. Social Science Journal,31, 29–38.
Kohut, A., Doherty, C., Dimock, M., & Keeter, S. (2012). Views of the news media: 1985-2011:
Press widely criticized, but trusted more than other information sources. Retrieved from
http://www.people-press.org/files/legacy-pdf/9-22-2011%20Media%20Attitudes%
20Release.pdf
Lavine, H. G., Johnston, C. D., & Steenbergen, M. R. (2012). The ambivalent partisan: How
critical loyalty promotes democracy. New York: Oxford University Press.
Lee, E. J. (2012). That’s not the way it is: How user-generated comments on the news affect
perceived media bias. Journal of Computer-Mediated Communication,18, 32–45.
Lee, C. S., & Ma, L. (2012). News sharing in social media: The effect of gratifications and prior
experience. Computers in Human Behavior,28, 331–339.
Lim, J. S., & Golan, G. (2011). Social media activism in response to the influence of political
parody videos on Youtube. Communication Research,38, 710–727.
Matsa, K. E., & Mitchell, A. (2014). 8 Key takeaways about social media and news. Retrieved
from http://www.journalism.org/2014/03/26/8-key-takeaways-about-social-media-and-
news
McLeod, J. M., Scheufele, D. A., Moy, P., Horowitz, E. M., Holbert, R. L., Zhang, W., ...
Zubric, J. (1999). Understanding deliberation the effects of discussion networks on partic-
ipation in a public forum. Communication Research,26, 743–774.
Min, S. J. (2007). Online vs. face-to-face deliberation: Effects on civic engagement. Journal of
Computer-Mediated Communication,12, 1369–1387.
Nermeen, S. (2011). Towards the Egyptian revolution: Activists’ perceptions of social media
for mobilization. Journal of Arab & Muslim Media Research,4, 273–209.
Pattie, C., & Johnston, R. (2009). Conversation, disagreement, and political participation.
Political Behavior,31, 261–285.
Preacher, K., & Hayes, A. (2004). SPSS and SAS procedures for estimating indirect
effects in simple models. Behavior Research Methods, Instruments, & Computers,
36, 717–731.
Preacher, K., Rucker, D., & Hayes, A. (2007). Addressing moderated mediation hypotheses:
Theory, methods, and prescriptions. Multivariate Behavioral Research,42, 185–227.
Price, V. (2006). Citizens deliberating online: Theory and some evidence. In T. Davies & B. S.
Noveck (Eds.), Online deliberation: Design, research and practice (pp. 37–58). Chicago,
IL: University of Chicago Press.
Price, V., & Cappella, J. (2002). Online deliberation and its influence: The electronic dialogue
project in campaign 2000. IT & Society,1, 303–329.
Purcell, K., Rainie, L., Mitchell, A., Rosenstiel, T., & Olmstead, K. (2010). Understanding of the
participatory news consumer: How Internet and cell phone users have turned news into social
experience. Retrieved from http://www.journalism.org/analysis_report/understanding_
participatory_news_consumer
Rahn, W. (1993). The role of partisan stereotypes in information processing about political
candidates. American Journal of Political Science,37, 472–496.
Lee and Song 19
Rainie, L., Smith, A., Schlozman, K. L., Brady, H., & Verba, S. (2012). Social media and
political engagement. Retrieved from http://pewinternet.org/Reports/2012/Political-engage
ment/Summary-of-Findings.aspx
Rogers, E. (1962). Diffusion of innovations. New York, NY: Free Press.
Rojas, H. (2010). “Corrective” actions in the public sphere: How perceptions of media and
media effects shape political behaviors. International Journal of Public Opinion Research,
22, 343–363.
Ruggiero, T. E. (2000). Uses and gratifications theory in the 21st century. Mass Communication
and Society,3, 3–37.
Smith, A. (2009). The Internet’s role in campaign 2008. Report, Pew Research Center, US,
April. Retrieved from http://www.pewinternet.org/Reports/2009/6–The-Internets-Role-in-
Campaign-2008.aspx
Smith, A. (2011). Why Americans use social media. Report, Pew Research Center, US, Novem-
ber. Retrieved from http://www.pewinternet.org/Reports/2011/Why-Americans-Use-
Social-Media.aspx?src¼prc-headline
Sobel, M. (1986). Some new results on indirect effects and their standard errors in covariance
structure. Sociological Methodology,16, 159–186.
Stanley, J., & Weare, C. (2004). The effects of Internet use on political participation: Evidence
from an agency online discussion forum. Administration & Society,36, 503–527.
Streiner, D. L. (2003). Being inconsistent about consistency: When coefficient alpha does and
doesn’t matter. Journal of Personality Assessment,80, 217–222.
Stutzman, F. (2008). The social media voter. In Techpresident. Retrieved from http://techpre
sident.com/blog-entry/social-media-voter
Tarde, G. (1989). Opinion and conversation (J. Ruth, Trans.). Unpublished translation of
L’opinion et la foule [Mass opinion]. Paris, France: Presses Universitaires de France. (Orig-
inal work published 1899)
Van Dijk, J. (2000). Widening information gaps and policies of prevention. In K. L. Hacker & J.
van Dijk (Eds.), Digital democracy: Issues of theory & practice (pp. 166–183). London,
England: Sage.
Venkatesh, V., Morris, M. G., Davis, G. B., & Davis, F. D. (2003). User acceptance of infor-
mation technology: Toward a unified view. MIS Quarterly,27, 425–478.
Vishwanath, A. (2009). From belief-importance to intention: The impact of framing on tech-
nology adoption. Communication Monographs,76, 177–206.
Weeks, B., & Holbert, R. L. (2013). Predicting dissemination of news content in social media: A
focus on reception, friending, and partisanship. Journalism and Mass Communication Quar-
terly,90, 212–232.
Zaller, J. (1992). The nature and origins of mass opinion. New York, NY: Cambridge Univer-
sity Press.
Author Biographies
Jayeon Lee (PhD, The Ohio State University, 2013) is an assistant professor at the Department
of Journalism & Communication, Lehigh University, PA. Her research and teaching interests
20 Electronic News XX(X)
revolve around social media and new technology, particularly users’ impression formation and
strategic self-presentation on social media in the context of political communication and
journalism.
Hyunjin Song (PhD, The Ohio State University, 2015) is a post-doc university assistant in the
Department of Communication at University of Vienna. His research interests include the
influence of interpersonal discussion on political engagement and statistical modeling of social
networks.
Lee and Song 21
