Social Media News Use and Political Cynicism: Differential Pathways Through “News Finds Me” Perception

Abstract

As citizens inadvertently “encounter” news and political information through digital media and social networking sites, they might perceive themselves to be well informed about politics without actively seeking political information, which has been labeled as the “News Finds Me” (NFM) perception. We attempt to explicate and further advance the conceptualization and corresponding measurement of “News Finds Me” perception. Using a nationally representative survey conducted in Austria, our analysis shows that social media news use may contribute to cynicism towards politics, and that such a relationship is critically mediated through NFM perceptions, but to a differential degree depending on specific sub-dimensions of NFM perceptions.

Full text

Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=hmcs20
Mass Communication and Society
ISSN: 1520-5436 (Print) 1532-7825 (Online) Journal homepage: https://www.tandfonline.com/loi/hmcs20
Social Media News Use and Political Cynicism:
Differential Pathways Through “News Finds Me”
Perception
Hyunjin Song, Homero Gil de Zúñiga & Hajo G. Boomgaarden
To cite this article: Hyunjin Song, Homero Gil de Zúñiga & Hajo G. Boomgaarden (2019):
Social Media News Use and Political Cynicism: Differential Pathways Through “News Finds Me”
Perception, Mass Communication and Society, DOI: 10.1080/15205436.2019.1651867
To link to this article: https://doi.org/10.1080/15205436.2019.1651867
© 2019 The Author(s). Published with
license by Taylor & Francis Group, LLC
View supplementary material
Accepted author version posted online: 02
Aug 2019.
Published online: 16 Aug 2019.
Submit your article to this journal
Article views: 53
View Crossmark data

Social Media News Use and Political
Cynicism: Differential Pathways Through
“News Finds Me”Perception
Hyunjin Song , Homero Gil de Zúñiga , and
Hajo G. Boomgaarden
Department of Communication
University of Vienna
As citizens inadvertently “encounter”news and political information through digital
media and social networking sites, they might perceive themselves to be well informed
about politics without actively seeking political information, which has been labeled as
the “News Finds Me”(NFM) perception. We attempt to explicate and further advance
the conceptualization and corresponding measurement of “News Finds Me”perception.
Hyunjin Song (Ph.D, The Ohio State University, 2015) is an assistant professor in the Department
of Communication at the University of Vienna. His research interests include the influence of media
use and interpersonal discussion on political engagement and computational methods.
Homero Gil de Zúñiga (Ph.D, University of Wisconsin, Madison, 2008) is Medienwandel
Professor in the Department of Communication at the University of Vienna. His research interests
include 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.
Hajo G. Boomgaarden (Ph.D, University of Amsterdam, 2007) is Professor for Empirical Social
Science Methods in the Department of Communication at the University of Vienna. His research
interests include the coverage and effects of political information on citizens’cognitions, attitudes and
behaviors in various domains of media and politics, and developments in automated content analysis
techniques.
Correspondence should be addressed to Hyunjin Song, Department of Communication, University
of Vienna, Rathausstraße 19/1/9, Vienna, 1010 Austria. E-mail: hyunjin.song@univie.ac.at
Color versions of one or more of the figures in the article can be found online at www.
tandfonline.com/HMCS.
This is an Open Access article distributed under the terms of the Creative Commons Attribution-
NonCommercial-NoDerivatives License (http://creativecommons.org/licenses/by-nc-nd/4.0/), which
permits non-commercial re-use, distribution, and reproduction in any medium, provided the original
work is properly cited, and is not altered, transformed, or built upon in any way.
Mass Communication and Society, 0:1–24
© 2019 The Author(s). Published with license by Taylor & Francis Group, LLC
ISSN: 1520-5436 print / 1532-7825 online
DOI: https://doi.org/10.1080/15205436.2019.1651867
1

Using a nationally representative survey conducted in Austria, our analysis shows that
social media news use may contribute to cynicism towards politics, and that such
a relationship is critically mediated through NFM perceptions, but to a differential
degree depending on specific sub-dimensions of NFM perceptions.
With the advance of digital media and social networking sites (SNS) such as Facebook
and Twitter, citizens across the globe are experiencing dramatic changes in the way they
receive political information. Increasingly, those platforms act as primary outlets
through which citizens encounter news stories. This trend is evident across the globe,
such that approximately more than one third –sometimes more than half –of the
citizens of 27 countries around the world use social media for discovering and
consuming news content (Newman, Fletcher, Kalogeropoulos, Levy, & Nielsen, 2018).
Many social media users also use social media to disseminate and discuss news
with others (e.g., Lee & Song, 2017). Inevitably, users within these social networks
will be exposed to news contents without any explicit or intentional effort of their
own (Fletcher & Nielsen, 2018;Kim,Chen,&deZúñiga,2013). While maintain-
ing connections with family and friends remains the primaryreason for using social
media, the increase in news circulation via social media or messaging applications
(e.g., Toff & Nielsen, 2018) may suggest that the chance of people being “inad-
vertently”exposed to and consuming news content becomes all the greater (e.g.,
Valeriani & Vaccari, 2016; Weeks, Lane, Kim, Lee, & Kwak, 2017).
As citizens increasingly encounter news and relevant political information without
any explicit effort through social media platforms, they also might perceive themselves
to be well informed about politics without actively seeking for political information,
which has been labeled as the “News Finds Me”(NFM) perception (Gil de Zúñiga,
Weeks, & Ardèvol-Abreu, 2017; also see Fletcher & Nielsen, 2018). Recent studies
grappling with this concept suggest that high NFM perceptions uniquely –and most
importantly, negatively –relate to citizens’acquisition of political knowledge, political
interest, and voting (Gil de Zúñiga & Diehl, 2019; Valenzuela, Bachmann, & Gil de
Zúñiga, 2018,May)–despite individuals with high NFM perceptions frequently
consuming news from social media.
Against this backdrop, the goal of this contribution is twofold. First, we
attempt to explicate and further advance the conceptualization and the corre-
sponding measurement of the “News Finds Me”perception. A more thorough
conceptual and empirical validation of the NFM scale is necessary, given the
complex nature of the construct. To this end, we propose that the NFM percep-
tion should be considered as a tripartite, higher-order construct consisting of
three subdimensions: (a) “being informed”(epistemic), (b) “not seeking”(moti-
vational), and (c) “reliance on peers”(instrumental dimension). We seek to
(theoretically) investigate and (empirically) confirm this internal factor structure
of the NFM measurement items by exploring its possible correlates and
2SONG, GIL DE ZÚÑIGA, BOOMGAARDEN

predictors, focusing on social media news use and one’s perceptions of their
peers’political expertise.
Second, we explore indirect consequences of social media news use through
NFM perception on individuals’political system-level perception concerning
political cynicism, taking the dimensionality of NFM into account. Political
cynicism is one of the key elements in assessing people’s overall confidence in
government and democratic systems, which subsequently relate to important
political outcomes such as participation and voting. A handful of previous
studies focused on the impact of media use on political cynicism, but so far,
they were largely confined to more active forms of media consumption on
traditional media (e.g., Hutchens, Hmielowski, Pinkleton, & Beam, 2016;
Pinkleton & Austin, 2002). With the notion of NFM perception —in particular
its motivational dimension —suggesting a relatively passive nature of news
consumption, it is largely unknown whether, and how, such a passive mode of
news consumption is related to citizen’s political system-level perceptions.
Given the increasing prominence of social networking sites for online news
consumption and increasing political disengagement across Europe and beyond,
our investigation sheds important insights on how the changing nature of online
news consumption affects citizen’s attitudes and satisfaction towards democratic
engagement in a more nuanced way. While prior contributions on this topic have
examined the (mostly negative) effect of NFM perception on knowledge and parti-
cipation primarily within the U.S. context, we extend this line of work by examining
the possible correlates and its effect of NFM perception within a different political
context (Austria), which has a much more traditional news environment than the U.S.
Using a nationally representative survey conducted in Austria, our analysis shows
that social media news use may contribute to lower levels of cynicism, and that such
a relationship critically depends upon NFM perceptions, but to a different degree
depending on specific sub-dimensions of NFM perceptions.
THE DIMENSIONALITY AND THE ANTECEDENTS OF
NEWS-FINDS-ME PERCEPTION
The News Finds Me (NFM) perception is defined as “[emphasis added] the
extent to which individuals believe they can indirectly stay informed about
public affairs –despite not actively following the news –through general
Internet use, information received from peers, and connections within online
social networks”(Gil de Zúñiga et al., 2017, p. 107). Contrary to avoidance of
(hard) news altogether (Ksiazek, Malthouse, & Webster, 2010) or preferences
toward entertainment contents over hard news (Prior, 2005), the NFM percep-
tion purportedly captures an individual’s idiosyncratic “passive”motivation of
NEWS FINDS ME PERCEPTION 3

consuming news contents, or lack thereof. At first glance, such perception
simply appears to stem from an individual’sepistemic belief that they them-
selves are being “well informed”about current events and political affairs (Gil
de Zúñiga et al., 2017). Yet the close reading of the initial conceptualization of
NFM perception presupposes a very specific set of conditions, compared to the
traditional notion of citizen competence coupled with active information seek-
ing behaviors (e.g., Valentino, Hutchings, Banks, & Davis, 2008).
Besides the epistemic belief that people rating high on NFM are being “well
informed”, another core element of NFM perception is the notion that one needs
not actively follow the news with explicit efforts. This motivational basis of
NFM perception presumably plays an important role in the current conceptua-
lization of NFM perception, highlighting the passive and “incidental”nature of
indirect news exposure within the contemporary high-choice information envir-
onment. In other words, the “incidental”nature of citizens’competence
expressed in NFM perception is squarely captured by this motivational sub-
dimension of NFM perception. In a highly saturated media environment, acquir-
ing politically relevant information may not need much active effort due to
abundant information availability (Elenbaas, Boomgaarden, Schuck, & de
Vreese, 2013; Prior, 2005). In such a setting, passive learning can occur
independently of one’s political motivation and attention to contents (Bode,
2016; Shehata, 2014). Therefore, individuals may believe they can sufficiently
stay informed about public affairs (epistemic “being informed”dimension)
despite not purposefully following news (motivational “not seeking”dimen-
sion). Based on this logic, we predict that individuals’frequent use of social
media for news will be positively related to two sub-dimensions of NFM
perception:
H1: Social media use for news will be positively related to the “being informed”
dimension (H1a) and the “not seeking”dimension (H1b) of NFM perception.
Another important feature of the NFM perception is the notion that people can rely
or delegate the process of acquiring necessary political information to their social
network peers (“reliance on peers”dimension), which captures the instrumental aspect
of NFM perception. Similar to the aforementioned availability-based logic, one can
easily obtain politically relevant information within (online) social networks due to
others sharing such contents. While individuals do not exclusively use social media as
a tool for satisfying their information-seeking and surveillance needs (Lee & Ma,
2012), inevitably they can satisfy such needs if other users within these social networks
share news contents that are relevant for them (Fletcher & Nielsen, 2018; Kim et al.,
2013;Toff&Nielsen,2018). Indeed, studies suggest that citizens are more likely to
engage with those whom they perceive to be able to provide a shortcut to political
4SONG, GIL DE ZÚÑIGA, BOOMGAARDEN

information (Huckfeldt, 2001; Sokhey & McClurg, 2012). This “division of labor”in
political information acquisition has long been recognized as a potential source of
political learning (e.g., Berelson, Lazarsfeld, & McPhee, 1954;Downs,1957). Based
on this logic, we expect that (a) those who perceive their social contacts (“alters”)to
have higher political expertise, and (b) those who heavily use social media for political
purposes may rely more on such delegation in acquiring political information.
Therefore:
H2: Alter political expertise perceptions (H2a) and social media news use (H2b) will
be positively correlated with the “reliance on peers”dimension of NFM perception.
DEMOCRATIC IMPLICATIONS OF NEWS-FINDS-ME PERCEPTION
For decades, citizens have voiced a persistent feeling of skepticism toward and
distrust in government (Kriesi et al., 2012). Individuals may perceive that
politicians deal with important issues in a superficial manner, or take decisions
based on strategic electoral positioning rather than based on the genuine interest
of the electorates. This type of perception leads to what has been coined as
political cynicism, or “a (perceived) gap between voters and their political
representatives”(Schuck, Boomgaarden, & de Vreese, 2013, p. 288), which
(typically) accompanies skepticisms and distrust towards a political system.
There is abundant literature documenting a large degree of political skepticism
towards elected officials among the general public, ranging from distrust in
government and political institutions (Cook & Gronke, 2005) to overall dis-
affection with the liberal democratic principles at the expense of populist
appeals (Fieschi & Heywood, 2004).
While the role of traditional media such as newspapers and television in
promoting political cynicism is now well documented (e.g., Cappella &
Jamieson, 1996; De Vreese & Elenbaas, 2008), some prior evidence (quite
conversely) suggests that increased access to websites of news media and
news organizations may increase trust toward political systems (therefore may
lower political cynicism: Bailard, 2014; Ceron, 2015). These studies generally
argue that increased access to political information afforded by the Internet may
offer better dissemination of information regarding the performances of govern-
ments and institutions to their citizens, therefore is likely to increase democratic
satisfaction (such as increased trust and decreased cynicism) especially in well-
functioning democratic countries (Bailard, 2014; Ceron, 2015). However, the
question of to what extent news exposure on social media in particular may
contribute to political cynicism is still insufficiently addressed. Research has
shown that when the information is shared by their peers, people tend to trust
given information more (Turcotte, York, Irving, Scholl, & Pingree, 2015), and
NEWS FINDS ME PERCEPTION 5

are more likely to engage with such contents even if such contents are counter-
attitudinal (Messing & Westwood, 2014). In addition, research suggests that
those who may not otherwise actively seek out news due to their low level of
interest and motivations (Prior, 2005) indeed often are influenced by mere
exposure to media contents independent of their level of interest and motiva-
tions (e.g., Cappella & Jamieson, 1996; Shehata, 2014). Given increasing trends
in incidental news exposure via social media (Toff & Nielsen, 2018; Valeriani &
Vaccari, 2016), we argue that such inadvertent exposure without explicit effort
may have important ramifications in understanding citizens’confidence towards
politics.
An increasing number of studies suggests that cynicism can be understood
as a function of specific –oftentimes, negative –contents such as negative
advertising and strategic and “game-framed”news (e.g., Cappella &
Jamieson, 1996; De Vreese & Elenbaas, 2008;Schucketal.,2013;
Shehata, 2014). News use therefore tends to increase political cynicism
especially among those who are politically interested, to the extent that
such attentive voters have more exposure to such negative coverage
(Cappella & Jamieson, 1996; De Vreese & Elenbaas, 2008; Shehata, 2014).
Yet past studies have also shown that political cynicism is negatively related
to “active”media consumption and political involvement (e.g., Pinkleton &
Austin, 2002), suggesting a relatively passive form of media use may
promote further political cynicism while active media use –and the percep-
tion of being sufficiently informed for that matter –may be negatively
related to level of political cynicism. This is because peripheral processing
(i.e., lack of active, systematic processing) of game-framed, strategic news
frame automatically activates negative considerations towards politics.
Prior research indicates that more controversial, conflictual, and negative
news stories are relatively more often shared online (e.g., Burggraaff & Trilling,
2017; Trilling, Tolochko, & Burscher, 2017), and relatedly, relatively passive
and effortless processing of such stories may further activate negative consid-
erations towards politics in general. For instance, Shehata’s(2014) analysis
shows that motivated news attention substantially diminishes negative effects
of game-framed news coverage on political trust and engagement, while Ceron’s
(2015) analysis suggests that an increased news consumption via social media
has a particularly negative consequence for citizens’trust towards political
institutions compared to news consumption via online news media (such as
websites of traditional news organizations). Since the two subdimensions of
NFM perception –“not seeking”and “reliance on peers”–represent relatively
passive, incidental consumption of (often negative and conflictual) news stories
on social networking websites without active engagement, this is one of the
main reasons why political cynicism and NFM should be positively associated.
6SONG, GIL DE ZÚÑIGA, BOOMGAARDEN

Since those who lack motivation to actively engage with news are likely to
effortlessly process negative and sensational news stories being available to
them via online social networks, we therefore may expect that “not seeking”and
“reliance on peers”dimensions of NFM perception, which represent relatively
passive and effortless motivations of consuming news contents, should further
promote political cynicism. Yet on the other hand, to the extent that individuals
perceive themselves to already be sufficiently informed about public affairs,
such individuals would be less likely to show cynicism towards political
systems –similar to those who are on top of the news, vigilant, always attentive
and functioning as opinion and information “watchdogs.”This implies the
“being informed”dimension of NFM perception would be negatively related
to political cynicism. Additionally, considering the fact that social media use
would positively predict NFM perception, we expect NFM perceptions to
mediate the influence of social media use on political cynicism. Therefore, we
posit:
H3: While “reliance on peers”(H3a) and “not seeking”(H3b) dimensions will
positively predict political cynicism, “being informed”(H3c) dimension will
negatively predict political cynicism.
H4: Social media use will indirectly increase political cynicism through “reliance
on peers”(H4a) and “not seeking”(H4b) dimension, whereas decrease political
cynicism through stay informed”(H4c) dimension of NFM perception.
METHODS
Sample
The study relies on a nationally representative survey conducted in Austria,
fielded from September 15th to October 31st, 2017. The survey was con-
ducted and administered by Statistics Austria, an independent public institu-
tion responsible for providing national-level census and polling. The survey
was commissioned by the Platform for Survey Research, Methods and
Empirical Analysis (PUMA), financed by the Austrian Federal Ministry of
Education, Science and Research. The protocol regarding research on human
subjects were reviewed and approved by PUMA prior to the fieldwork.
Based on a central population register of the Austrian population, Statistics
Austria used multistage stratified random sampling based on key demographics
to draw a total of 1500 representative participants. Eligible participants were
given a mail invitation, where were given a code for online access to the survey
and identical paper-and-pencil questionnaire (either with the initial invitation
code or as a follow-up reminder). Among them, a total of 641 respondents have
NEWS FINDS ME PERCEPTION 7

participated in the survey (AAPOR RR1: 42.7%). Our sample contained slightly
more females (sample = 54% female; population = 51% female) and was
slightly older (sample: mean = 45.65, median = 47; population: median
range = 40 to 44 years) than the general population, but largely representative
in terms of education (both sample and population median = Intermediate
technical and vocational schools) and net household income (sample med-
ian = 2201 to 2600 EUR, population: mean = 2158 EUR).
1
Measurements
News-Finds-Me Perception. The NFM perception was measured
based on agreement (on a 10-point scale ranging from “strongly
disagree”=1to“strongly agree”= 10) with a series of statements from Gil
de Zúñiga et al. (2017). Given that the original construct generated in that
study included four items –two items for reliance on peers, and one item for
informed and not seeking, respectively, this study included two additional
statements that closely capture those single-item sub-dimensions to improve
the measurement (the informed dimension now includes “I do not worry about
keeping up with news because I know news will find me,”and the not seeking
dimension includes “I do not have to actively seek news because when
important public affairs break, they will get to me through social media”).
Accordingly, all three sub-dimensions of the NFM construct have now two
items each. Panel A in Tab le 1 reports item wordings and descriptive statistics.
For subsequent regression analysis (including propensity score weighting), we
constructed scales for each sub-dimension (Reliance on peers: M = 3.73,
SD = 2.30, Spearman-Brown ρ= .658; Informed: M = 4.60, SD = 2.43,
ρ= .494; Not seeking: M = 4.89, SD =2.39,ρ= .690).
2
Social Media News Use. Following Gil de Zúñiga et al. (2017), social
media use for news purpose was captured as an index of eight items measured
on a 10-point scale (1 = never to 10 = all the time, M= 3.67, SD = 1.91,
α= .76). Five items asked about respondents’exposure to news through
Facebook, YouTube, WhatsApp, Twitter, and Google (“how often you use
1
All reported population characteristics are from 2015 labor market statistics, provided by the
Federal Bureau of Statistics Austria.
2
Some of the item wordings in NFM scale appears to address two subdimensions at the same
time. To address this issue, we validate our main analysis by using extracted factor scores from EFA
(instead of a scale mean of relevant items) as the score for sub-dimensions. This robustness check,
which appears in Table A5 to A6 of the online appendix, suggests that our multidimensional
conceptualization itself and corresponding results are reasonably robust, although our NFM percep-
tion question wordings are less than ideal.
8SONG, GIL DE ZÚÑIGA, BOOMGAARDEN

following social media [via computer, tablet, mobile, or any device] for getting
news online”). In addition, three more general items (“I use social media to stay
informed about current events and public affairs,”“I use social media to get
news about current events from mainstream media,”and “I use social media to
get news from online news sites,”all of which were also based on the same 10-
point scale) captured their attention and motivations, therefore collectively
tapping general online news use.
3
Political Cynicism. Political cynicism is captured by six items (on
a 5-point scale from “strongly disagree”=1to“strongly agree”=5)asking
respondents’attitudes towards elected politicians in general (M=2.28,SD = .64,
TABLE 1
Question Wordings and Model Fit Comparisons
Panel A. Question Wording for NFM Perception and Its Sub-dimensions (N= 641)
Items Dimension Item wordings M(SD)
1Reliance on peers I rely on my friends to tell me what’s important
when news happen
4.57 (2.95)
2Informed I can be well-informed even when I don’t actively
follow the news
5.28 (2.73)
3Informed I do not worry about keeping up with news because
I know news will finds me
4.55 (2.75)
4Reliance on peers I rely on information from my friends based on
what they like or follow through social media
2.92 (2.38)
5Not seeking I do not have to actively seek news because when
important public affairs break, they will get to me
in social media
4.04 (3.08)
6Not seeking I’m up-to-date and informed about public affairs
news, even when I do not actively seek news
myself
5.13 (2.88)
Panel B. A Comparison of the Single Factor Model Vs. Three-factor Model of NFM
Models Chisq (df) CFI TLI RMSEA [90% CIs] SRMR
Single factor 106.20 (8)*** .872 .759 .139*** [.116, .163] .056
Three-factor 8.71 (5) .995 .985 .029 [0, .071] .016
Note. Model fit comparison yields χ
2
-diff = 97.48, df =3,p< .001, showing hypothesized three-
factor model fits significantly better to the data than the single-factor model. All model N= 638.
3
The combined measure of exposure and attention is regarded as one of the best approaches of
measuring a general concept of media “use,”which reflects the correspondence between exposure to
a source and devotion of cognitive effort to that source at the same time (Eveland, Hutchens, &
Shen, 2009).
NEWS FINDS ME PERCEPTION 9

α= .74), adopted from Boukes and Boomgaarden (2015). Items include:
“Politicians often argue too much with each other,”“MPs and ministers do not
care much about what people like me think,”“Politicians are particularly
interested in their own interest,”“Politicians spend too much time discussing,
and too little governing,”“Most public officials are truly interested in what the
people think”(recoded), and “Most politicians are competent people who know
what they are doing”(recoded).
Alter Political Expertise Perceptions. Respondents were also asked
about the perception of each of their social network ties’political expertise
levels (based on a 5-point scale from “not at all knowledgeable”=1to“very
much knowledgeable”= 5) for strong and weak ties. A summary index of
alter political expertise perception was derived by averaging the two
(M= 3.14, SD = .64, Spearman-Brown ρ= .630).
4
Control Variables. In order to account for potential confounders, a host of
control variables were included in the analysis, including respondents’gender
(54% female, coded as 1, otherwise zero), age (in 10-year increments, M= 4.55,
SD = 1.65), and education attainment (Median =7[“Teaching or vocational
school”], range = 1 “from no school attended”to 14 “Doctor/PhD”). We control
for talk frequency (M= 2.72, SD = .87, Spearman-Brown ρ= .684), which was
assessed by asking how often they talk about politics with their strong (i.e.,
family and close friends) and weak ties (i.e., coworkers, and acquaintances)
whether online or offline (based on a 5-point scale from “never”=1to“very
often”= 5). In addition, we also include political interest (M= 2.63, SD = .87,
from 1 = “not at all interested”to 4 = “very interested”) and internal efficacy
(M= 3.61, SD = 1.21, from 1 = “strongly disagree”to 5 = “strongly agree”),
measured by one item with the following statement: “People like me don’t have
any say in what public officials do”(recoded).
Analysis Strategy
First, we employed a series of confirmatory factor analysis (CFA) to empiri-
cally test the three-dimensional structure of the NFM perception. We do so
by comparing a baseline single-factor model with the proposed three-factor
model using χ
2
test statistics. We further validate the proposed three-factor
model of the NFM perception by examining NFM perception’s convergent
and discriminant validity with other logically similar constructs. Convergent
4
The zero-order correlation between strong vs. weak talk frequency was .52, and the zero-order
correlation between strong vs. weak alter expertise perception was .46.
10 SONG, GIL DE ZÚÑIGA, BOOMGAARDEN

validity was examined via its correlation structure between NFM sub-
dimensions and theoretically related variables, whereas a discriminant valid-
ity is examined by comparing models with and without equality constraints
being imposed between each dimension of the NFM and other logically
similar constructs (Hayes, Glynn, & Shanahan, 2005).
Second, we conducted a series of OLS regression analysis, coupled with
a propensity score weighting analysis (Imai & Ratkovic, 2014) and Monte-Carlo
inferences for mediation (Hayes, 2017; Preacher & Selig, 2012). In order to address
the limitations of cross-sectional data, we employed the Covariate Balancing
Generalized Propensity Score (CBGPS) methodology proposed by Fong, Hazlett,
and Imai (2018; also see Imai & Ratkovic, 2014). The CBGPS weighting attempts
to reduce self-selection bias by reducing correlations between a continuous “treat-
ment”variable and a set of relevant pre-treatment covariates. This can be seen as
a substitute for randomization, where the goal of such a propensity score procedure
is to produce an unbiased comparison between cases that are indistinguishable in
terms of their relevant background characteristics except a“treatment”variable of
interest. When applied to observational data, post-adjusted data (using estimated
propensity score weights) would have similar balancing properties that can be found
in a randomized experiment. We estimate two CBGPS models predicting two
“treatment”variables (i.e., social media news use and NFM perception, respec-
tively) and subsequently utilize CBGPS weights in the outcome analysis (see below
for more detail). Given a host of other demographic and political correlates of the
NFM perception, and for the ease of application of propensity score weighting, we
opted to rely on OLS regression-based procedures rather than structural equation
models. All analyses were conducted using Rstatistical software, with Lavaan
(Rosseel, 2012)andCBPS package (Fong, Ratkovic, & Imai, 2018).
RESULTS
Confirmatory Factor Analysis of News Finds Me Sub-dimensions
First, we compare the model fit between the single-factor CFA model (assuming
all observed variables belong to the single “NFM”latent construct) and that of the
multi-factor CFA model.
5
In the single-factor CFA model, covariances among
sub-dimensions are fixed to 1 (therefore all sub-dimensions are treated as the
5
We allow residuals between the items mentioning social media (item 4 and item 5 in NFM
scales) to be correlated, in order to capture any residual covariation due to question wordings. Also,
we scale the variances of the latent construct to be 1 (with their residual error variances set to zero),
so any covariance estimates between latent constructs are to be interpreted as correlations among
those latent factors.
NEWS FINDS ME PERCEPTION 11

single latent factor), whereas in the multi-factor CFA model, such constraints are
released. Although both the single-factor model and the three-factor model indeed
showed acceptable model fit (as can be seen in Panel B of Table 1), the difference
in the model fit statistics between the single-factor and the hypothesized three-
factor model was statistically significant, χ
2
-diff = 97.48, df =3,p< .001,
confirming that the three sub-dimensions of NFM perceptions are unique.
As an initial examination of the validity, we further assessed within- and
across-correlations of each NFM scale item with the three sub-dimensions. As
can be seen in Table A1 of the online appendix, each observed variable that
belongs to each of the sub-dimensions most strongly correlated with that
specified sub-dimension but not with others, further supporting the appropriate-
ness of our three-dimensional structure of NFM.
Convergent and Discriminant Validity of News Finds Me Sub-dimensions
Having established the appropriateness of the three-factor structure of NFM
perception, we now assess their additional evidence of convergent and divergent
validity. As previously noted, convergent and discriminant validity was exam-
ined via (a) the correlations between latent variables of NFM sub-dimensions
with each of the related latent variables, and (b) model fit comparisons between
a single-factor vs. a two-factor model.
First, we computed the bivariate latent covariance among latent variables
(with their error variance constrained to be 1, so this is interpreted as correla-
tions), as reported in Panel A of Table A2 in the online appendix. The results
show that, consistent with prior research on this topic (e.g., Gil de Zúñiga et al.,
2017), social media news use was positively correlated with all of the sub-
dimensions of NFM perception. Though, at the same time, the result also
suggested that there are some noticeable differences in their magnitudes of zero-
order correlations across three sub-dimensions. Social media news use was most
strongly correlated with the “not seeking”dimension (Φ= .769), but to a lesser
extent with the “reliance on peers”dimension (Φ= .361), and least strongly
with the “being informed”dimension (Φ= .124, all p< .05). This is a particu-
larly noteworthy pattern, given that bivariate correlations of the three NFM sub-
dimensions are very much similar to each other (all Φs about .5 to .6 range).
Also, the “being informed”dimension was negatively correlated with cynicism
(Φ=−.258, p< .001), which was indeed expected based on the theoretical
meanings of the two constructs. As such, results showed initial evidence of
convergent validity of our multidimensional conceptualization of NFM percep-
tion, although we would be cautious, as this result does not correct for self-
selection bias in cross-sectional data.
12 SONG, GIL DE ZÚÑIGA, BOOMGAARDEN

Next, the covariance between latent variables (again with error variances set
to zero) was constrained to 1 to yield a constrained model (this is equivalent to
specifying the single-factor model, treating each of the sub-dimensions of NFM
to be equal to each of the related latent variables). Subsequently, differences in
fit statistics between the single-factor and two-factor models (i.e., the model that
freely estimates covariance without equality constraints) were computed, with 1
df for χ2-difference tests, as reported in Panel B of Table A2 in the online
appendix. A substantial improvement in fit between the single-factor and two-
factor model suggested that, although they are meaningfully correlated, the
latent variables are unique to each other, thereby tapping different constructs.
Coupled with earlier results showing that each of NFM subdimensions is
correlated with the other variables differently, this provides evidence of dis-
criminant validity.
Effects of Social Media Use on Cynicism Through NFM: Propensity Score
Weighting
Having established initial evidence of convergent and divergent validity, we now
move to the correlates and its possible consequences of NFM perception. Since
we are interested in effects of social media news use on three NFM sub-
dimensions, and in turn effects of those sub-dimensions on political cynicism,
we estimated two CBGPS models predicting continuous “treatment”variable,
social media news use and NFM perception respectively, as a function of the host
of pre-treatment covariates outlined above. Figure 1 below reports the covariate
balance (i.e., absolute values of Pearson correlations between a “treatment”
variable and a set of covariates) before (each upper panel) and after (each lower
panel) weighting. As can be seen in Figure 1, the estimated CBGPS weights
substantially reduced the degree of imbalance in the data, such that the maximum
covariate imbalance in post-weighted data does not exceed 0.1 (which is an
adequate level of covariate balance in continuous treatment cases: Zhu,
Coffman, & Ghosh, 2015). Resulting CBGPS weights were then further utilized
in outcome analysis estimating NFM perception (mediator) and political cynicism
(the dependent variable), and later in the Monte-Carlo simulation method for
estimating mediation effects. Detailed information regarding the propensity score
weighting procedures are reported in Table A3 of the online appendix (we also
present side-by-side results of identical models without propensity score weight-
ing in Table A4 of the online appendix). Table 2 below reports the results adjusted
for propensity score weighting, which produces more accurate (and often more
conservative) estimates compared to typical OLS estimates.
Our first set of hypotheses (H1a and H1b) predicted that social media use for
news would be positively associated with “being informed”and “not seeking”
NEWS FINDS ME PERCEPTION 13

sub-dimensions of the NFM perception. These expectations received strong
support by our data, in that social media news use significantly and positively
predicted the informed (H1a: b= .175, SE = .053), not seeking (H1b: b= .541,
SE = .047), and the reliance on peers dimension as well (H2b: b= .324,
FIGURE 1 Covariate balance plots. The “treatment”variable in upper panel is social media news
use, and the “treatment”variable in lower panel is NFM perception. The figures show that covariate
balance (in terms of absolute Person correlation between treatment variable and the pre-treatment
covariates) is substantially improved by CBPS weighting, such that none of the pre-treatment
covariates are correlated higher than 0.1 with the treatment variable after weighting. A detailed
information about each covariate balance statistics is reported in Table A3 of the online appendix.
14 SONG, GIL DE ZÚÑIGA, BOOMGAARDEN

SE = .049, all p<. 001), even after controlling for a set of covariates (such as
political interest, internal efficacy, and demographics) and possible self-
selection bias (via propensity score weighting). Additionally, H2a predicted
that alter political expertise would be significantly related to the reliance on
peers dimension, which is strongly supported by our data (H2a: b= .478,
SE = .164, p< .01).
Besides those hypothesize effects, a few results deserve further mention.
First, we found that education and political interest all strongly and nega-
tively predict all three NFM sub-dimensions. Given the passive nature of the
NFM perception, these results indeed make sense, in that more educated and
interested citizens tend to actively consume political information (Ksiazek
et al., 2010; Prior, 2005). Furthermore, we observe that, in addition to the
hypothesized reliance on peers dimension, alter political expertise is also
positively related to the being informed (b=.473,SE =.175,p<.01)but
not to the not seeking dimension (b= .088, SE = .158, p>.05).When
surrounded with knowledgeable peers, one may also come to believe that
they themselves are also sufficiently informed as a result of relying on such
contacts in acquiring political information; but it seems that merely being
TABLE 2
Predictors of NFM Perception and Political Cynicism, Using Post-CBGPS Weighting
(Standard Errors in Parentheses)
NFM reliance NFM informed NFM not seeking DV cynicism
(Intercept) 3.255 (.738)*** 4.908 (.788)*** 2.785 (.711)*** 3.465 (.210)***
Focal predictors
Social media news use .324 (.049)*** .175 (.053)*** .541 (.047)*** −.035 (.014)*
Alter expertise .478 (.164)** .473 (.175)** .088 (.158) −.135 (.042)**
Mediators
NFM - Reliance .033 (.013)**
NFM - Informed −.038 (.012)**
NFM - Not seeking .011 (.012)
Controls
Interest −.429 (.145)** −.592 (.154)*** −.345 (.139)* .108 (.038)**
Discussion freq −.092 (.144) −.100 (.154) .104 (.139) −.030 (.038)
Internal efficacy −.085 (.081) .085 (.087) −.013 (.078) −.195 (.022)***
Demographics
Gender (Female = 1) .135 (.193) .144 (.206) .178 (.186) .060 (.051)
Education −.086 (.036)* −.139 (.039)*** −.151 (.035)*** .004 (.009)
Age (in 10 years) −.020 (.062) .073 (.066) .284 (.059)*** −.042 (.018)*
R
2
(Adjusted R
2
) .125 (.112) .117 (.103) .276 (.265) .213 (.197)
Number of observations 532 532 532 532
RMSE 0.094 0.100 0.090 0.025
Note. Significant unstandardized coefficients are bolded. ***p< .001, **p< .01, *p< .05.
NEWS FINDS ME PERCEPTION 15

surrounded by such contacts does not lower one’s motivation for active
information seeking.
Turning to the impact of each NFM sub-dimension on political cynicism, the
results show that while the being informed dimension (H3c: b=−.038,
SE = .012, p< .01) is negatively associated with political cynicism, the reliance
on peers dimension (H3a: b= .033, SE = .013, p< .01) is positively associated
with political cynicism. The not seeking dimension did not reach the conven-
tional level of statistical significance as predicted by H3b.
To better understand the consequences of different dimensions of NFM
perception on political cynicism (and the value of such a multidimensional
conceptualization over a monotonic conceptualization), we conducted Monte
Carlo (MC) method-based mediation inferences as described in Hayes (2017)
and in Preacher and Selig (2012). The MC method is a particularly appropriate
choice here since it can easily accommodate complex mediation models that
may take a long time to converge or may randomly generate a constant variance
of test statistics when directly bootstrapped (such as multilevel models or
regression models with propensity score weights as here). We take parameter
estimates and their asymptotic variances and covariance from regression models
reported in Table 2, and then generated the joint distribution of indirect effects
from 20,000 replicated draws to construct 95% confidence intervals for indirect
effects. The results are reported in Table 3 below.
When we look at the simple mediation model with a single composite scale of
NFM as a mediator, we find that NFM perception as a whole is not significantly
associated with cynicism (indirect effect =−.0003, 95% MC CIs = [−.0101, .0093]).
Yet when we consider potentially differential pathways by utilizing three sub-
dimensions separately, we find that the “reliance on peers”dimension (indirect
TABLE 3
Estimated Indirect Effects of Social Media News Use on Cynicism through NFM
Perception Dimensions and Their 95% Distributions from Monte Carlo Methods
DV: Political Cynicism
IV: Social media news use NFM Composite NFM Sub-scales
Direct effect −.0251 (.0135) −.0352 (.0142)
Indirect effect
NFM: Composite −.0003 [−.0101, .0093]
Reliance .0057 [.0025, .0200]
Being informed −.0065 [−.0133, −.0016]
Not Seeking .0018 [−.0071, .0191]
Note. Unstandardized coefficients in bold are significant at p< .05 level, or 95% confidence
intervals does not include zero. Standard errors in parentheses (for direct effects), and 95%
Confidence Intervals from Monte Carlo simulation methods in brackets (for indirect effects).
16 SONG, GIL DE ZÚÑIGA, BOOMGAARDEN

effect = .0057, 95% MC CIs = [.0025, .0200]) and “being informed”dimension
(indirect effect =−.0065, 95% MC CIs = [−.0133, −.0016]) significantly mediate
the influence of social media use on cynicism –and more importantly, in the
opposite direction to each other. While this confirms H4a and H4c, we do not
find compelling evidence that the “not seeking”dimension significantly mediates
the influence of social media use (H4b). Nevertheless, the overall pattern suggests
that different sub-dimensions of the NFM work against each other in predicting
political cynicism, and thus, that the use of a single composite measure of NFM
perception may actually mask an important yet much nuanced relationship between
social media use and political cynicism.
DISCUSSION AND CONCLUSION
Following previous contributions on this topic, our study examined what is
called the “News Finds Me”perception (Gil de Zúñiga et al., 2017). We
specifically focused on its scale refinement in terms of conceptual clarifications
by proposing –and empirically establishing –that the NFM should be consid-
ered as a tripartite, higher-order construct consisting of “being informed”(epis-
temic), “not seeking”(motivational), and “reliance on peers”(instrumental)
sub-dimensions. We then systematically examined its correlates such as social
media use for news and one’s social network characteristics, and further uncov-
ered different effects of three sub-dimensions of the NFM perception on citi-
zens’system-level beliefs focusing on political cynicism.
Representing a detachment from more active surveillance needs, the notion of
NFM suggests that incidental exposure to news, mainly through day-to-day social
media use and through interpersonal interactions with their peers, creates the
perception that they are already sufficiently informed about important political
problems despite not actively seeking such information (e.g., Toff & Nielsen,
2018). Largely focusing on the U.S population, previous studies found that social
media news use positively contributes to the development of the NFM perception
(e.g., Gil de Zúñiga & Diehl, 2019; Gil de Zúñiga et al., 2017). Our study not only
directly corroborate such previous findings in a different geographical context
with a considerably different media landscape and internet news use in general
(the U.S vs. Austria), but also found that social media news use was significantly
correlated with all of the hypothesized sub-dimensions of NFM perception.
Besides, our study shows that one’s network characteristics also importantly
shape NFM perception; individuals are likely to hold a higher NFM perception
when they also perceive themselves surrounded with their social network peers
who are politically knowledgeable and interested. Consistent with the theoretical
perspective that such a “passive”mode of news use may further promote negative
NEWS FINDS ME PERCEPTION 17

political consequences, we also found that social media news use may increase
political cynicism through the “reliance on peer”dimension. Through careful
inspection of the dimensionality of NFM perception, our study therefore con-
tributes to a more nuanced understanding of the relationship between social media
news use, one’s social network characteristics, and their implications on demo-
cratic perceptions through NFM perception.
Despite the fact that inadvertent exposure to political information on social
media makes citizens feel confident in keeping up with current affairs, previous
studies on this topic suggested that relying on such a passive, incidental mode of
news consumption without active involvement does not compensate for the lack
of active use of traditional media –those who incidentally encounter news
contents without active engagements are actually less likely to be knowledge-
able about politics, and for that matter, less likely to actively participate in
political processes (e.g., Gil de Zúñiga & Diehl, 2019; Gil de Zúñiga et al.,
2017). This may suggest the possibility that those who score higher on NFM
would have higher levels of cynicism as well. However, we instead found that
perceptions of being informed –one of the sub-dimensions of NFM –actually
decrease citizen’s political cynicism. As its positive effect on cynicism suggests,
“passive”news consumers may not necessarily become more detached from
politics, but may play an active role in democracy despite their illusion of being
informed. These contradicting patterns further suggest the multidimensional
nature of “News Finds Me Perception”and its complex implications for demo-
cratic engagement.
Furthermore, the pattern suggests that the “reliance on peers”dimension
increases cynicism, which highlights a much broader implication of the NFM
perception. While this dimension captures the passive nature of information
“encounter”strategies, it additionally represents “delegation”of deciding what
is important and meaningful for individuals. Traditionally, such peer delegation –
or a general reliance on indirect information seeking via opinion leaders or peers
from one’s social network –has been regarded as an efficient and costless
shortcut for information acquisition, enabling those who are not that interested
in politics to behave as if they are fully informed (e.g., Downs, 1957; Sokhey &
McClurg, 2012). Our analysis suggests that instead of enabling those who are
less interested in politics to meaningfully participate in politics, such delegation
might actually result in a further detachment from the political process –not
only in terms of political knowledge acquisition (as evidenced in Gil de Zúñiga
& Diehl, 2019; Gil de Zúñiga et al., 2017) but also in terms of increasing
political cynicism, as documented in this study.
This empirical pattern in particular seems to suggest possible negative con-
sequences of the NFM perception may not necessarily be driven by the “not
seeking”dimension, but is rather based on the “reliance on peers”dimension.
18 SONG, GIL DE ZÚÑIGA, BOOMGAARDEN

Based on our own results here and on some recent qualitative evidence on this
topic (e.g., Toff & Nielsen, 2018), we suppose that the “not seeking”dimension
is rather a consequence of both the “reliance on peers”and “informed”dimen-
sions –because I feel confident that I am sufficiently informed by such
information delegation, I think I need not further look at what is needed. This
observation further suggests that the specific consequences of NFM perception
may pan out very differently based on the internal dynamics of NFM sub-
dimensions (e.g., when people perceive that they are not sufficiently informed
despite the fact that their friends in social networks provide all necessary
information). In other words, we suspect that a specific pattern of interaction
between those two sub-dimensions is likely to be key in understanding the
particular consequences of NFM perception.
The present findings also suggest the possibility of “channel”differences (e.g.,
through online communities vs. through the use of news aggregators and websites
of news organizations) in the Internet’s effect on political cynicism. As suggested
by Ceron (2015), while the exposure to news through online communities –mainly
through “encountering”news shared by one’s peers, much like the use of social
media –may represent a “passive”mode of consuming news, using a portal or the
websites of news organizations may still involve an active mode of news use. Our
results partly collaborate this perspective, such that passive forms of news use (i.e.,
“reliance on peers”) in particular, indeed has a negative consequence for cynicism.
Further, this suggests that other forms of using social media may indeed have
distinct consequences through NFM perception. For instance, consuming “ambi-
ence news”within the current context of second screening –using multiple devices
and screens to interact with information at the same time such as news on TV, on
smartphones, and/or laptops (Gil de Zúñiga & Liu,2017, Barnidge, Diehl, & Rojas,
2019;Chadwick,O’Loughlin, & Vaccari, 2017)–may lead to similar conse-
quences. In such a context, people may feel they are well informed, and can rely
on peers to be informed, although they may feel they still need to actively seek news
to further verify facts, discuss public affairs issues, or simply to learn more about
political issues. This theoretical suggestion, of course, remains empirically
untested; future research is therefore advised to further explore this possibility.
Methodologically, our analysis also uncovers more nuanced patterns of
association between sub-dimensions of NFM and their consequences. We
demonstrate that separately considering the three NFM sub-dimensions contri-
butes to a much better understanding of the impact of social media. While
relying on a single composite scale may reasonably work well if a researcher is
interested in its monotonic relationship with other variables (such as political
knowledge: Gil de Zúñiga et al., 2017), doing so may mask important yet
nuanced dynamics among those different sub-dimensions, especially when
there are theoretical reasons to suspect such differential associations.
NEWS FINDS ME PERCEPTION 19

We acknowledge that a single-country, cross-sectional data with a rather
small sample size does not permit us to make very strong tests of our hypoth-
eses, and is particularly unable to completely resolve the issue of endogeneity,
although our use of propensity score weighting may reasonably alleviate such
concerns. While prior evidence on this topic suggests the specific chain of
causality from social media use to NFM perception on the one hand, and
from the NFM perception to its consequences (e.g., Gil de Zúñiga et al.,
2017; Valenzuela et al., 2018, May), establishing a robust causal evidence is
a tricky endeavor, especially when our measurement of the key concept relies on
participants’retrospective self-reports. Future work on this topic is therefore
advised to rely on cross-national, longitudinal data with better measurement
techniques to robustly explore the antecedents and consequences of NFM
perceptions, and its implications for democracy.
The limited number of measurement items per sub-dimension (i.e., two
per sub-dimension) also provided us a less-than-ideal setup for rigorously
testing our expectations. Given the complicated nature of the NFM construct,
future studies are advised to employ qualitative, in-depth interviews (e.g.,
Toff & Nielsen, 2018) or cross-national comparisons (e.g., Fletcher &
Nielsen, 2018) in the development of more comprehensive sets of measure-
ment items and advancing possible additional sub-dimensions. In addition,
exploring possible consequences of NFM on different system-level beliefs,
such as trust towards various political institutions, would be a fruitful endea-
vor for future studies.
Lastly, in establishing convergent and divergent validity, we have relied on
a somewhat limited set of variables (that were also used later on the main
analysis) rather than utilizing more directly related yet auxiliary constructs such
as avoidance of news, incidental news motivation, or in general attention and
elaboration during news exposure. While the nature of our survey setting has
practically limited us to a small set of variables available in our dataset, our
results nevertheless provide very initial evidence towards establishing the multi-
dimensionality of the NFM construct. Future studies will be benefitted from the
use of more theoretically related auxiliary constructs in further clarifying the
nature and meanings of the NFM perception and its implications on contem-
porary news environments.
News media has long been the core mechanism of how citizens acquire and
make sense of political information, yet an increasing trend of inadvertent
exposure, at the expense of direct and active engagements with conventional
forms of news, brings about new opportunities and unforeseen challenges to
a healthy democratic citizenry. Given social media’s likely rising significance in
terms of news provision in the future, it is important to understand cognitive and
behavioral implications of such news consumption patterns.
20 SONG, GIL DE ZÚÑIGA, BOOMGAARDEN

FUNDING
The author(s) disclosed receipt of the following financial support for the
research, authorship, and/or publication of this article: This work was supported
and funded by the Platform for Surveys, Methods and Empirical Analyses
(PUMA), financed by the Austrian Federal Ministry of Education, Science
and Research. The funders had no role in study design, data collection and
analysis, decision to publish, or preparation of the manuscript. Any opinions,
findings, and conclusions or recommendations expressed in this material are
those of the author(s) and do not necessarily reflect the views of the PUMA and
Austrian Federal Ministry of Education, Science and Research.
SUPPLEMENTAL DATA
Supplemental data for this article can be accessed here.
ORCID
Hyunjin Song http://orcid.org/0000-0001-7752-3035
Homero Gil de Zúñiga http://orcid.org/0000-0002-4187-3604
Hajo G. Boomgaarden http://orcid.org/0000-0002-5260-1284
REFERENCES
Bailard, C. S. (2014). Democracy’s double-edged sword: How Internet use changes citizens’views
of their government. Baltimore, MD: Johns Hopkins University Press.
Barnidge, M., Diehl, T., & Rojas, H. (2019). Second screening for news and digital divides. Social
Science Computer Review,37(1), 55–72. doi:10.1177/0894439317750840
Berelson, B. R., Lazarsfeld, P. F., & McPhee, W. N. (1954). Voting: A study of opinion formation in
a presidential campaign. Chicago, IL: University of Chicago Press.
Bode, L. (2016). Political news in the news feed: Learning politics from social media. Mass
Communication and Society,19,24–48. doi:10.1080/15205436.2015.1045149
Boukes, M., & Boomgaarden, H. G. (2015). Soft news with hard consequences? Introducing
a nuanced measure of soft versus hard news exposure and its relationship with political
cynicism. Communication Research,42, 701–731. doi:10.1177/0093650214537520
Burggraaff, C., & Trilling, D. (2017). Through a different gate: An automated content analysis of
how online news and print news differ. Journalism. Advance online publication. doi:10.1177/
1464884917716699
Cappella, J. N., & Jamieson, K. H. (1996). News frames, political cynicism, and media cynicism.
The Annals of the American Academy of Political and Social Science,546,71–84. doi:10.1177/
0002716296546001007
NEWS FINDS ME PERCEPTION 21

Ceron, A. (2015). Internet, news, and political trust: The difference between social media and online
media outlets. Journal of Computer-Mediated Communication,20, 487–503. doi:10.1111/
jcc4.2015.20.issue-5
Chadwick, A., O’Loughlin, B., & Vaccari, C. (2017). Why people dual screen political debates and
why it matters for democratic engagement. Journal of Broadcasting & Electronic Media,61(2),
220–239. doi:10.1080/08838151.2017.1309415
Cook, T. E., & Gronke, P. (2005). The skeptical American: Revisiting the meanings of trust in
government and confidence in institutions. The Journal of Politics,67, 784–803. doi:10.1111/
j.1468-2508.2005.00339.x
De Vreese, C. H., & Elenbaas, M. (2008). Media in the game of politics: Effects of strategic
metacoverage on political cynicism. The International Journal of Press/Politics,13, 285–309.
doi:10.1177/1940161208319650
Downs, A. (1957). An economic theory of political action in a democracy. Journal of Political
Economy,65, 135–150. doi:10.1086/257897
Elenbaas, M., Boomgaarden, H. G., Schuck, A. R., & de Vreese, C. H. (2013). The impact of media
coverage and motivation on performance-relevant information. Political Communication,30,
1–16. doi:10.1080/10584609.2012.737411
Eveland, W. P., Jr, Hutchens, M. J., & Shen, F. (2009). Exposure, attention, or “use”of news?
Assessing aspects of the reliability and validity of a central concept in political communication
research. Communication Methods and Measures,3, 223–244. doi:10.1080/19312450903378925
Fieschi, C., & Heywood, P. (2004). Trust, cynicism and populist anti-politics. Journal of Political
Ideologies,9, 289–309. doi:10.1080/1356931042000263537
Fletcher, R., & Nielsen, R. K. (2018). Are people incidentally exposed to news on social media?
A comparative analysis. New Media & Society,20, 2450–2468. doi:10.1177/1461444817724170
Fong, C., Hazlett, C., & Imai, K. (2018). Covariate balancing propensity score for a continuous
treatment: Application to the efficacy of political advertisements. The Annals of Applied
Statistics,12, 156–177. doi:10.1214/17-AOAS1101
Fong, C., Ratkovic, M., & Imai, K. (2018). CBPS: Covariate balancing propensity score. R package
version 0.19. Retrieved from: https://CRAN.R-project.org/package=CBPS
Gil de Zúñiga, H., & Diehl, T. (2019). News finds me perception and democracy: Effects on
political knowledge, political interest, and voting. New Media & Society,21, 1253–1271.
doi:10.1177/1461444818817548
Gil de Zúñiga, H., & Liu, J. H. (2017). Second screening politics in the social media sphere:
Advancing research on dual screen use in political communication with evidence from 20
countries. Journal of Broadcasting & Electronic Media,61(2), 193–219. doi:10.1080/
08838151.2017.1309420
Gil de Zúñiga, H., Weeks, B., & Ardèvol-Abreu, A. (2017). Effects of the news-finds-me perception
in communication: Social media use implications for news seeking and learning about politics.
Journal of Computer-Mediated Communication,22, 105–123. doi:10.1111/jcc4.2017.22.issue-3
Hayes, A. F. (2017). Introduction to mediation, moderation, and conditional process analysis:
A regression-based approach (2nd ed.). New York, NY: The Guilford Press.
Hayes, A. F., Glynn, C. J., & Shanahan, J. (2005). Willingness to self-censor: A construct and
measurement tool for public opinion research. International Journal of Public Opinion Research,
17, 298–323. doi:10.1093/ijpor/edh073
Huckfeldt, R. (2001). The social communication of political expertise. American Journal of
Political Science,45, 425–438. doi:10.2307/2669350
Hutchens, M. J., Hmielowski, J. D., Pinkleton, B. E., & Beam, M. A. (2016). A spiral of skepticism?
The relationship between citizens’involvement with campaign information to their skepticism
and political knowledge. Journalism & Mass Communication Quarterly,93,1073–1090.
doi:10.1177/1077699016654439
22 SONG, GIL DE ZÚÑIGA, BOOMGAARDEN

Imai, K., & Ratkovic, M. (2014). Covariate balancing propensity score. Journal of the Royal
Statistical Society: Series B (statistical Methodology),76, 243–263. doi:10.1111/rssb.12027
Kim, Y., Chen, H. T., & de Zúñiga, H. G. (2013). Stumbling upon news on the Internet: Effects of
incidental news exposure and relative entertainment use on political engagement. Computers in
Human Behavior,29, 2607–2614. doi:10.1016/j.chb.2013.06.005
Kriesi, H., Grande, E., Dolezal, M., Helbling, M., Höglinger, D., Hutter, S., & Wüest, B. (2012).
Political conflict in western Europe. Cambridge, UK: Cambridge University Press.
Ksiazek, T. B., Malthouse, E. C., & Webster, J. G. (2010). News-seekers and avoiders: Exploring
patterns of total news consumption across media and the relationship to civic participation.
Journal of Broadcasting & Electronic Media,54, 551–568. doi:10.1080/08838151.2010.519808
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. doi:10.1016/j.chb.2011.10.002
Lee, J., & Song, H. (2017). Why people post news on social networking sites: A focus on
technology adoption, media bias, and partisanship strength. Electronic News,11,59–79.
doi:10.1177/1931243117692084
Messing, S., & Westwood, S. J. (2014). Selective exposure in the age of social media: Endorsements
trump partisan source affiliation when selecting news online. Communication Research,41,
1042–1063. doi:10.1177/0093650212466406
Newman, N., Fletcher, R., Kalogeropoulos, A., Levy, D. A. L., & Nielsen, R. K. (2018). Reuters
Institute Digital News Report 2018. Retrieved from: https://reutersinstitute.politics.ox.ac.uk/sites/
default/files/digital-news-report-2018.pdf
Pinkleton, B. E., & Austin, E. W. (2002). Exploring relationships among media use frequency,
perceived media importance, and media satisfaction in political disaffection and efficacy. Mass
Communication & Society,5, 141–163. doi:10.1207/S15327825MCS0502_3
Preacher, K. J., & Selig, J. P. (2012). Advantages of Monte Carlo confidence intervals for indirect
effects. Communication Methods and Measures,6,77–98. doi:10.1080/19312458.2012.679848
Prior, M. (2005). News vs. entertainment: How increasing media choice widens gaps in political
knowledge and turnout. American Journal of Political Science,49, 577–592. doi:10.1111/
ajps.2005.49.issue-3
Rosseel, Y. (2012). Lavaan: An R package for structural equation modeling and more. Version
0.5–12 (BETA). Journal of Statistical Software,48,1–36. doi:10.18637/jss.v048.i02
Schuck, A. R., Boomgaarden, H. G., & de Vreese, C. H. (2013). Cynics all around? The impact of
election news on political cynicism in comparative perspective. Journal of Communication,63,
287–311. doi:10.1111/jcom.12023
Shehata, A. (2014). Game frames, issue frames, and mobilization: Disentangling the effects of frame
exposure and motivated news attention on political cynicism and engagement. International
Journal of Public Opinion Research,26, 157–177. doi:10.1093/ijpor/edt034
Sokhey, A. E., & McClurg, S. D. (2012). Social networks and correct voting. The Journal of
Politics,74, 751–764. doi:10.1017/S0022381612000461
Toff, B., & Nielsen, R. K. (2018). “I just Google it”: Folk theories of distributed discovery. Journal
of Communication,68, 636–657. doi:10.1093/joc/jqy009
Trilling, D., Tolochko, P., & Burscher, B. (2017). From newsworthiness to shareworthiness: How to
predict news sharing based on article characteristics. Journalism & Mass Communication
Quarterly,94,38–60. doi:10.1177/1077699016654682
Turcotte, J., York, C., Irving, J., Scholl, R. M., & Pingree, R. J. (2015). News recommendations
from social media opinion leaders: Effects on media trust and information seeking. Journal of
Computer-Mediated Communication,20, 520–535. doi:10.1111/jcc4.2015.20.issue-5
Valentino, N. A., Hutchings, V. L., Banks, A. J., & Davis, A. K. (2008). Is a worried citizen a good
citizen? Emotions, political information seeking, and learning via the internet. Political
Psychology,29, 247–273. doi:10.1111/pops.2008.29.issue-2
NEWS FINDS ME PERCEPTION 23

Valenzuela, S., Bachmann, I., & Gil de Zúñiga, H. (2018). Positive, negative or null effects? A panel
study on political learning from social media. Prague, Czech Republic: International
Communication Association.
Valeriani, A., & Vaccari, C. (2016). Accidental exposure to politics on social media as online
participation equalizer in Germany, Italy, and the United Kingdom. New Media & Society,18,
1857–1874. doi:10.1177/1461444815616223
Weeks, B. E., Lane, D. S., Kim, D. H., Lee, S. S., & Kwak, N. (2017). Incidental exposure, selective
exposure, and political information sharing: Integrating online exposure patterns and expression
on social media. Journal of Computer-Mediated Communication,22, 363–379. doi:10.1111/
jcc4.2017.22.issue-6
Zhu, Y., Coffman, D. L., & Ghosh, D. (2015). A boosting algorithm for estimating generalized
propensity scores with continuous treatments. Journal of Causal Inference,3,25–40. doi:10.1515/
jci-2014-0022
24 SONG, GIL DE ZÚÑIGA, BOOMGAARDEN