Content uploaded by Homero Gil de Zúñiga
Author content
All content in this area was uploaded by Homero Gil de Zúñiga on Mar 04, 2019
Content may be subject to copyright.
https://doi.org/10.1177/1461444818817548
new media & society
1 –19
© The Author(s) 2018
Article reuse guidelines:
sagepub.com/journals-permissions
DOI: 10.1177/1461444818817548
journals.sagepub.com/home/nms
News finds me perception
and democracy: Effects on
political knowledge, political
interest, and voting
Homero Gil de Zúñiga
University of Vienna, Austria
Universidad Diego Portales, Chile
Trevor Diehl
Central Michigan University, USA
Abstract
Recent scholarship suggests that overreliance on social networks for news and public
affairs is associated with the belief that one no longer needs to actively seek information.
Instead, individuals perceive that the “news will find me” (NFM) and detach from the
regular habit of traditional news consumption. This study examines effects of the
NFM perception on political knowledge, political interest, and electoral participation.
Drawing on a nationally representative panel survey from the United States (N = 997),
this study finds that the NFM perception is negatively associated with both political
knowledge and political interest across two time periods. The NFM perception also
leads to negative, indirect effects on voting as the relationship is mediated through
lower reported levels of political knowledge and interest in politics. The findings add
to current conversations about the ability of personalized information networks to
adequately inform and engage the public.
Keywords
News finds me perception effects, political interest, political knowledge, political
participation, social media news, voting behavior
Corresponding author:
Homero Gil de Zúñiga, Department of Communication, University of Vienna, Althanstraße 14 (UZA II),
Vienna 1090, Austria.
Email: homero.gil.de.zuniga@univie.ac.at
817548NMS0010.1177/1461444818817548new media & societyGil de Zúñiga and Diehl
research-article2018
Article
2 new media & society 00(0)
As news audiences shift attention away from broadcast era media platforms and toward
the socially mediated, multi-platform media environment, individuals may no longer
feel that they need to regularly follow traditional news media. Instead, people increas-
ingly rely on their extended friend networks for news and information relevant to them
(Matsa and Shearer, 2018). This practice is associated with the perception that active
news use is not necessary because the “news finds me” (NFM) without regular atten-
tion to professional news outlets (Gil de Zúñiga et al., 2017). NFM is a perception that
one will remain well informed through peers and online social networks, and it is a
direct outcome of emergent media practices (Gil de Zúñiga et al., 2017: 105). This
perception has developed, in part, because news feeds and social media offer a greater
potential for inadvertent and passive exposure to news online (Haim et al., 2018; Lee
and Kim, 2017). In addition, the contemporary news environment is characterized by
“ambient awareness” through smart phones and mobile devices where individuals
have instant access to a variety of personally relevant news sources, content updates,
and notifications (Hermida, 2010; Levordashka and Utz, 2016). Recent scholarship
suggests that the ability to shift from traditional news media to various online media,
along with the elevated role of relying on friends to provide information needs, may
have important outcomes for political behavior (Aldrich et al., 2016; Bennett et al.,
2018; Lu and Lee, 2018).
Reliance on the ambient information environment might lead to a general attitude of
detachment from traditional surveillance uses of news media. Surveillance use of media
is characterized by regular attention to news for the purpose of gathering information
about one’s environment (David, 2009; Norris, 1996). The displacement of traditional
surveillance practices is important since regular, even passive attention to traditional
news media has long been associated with higher levels of political knowledge, political
interest, and political participation (Boulianne, 2011; Chaffee and Frank, 1996). As
media choice becomes more prevalent, scholars argue that surveillance uses of news
media are in decline (Hopmann et al., 2016). One area of research has argued that this
shift leads to pro-democratic outcomes, like a tendency to participate in politics (Bennett
et al., 2018) or stumble across counter-attitudinal discussion (Lu and Lee, 2018). Yet
another line of research suggests the opposite is true; people who rely on social media for
news are less knowledgeable about current events and public affairs (Lee and Xenos,
2019; Yamamoto et al., 2018) and they are more susceptible to tabloid news and misin-
formation (Chadwick et al., 2018).
One reason for these conflicting narratives is that changes in the media environment
correspond with changing orientations toward the media itself (for a splendid example
of uses of social media, for instance, see Boczkowski et al., 2018; Diehl et al., 2018).
The NFM perception may account for how a shift in one’s surveillance practices—
from traditional news media to social networks and online information—influences
political attitudes and behavior. Previous work introduced the NFM concept and exam-
ined its relationship with patterns of attention to news and political knowledge (Gil de
Zúñiga et al., 2017). In response to that literature, this study employs a nationally
representative panel survey from the United States to explore how the NFM perception
relates to the following political outcomes: (1) interest in politics, (2) knowledge of
current events and political institutions, and (3) voting in national and local elections.
Gil de Zúñiga and Diehl 3
The study proposes a theoretical model, where voting is influenced by an individual’s
NFM perception, and the influence is mediated by individual levels of political interest
and knowledge of public affairs.
Literature review
Habits of news consumption and the shift in surveillance practices
Most adults in the United States (68%) get some news from social media, with about half
using Facebook (48%), a fifth using YouTube (21%), and one tenth using Twitter as a
gateway to news (12%; Matsa and Shearer, 2018). People around the world are increas-
ingly turning to private messaging apps and services like WhatsApp to communicate
(Newman, 2018). These news sources are also mobile, thus reducing the effort required
to follow the news, and increasing the speed at which one can stay informed. Young
people now get their news online or through social media, and when they do, it is spo-
radic; young people are less likely to read newspapers and are more likely to say that they
only check the news from time to time (Molyneux, 2018). The algorithms that filter
content on these platforms also tend to favor friend influence over news values in story
selection (DeVito, 2017).
Changes in news consumption patterns are a regular concern for scholars and politi-
cal analysts, because news consumption is thought to foster an informed, politically
engaged electorate (Bachmann and Gil de Zúñiga, 2013; Chambers, 2003; Norris,
1996). The connections between patterns of news consumption and pro-democratic
behaviors depend, in part, on one’s motivations for media use. Motivations are an
internal form of implied reasoning that individuals apply to make decisions related to
attaining a goal or achieving a desired outcome (Atkinson, 1964; Bandura, 2001).
Motivations for news use have long been associated with the goal of learning about
one’s environment; a motivation referred to as the surveillance need (Wright, 1960).
Surveillance needs are connected to increased attention to the news because they fulfill
certain psychological goals. For example, prior work suggests that both the needs for
cognition (thinking about things) and the need for evaluation are pre-cursors to surveil-
lance motivations (David, 2009). Individuals who are motivated by surveillance needs
tend to consume news media in order to monitor their social, political, and community
surroundings (Shoemaker, 1996).
Surveillance activities illustrate a basic normative feature of democratic citizenship.
Few people spend their time with direct political concerns. Instead, the so-called “moni-
torial citizen” uses media to scan the environment and only acts when the “alarm bell”
is heard (Schudson, 1998). Several studies show that surveillance uses of traditional
news media not only predict news consumption but also stimulate interest in politics
and political knowledge (David, 2009; Eveland, 2001). Those who use the media to
fulfill information needs tend to learn more from the news, because they are more moti-
vated to reflect upon and act on that information (Eveland et al., 2003; Kim et al., 2018).
Much of the literature on surveillance practices and their effects assume that people
should pay at least passive attention to news media.
4 new media & society 00(0)
In the modern media environment, professional news outlets must compete for the
public’s attention with a myriad of information sources, from online news and entertain-
ment to social information and constant updates and notifications. This leads to a sort of
“ambient awareness” of one’s information environment (Hermida, 2010; Levordashka
and Utz, 2016). In response to this experience, people are inclined to shift their attention
from traditional news sources toward their peers and social networks. This shift is borne
of a qualitatively different type of media experience. Accordingly, surveillance practices
are displaced from traditional news media. Instead, individuals develop the perception
that they can remain informed without monitoring the news per se, as they use their net-
works to navigate an ever-complex flow of information.
These changes in attitudes about one’s media environment—namely, the shift in sur-
veillance practices from traditional news media sources to peer groups and networks—
have a direct impact on how and why people respond to their information environment.
If the monitorial citizen waited for the news media to alert them about political issues or
opportunities, the modern media consumer is one step detached from this type of surveil-
lance. They monitor their social feeds and online networks instead. Surveillance prac-
tices are associated with cognition about one’s environment and therefore precipitate, or
at least influence, political behavior. Therefore, any change in surveillance practices and
associated orientations toward the news media should have a direct impact on political
behavior as well.
Political knowledge, political interest, and voting
Informed voters are able to identify their political preferences and opportunities, and
then translate those preferences into support for a candidate or issue (Carmines and
Stimson, 1980). As Delli Carpini and Keeter (1996) point out, those who are more
knowledgeable about politics tend to participate in elections more than less knowledge-
able people. There are two simple reasons for this. First, knowledge of politics promotes
other pro-democratic behaviors, like political interest and efficacy. This is because citi-
zens are able to discover politically relevant issues and problems. Second, knowledge
enables individuals to identify opportunities and connect them to their political needs.
These decisions are guided by information gathered from the news media. In this sense,
political activity is an outcome of so-called “mobilizing information” (Lemert, 1981).
Where, when, and how to vote are all possible when one has some basic knowledge of
and interest in politics.
Several studies provide empirical evidence for the relationship between political
knowledge and political interest to voting behaviors. In particular, media uses for news
and information often lead to a mutually re-enforcing relationship between news con-
sumption, levels of knowledge, and interest, particularly in those already interested in
politics (Norris, 1996; Verba et al., 1995). Based on the centrality of knowledge and
interest in politics to voting in elections, the following hypotheses are proposed:
H1: Political knowledge (W¹) is positively associated with voting in time (W²).
H2: Political interest (W¹) is positively associated with voting in time (W²).
Gil de Zúñiga and Diehl 5
NFM perception and political knowledge
As media technologies change, the underlying social, political, and institutional
forces that shape news consumption also change (Bimber, 2003; Williams and Delli
Carpini, 2011). These changes might also influence an individual’s attitudes toward
their needs for surveillance. As Bimber et al. (2015) argue, there is no reason to think
that all media technologies stimulate political interest and electoral participation in
the same way. Depending on how people pay attention to media, surveillance needs
alone may not lead to electoral participation. Individuals instead take cues about what
is, and what is not important from their extended social networks (Kaiser et al., 2018).
That is, individuals may shift from a surveillance use of traditional news media, and
instead rely on peers and social information generated in their self-curated networks.
Exposure to more passive forms of news filtering through algorithms and friend con-
nections may lead to the perception that one is informed about issues of the day—
even when they are not.
In response to these changes in the patterns of attention to and use of new media tech-
nologies, people may develop a corresponding perception about their need to actively
seek news. This attitude has been called the “news finds me” (NFM) perception, and is
theorized as an intersection of three related concepts. The NFM perception—in contrast
to other related concepts like incidental exposure or entertainment preferences—is an
individual’s perception that (1) they are well informed about current events, despite not
purposely following the news, because (2) the important information, “finds them” any-
way through their general media use, peers, and social connections; and (3) they rely on
social media for news (Gil de Zúñiga et al., 2017).
The reliance on self-curated networks and news feeds on social media platforms
can lead individuals to perceive that they will get the news that they personally find
important, and they can be well informed about public affairs by doing so. However,
empirical evidence suggests that few people completely avoid mainstream news
(Trilling and Schoenbach, 2013). Therefore, we expect that people might be inadvert-
ently exposed to some news through their social networks. This is particularly the
case for selective exposure to news online. One may also distinguish between popular
topical news and in-depth reporting, where the former is more likely to be transmitted
through social networks. These experiences reinforce one’s assumption that the “news
will find them” through their social media habits. The more one gets exposed to infor-
mation through their networks, the less likely they are to actively seek news. This
tendency leads to the perception that the media environment will fulfill their surveil-
lance needs.
Findings from surveys on this concept show that the NFM perception is associated
with increased reliance on social media for news and a decrease in print and television
news consumption (Gil de Zúñiga et al., 2017). Findings from experiments show that
news exposure on social media depends on the political sophistication and homogeneity
of their networks (Kaiser et al., 2018). These developments are particularly troubling for
those concerned with the potential of social media to create an informed electorate. It
stands to reason that if one overly relies on their extended social networks for news (and
avoids traditional news), the extent to which they are able to identify their basic political
6 new media & society 00(0)
needs and opportunities would also decline over time. NFM perceptions compound this
problem because these individuals hold a false sense of their needs for surveillance uses
of news media.
Although political learning can take place for passive attention to political information
online (Elenbaas et al., 2014), reliance on social platforms and messaging apps for news
has also been empirically linked to lower levels of political knowledge (Lee and Xenos,
2019; Yamamoto et al., 2018). One explanation for the inconsistent findings in this area is
that orientations toward the media are not considered. Since online and social media news
exposure is usually more passive, when these patterns of media use are combined with the
NFM perception, political learning is less likely to take place (Gil de Zúñiga et al., 2017).
The implication for political knowledge is that NFM represents a type of self-delusion
where people think they are informed, even though they are not. Therefore, we expect the
NFM perception to be negatively related to political knowledge:
H3: “NFM” perception (W¹) is negatively associated with political knowledge in
time (W²).
NFM perception and political interest
Most measures of political involvement are dependent upon the extent to which indi-
viduals report that they pay attention to politics. In particular, those with interest in poli-
tics tend to have the motivations and knowledge to make the best use of information they
come across online (Bimber, 2003; Verba and Nie, 1987; Xenos and Moy, 2007). This
leads to divergent possibilities for the impact of the NFM perceptions on political inter-
est. On one hand, the perception that one can stay informed through their networks is
influenced by the political sophistication of others in the social network (Kaiser et al.,
2018). Thus, highly sophisticated groups can isolate themselves online (Sunstein, 2018).
These isolated networks might create their own clusters, thus stimulating a level of polit-
ical interest not possible without the Internet and social media (e.g. Prior, 2005). Thus,
NFM perceptions might be associated with increased interest in politics.
On the other hand, NFM perceptions are more likely to be inversely related to politi-
cal interest in the general public. The displacement of surveillance uses of media suggest
that if one relies on their networks and curated information filters, they are more likely
to consume news sporadically (Molyneux, 2018). Without regular exposure to news,
individuals will be less able to identify politically relevant information, and even less
likely to translate that information into political interest. Politics is only rarely a topic of
centrality on social networking sites, and few use politics as a filter when they build and
maintain their networks (Bisgin et al., 2012). Thus, it follows that NFM perception
implies that people will assume that politically relevant information will find them,
regardless of whether they are actually interested in politics at all. Based on this theoreti-
cal rationale, the following hypothesis is proposed:
H4: “NFM” perception (W¹) is negatively associated with political interest in time
(W²).
Gil de Zúñiga and Diehl 7
NFM perception and voting
Literature on voting tends to offer at least two explanations for why people decide to
vote: resources and the cultivation of a civic norm. The resources tradition argues
that political participation is an expensive enterprise requiring time, money, and
organizational skills (Verba et al., 1995). Therefore, large gaps exist between those
who participate and those who do not, based on their use of and access to those
resources. The ability for the digital media environment to spur civic engagement is
dependent upon informational uses of news media. The rationale cited in conceptual
work in this area suggests that networked communication technology provides addi-
tional resources for getting involved in politics. In this context, media choice seems
to be accelerating gaps in electoral participation between high and low social status
groups (Prior, 2007).
Other scholars have explored voting as a product of cultivating civic norms. This
tradition suggests that voting is a civic duty, cultivated early in one’s life. Individuals
vote because their parents did, or they pick up the notion of voting as a civic duty in high
school (Campbell, 2006). Civic norms may also be transmitted through the media.
Community norms are transmitted through the news media by setting standards of
behavior for social groups (Bandura, 2001). For example, it is hard to imagine a case
where the news media did not cover an election. The implicit transmission of voting
norms is also apparent in the ubiquity of campaign coverage and staged media events. In
this context, the “dutiful” or “monitorial” citizen takes cues about when it is time to par-
ticipate in politics from the news media (Schudson, 1998; Gil de Zúñiga, 2012; 2017).
Implications of the NFM perception as hypothesized here are that individuals will
be less informed and less interested in politics. News avoidance (as reflected in the
increased reliance on cues from friend connections and social news feeds), in combi-
nation with less consumption of traditional media, lends theoretical support to the
idea that NFM individuals are not being exposed to civic norms as they are transmit-
ted by the mass media. Displacement of surveillance uses of media denies people the
benefits from the so-called “virtuous circle” (Gastil and Xenos, 2010; Norris, 1996)
where news use stimulates increased political interest and political participation. In
addition, those who disproportionately rely on social networks for their information
are more likely to prefer alternative modes of political engagement (Aldrich et al.,
2016; Bennett et al., 2018). This is because the nature of participatory resources ena-
bled by mobile and social platforms favors forms of online mobilization and protest
(Tufekci and Wilson, 2012).
Based on the conceptual implications of the NFM perception, it is likely those who
report higher levels of the NFM perception will also be less likely to say that they voted
in the last election. They are less likely to be exposed to the normative prompts for par-
ticipation provided by traditional news media. The resources provided by online and
social platforms are situated away from institutionalized politics. Finally, the proposed
inverse relationship between the NFM perception and political interest and knowledge
implies that there is an indirect (mediated) influence on voting. Scholars have yet to
address the direct and indirect relationships between NFM perceptions and voting.
Therefore, the following are proposed as research questions:
8 new media & society 00(0)
RQ1: What, if any, is the relationship between “NFM” perception (W¹) and voting in
time (W²)?
RQ2: What, if any, are the indirect effects of “NFM” perception (W¹) on voting (W²),
through political knowledge and political interest (W¹)?
Methods
Sample
This study draws from an online, custom panel survey created at the University of Texas
at Austin. The data collection was contracted to the media polling company Nielsen. For
an accurate representation of the US population, Nielsen specified a quota based on gen-
der, age, education, and income. This sampling procedure, and its limits, has been dis-
cussed at length elsewhere (e.g. Callegaro et al., 2014). The first wave of the survey (W¹)
was conducted between 15 December 2013 and 5 January 2014. The final number of
valid cases in Wave 1 was 1813. The response rate was 34.6%, which is within the
acceptable parameters for online panel surveys (American Association for Public
Opinion Research [AAPOR], 2016; Bosnjak et al., 2016; RR3). Second wave (W2) data
were collected from 15 February to 5 March 2014, with 1024 cases (retention rate of
57%).1 The sample was slightly older than current population estimates for the United
States, more educated, and included fewer Hispanics than the US Census. Yet, the overall
sample is sufficiently similar to the US census, and also comparable to other surveys
utilizing similar sampling strategies.
Outcome variables in the model
NFM perception. The independent variable of interest, NFM perception, was measured
with the following items. Respondents were asked to respond to the following prompt:
The way people get their news may have changed because of social networking sites and
digital media. Please tell us how much you agree or disagree with the following state-
ments where 1 = strongly disagree and 10 = strongly agree; “I rely on my friends to tell
me what’s important when news happens,” “I can be well informed even when I don’t
actively follow the news,” “I don’t worry about keeping up with the news because I
know news will find me,” and “I rely on information from my friends based on what they
like or follow through social media” (four items averaged scale, W¹ Cronbach’s α = .73;
M = 3.58; SD = 1.76).
Political knowledge. Knowledge of public affairs and political facts was measured in
accordance with prior literature (Delli Carpini and Keeter, 1996). A scale was created
that included measures of respondents’ awareness of current policy issues, rules of the
US political system, and institutional actors. Questions were a mix of multiple choice
and fill-in-the-blanks. Correct responses to each item were coded as 1, and incorrect
responses were coded as 0, range = 8, eight items additive scales; W¹ Kuder–Richardson
20 (KR20) = .75; M = 4.58, SD = 2.17; W2 KR20 = .71; M = 4.48, SD = 2.07.
Gil de Zúñiga and Diehl 9
Political interest. Respondents were asked about their willingness to pay attention to poli-
tics (Lupia and Philpot, 2005; Verba and Nie, 1987) and rate their degree of interest in
“information about what’s going on in politics and public affairs,” as well as their level
of attention to “information about what’s going on in politics and public affairs,” two
items averaged scale, W¹ Spearman–Brown (SB) coefficient = .96; M = 6.67, SD = 2.70;
W2 SB coefficient = .97; M = 6.29; SD = 2.63.
Voting. The dependent variable of interest in this study asked respondents about their
voting habits (Pew, 2012). Two items asked, “How often do you vote in local or state-
wide elections” and “How often do you vote in federal or presidential elections” (W2 SB
coefficient = .93; M = 8.2, SD = 2.91).2
Demographics and control variables
Demographics. The survey recorded respondent’s gender (49.7% females), age
(M = 52.71, SD = 14.72), and race (77.9% White). Level of education was operationalized
as highest level of formal education completed (M = 3.61, Mode = some college); and
income, which was measured from eight categories related to annual household income
(M = 4.46, Mdn = $50,000–$99,999).
Control variables. Because voting and political antecedents are related to a range of indi-
vidual characteristics, we included a full battery of control variables from Wave 1, using
survey items based on previous research. Most items rely on 10-point scales, except where
noted. They are discussion network size (2-item total scale, logged: M = 0.33, Mdn = 0.24,
SD = 0.37, skewness = 1.32), frequency of political discussion (4 items: Cronbach’s α = .805;
M = 4.46; SD = 2.28), internal political efficacy (2 items: SB coefficient = .87; M = 5.34,
SD = 2.56), strength of partisanship (single-item folded, 6-point scale: M = 2.1, SD = 1.87),
social media news use (11-item averaged index: Cronbach’s α = .89; M = 2.0, SD = 1.36);
overall news use: TV (7 items, Cronbach’s α = .79; M = 5.1, SD = 2.0), newspapers (3 items,
Cronbach’s α = .68; M = 4.5, SD = 2.5), radio news (2 items, SB coefficient = 84; M = 4.4,
SD = 2.8), news aggregators (2 items, SB coefficient = 55; M = 2.5, SD = 1.8), and citizen
journalism websites (including blogs) (2 items, SB coefficient = 65; M = 1.9, SD = 1.7); and
finally, incidental news exposure (7 items, Cronbach’s α = .82; M = 4.2, SD = 1.8).
Statistical analyses
Pearson’s correlations were employed to examine relationships between demographic
characteristics and the variables of interest (Table 1). To answer the hypotheses and research
questions, a series of regression analyses were conducted. First, a lagged ordinary least
squares (OLS) regression was used to examine the effect of NFM perceptions in Time 1
(Wave 1, W1) on political knowledge in Time 1 (W1), interest in politics in Time 1 (W1),
and voting in Time 2 (Wave 2, W2). Second, panel survey data allowed for an autoregres-
sive test that accounts for the effects of political interest and knowledge in Time 1, on the
same relationships in Time 2. The best predictor of future behavior is past behavior, and
therefore the modeling strategy tests knowledge and interest measured at Time 1 as
10 new media & society 00(0)
predictors of the same variables at Time 2 (autoregressive design). These tests provide a
more rigorous result. Finally, a parallel mediation path model was employed to test the
mediating effects of political knowledge and political interest on the NFM perception and
voting. Mediation analysis used the PROCESS macro, Model 4 in SPSS (Hayes, 2013).3
Results
According to the correlation tests (Table 1), the NFM perception is associated with
younger adults (r = –.27, p < .001), women (r = .11, p < .001), and those who report
lower incomes (r = –.06, p < .05). As expected, NFM is also positively associated with
social media for news (r = .44, p < .001) (Table 2) and incidental exposure to news
(r = .33, p < .001), while there is no relationship with television or newspaper use. The
positive relationship between both political knowledge and voting (H1), and political
interest and voting (H2), were confirmed in the OLS regression models (Table 2).
Political knowledge (W1) was the strongest predictor of voting (W2) in the model (Model
1; β = .201, p < .001), thus confirming H1. Political interest (W1) was almost as powerful
of a predictor (β = .164, p < .001). As the resources model of voting predicts, other posi-
tive, statistically significant predictors of voting (W2) include age, education, income,
and race (R2 = .19), reading printed newspapers (β = .051, p < .05), and political anteced-
ents (R2 = .12) (total model R2 = .374). Consistent with decades of findings on voting
behavior, demographics, political knowledge, and political interest are powerful predic-
tors that an individual will report he or she has cast a ballot.
The second set of research questions and hypotheses addresses the effects of the NFM
perception on political attitudes and behavior. In the cross-sectional framework, NFM
perception was negatively related to political knowledge (W1) (H3) (Table 2, Model 4;
β = –.110, p < .001). The more one believes that the news will find them, the less knowl-
edgeable they tend to be about basic actors, rules, and events in public life (total R2 = .50).
As with voting, the strongest predictor that one will be informed about public affairs is
political interest (β = .469, p < .001). The NFM perception is also negatively associated
Table 1. Zero-order Pearson’s correlations between demographics, political orientations, and
variables of interest.
1 2 3 4 5 6 7 8
1. Age
2. Gender (f) −.11***
3. Education −.02 −.03
4. Income .01 −.11*** .29***
5. Race (White) .14*** −.11*** −.01 .10**
6. NFM perception −.27*** .11*** −.04 −.06* −.08*
7. Political knowledge .23*** −.35*** .24*** .27*** .15*** −.27***
8. Political interest .31*** −.21*** .10** .19*** .10** −.20*** .60***
9. Voting .34*** −.09** .15*** .21*** .19*** −.14*** .44*** .48***
N = 1002. NFM, news finds me.
*p < .05; **p < .01; ***p < .001.
Gil de Zúñiga and Diehl 11
with political interest (W1) (H4) (Table 2, Model 2; β = –.102, p < .001). Thus, those that
hold the perception about being informed through their networks tend to be less interested
in politics in general. Those that are interested in politics tend to be older, male, watch TV
news, and have higher levels of political efficacy and knowledge (total model R2 = .68).
The negative effects of NFM perception held in the autoregression analysis for both
knowledge (W2) (Model 5; β = –.036; p < .05) and political interest (W2) (Model 3;
β = –.044, p < .05), providing some statistical support for the idea that NFM leads to
Table 2. Cross-sectional, lagged, and autoregression models estimating the news find me
perception effects on political interest, political knowledge, and voting.
Vote (W2) Political
interest (W1)
Political
interest (W2)
Political
knowledge
(W1)
Political
knowledge
(W2)
Model 1 2 3 4 5
Demographics
Age .174*** .062*** .020 .031 .001
Gender (f) .076*** .038* −.002 −.179*** −.054***
Education .059** −.031 .013 .118*** .033*
Income .067** −.034* .005 .115*** .030*
Race (White = 1) .119*** .030 .007 .027 .027*
% R219 16.4 16 25.4 25.4
News use
TV news .006 .160*** .057*** −.125*** −.020
Print news .051* .016 .021 −.019 .001
Radio news .033 .030 .006 .030 −.025
Aggregators −.015 .001 −.013 .050* .021
Citizen journalism −.037 .057** −.009 −.093*** −.003
Social media −.005 −.021 .038 −.050 −.026
Incidental exposure −.031 .009 −.001 .024 .009
∆R22.6% 13.1% 12.6% 1.9% 1.5%
Voting antecedents
Political efficacy .076** .378*** .103*** .073** .034
Discussion network size −.007 .076*** .028 .099*** .001
Discussion frequency .080** .176*** .059*** −.048 −.008
Strength of partisanship .179*** .011 .031* .018 −.005
∆R212.3% 32.1% 28.7% 13.5% 12.2%
Predictors in the model
Political interest (W1) .164*** − .640*** .469*** .087***
Political knowledge (W1) .201*** .302*** .084*** — .747***
News find me .028 −.101*** −.044** −.110*** −.036*
∆R24.5% 7% 19.1% 10.5% 36.8%
Total adjusted R2.374 .680 .760 .504 .753
N = 997. Cell entries are final-entry ordinary least squares (OLS) standardized Beta (β) coefficients.
Lagged = Model 1; cross-sectional = Models 2 and 4; autoregression = Models 3 and 5.
*p < .05; **p < .01; ***p < .001.
12 new media & society 00(0)
lower levels of political knowledge and less interest in politics over time (Table 3). Note
that regression coefficients in the autoregressive models (Models 3 and 5) for many of
the explanatory variables decrease, since much of the variance is explained by the autore-
gressive variables (political interest [W1], Model 3; political knowledge [W1], Model 4).
The NFM perception (W1) is not directly related to voting (W2) (RQ1) (β = .028,
p = .34) in the OLS model (Table 2, Model 1). The final research question (RQ2) asked
whether there was an indirect effect of the “NFM” perception on voting, through political
knowledge and political interest. In the parallel mediation model (Figure 1), political
interest and knowledge are included as mediators between the NFM perception (W1) and
voting (W2). The figure shows both the direct path, and indirect paths based on OLS
regression models. Results are reported as unstandardized regression coefficients.
Concurrent with the OLS regression models in Table 2, the NFM perception (W1) is nega-
tively associated with both political knowledge (W2) and political interest (W1). As
expected, both interest in politics (W1), and knowledge of current affairs and political
institutions (W1) are positively associated with voting (W2). The NFM (W1) perception
had no direct relationship with voting (W2) (Figure 1).
The indirect paths show a negative, compounding effect of “NFM” perceptions on
voting (Table 3). In other words, the direct, negative impact of this perception on politi-
cal knowledge and interest leads one to be less likely to vote in the future. The indirect
path estimate (Table 3) shows the negative, statistically significant effect of the NFM
perception (W1) on voting (W2) through political knowledge (W2), point estimate = –.008,
confidence interval (CI) = [–.024, –.0001]. The indirect effect through political interest
(W1) is also negative, and statistically significant (point estimate = –.009, CI = [–.04,
–.003]). The presence of statistically significant mediation effects, even after controlling
for demographics and political knowledge and interest in Time 1, provides support for an
indirect, negative relationship of the NFM perception on electoral behavior. Even with-
out a direct effect between voting and NFM perceptions, path models can still detect the
presence of indirect, mediated effects (see Hayes, 2013).4
Discussion
This study explored the potential effects of the NFM perception on attitudes toward demo-
cratic norms and values. We tested the NFM perception with respect to traditionally pro-
democratic benchmarks: political interest, political knowledge, and participating in elections.
Table 3. Indirect lagged and autoregressive effects of news find me perception on voting
through political interest and knowledge.
Indirect effects path Point
estimate
Confidence
interval
News find me (W1) → political interest (W2) → vote (W2) −.009 [–.039, –.003]*
News find me (W1) → political knowledge (W2) → vote (W2) −.008 [–.024, –.0001]*
Path estimates are unstandardized coefficients. Indirect effects derived from the mediation model in Figure
1. Bias-corrected confidence intervals based on bootstrapping to 10,000 samples. Asterisks (*) represent a
statistically significant indirect effect at the 95% interval, as the interval does not contain zero.
Gil de Zúñiga and Diehl 13
Results affirm previous findings on the importance of knowledge and interest in determin-
ing voting habits (H1; H2). In contrast, NFM perceptions are inversely related to political
interest and political knowledge (H3; H4). These findings were confirmed in autoregressive
tests. That is, by isolating the effects of political interest and knowledge in Time 1, we are
able to make a stronger argument for the continued, negative impact of NFM perceptions on
pro-democratic attitudes in time. Political knowledge and interest mediate the relationship
between NFM and voting, suggesting that the NFM effects are indirect, but strongly associ-
ated with a turn away from participation in institutionalized politics.
No direct relationship between NFM perceptions and voting was found. However,
and more importantly, we find evidence for robust negative relationships between NFM
perceptions and both political knowledge and interest. These negative relationships, in
turn, seem to remove, or at least decrease, pro-social behaviors that often precede partici-
pation in democratic life. In other words, the negative associations between NFM and
political knowledge and interest also decrease the likelihood one will cast a ballot. This
lack of knowledge and interest in politics then serves as a barrier to the most basic form
of political participation. Displacement of surveillance uses suggests that, within time,
individuals eventually lack the informational resources and interests needed to make
Figure 1. Indirect effects of the news finds me perception on political interest, knowledge, and
voting.
N = 997, path entries are unstandardized coefficients. The effects of demographic variables (age, gender,
education, race, and income), sociopolitical antecedents (political efficacy, strength of partisanship, discus-
sion network size, and political discussion frequency), news consumption variables, and autoregressive
terms (political knowledge and interest in Time 1) were included as control variables. Solid arrows indicate
statistically significant paths at p = .05 or below.
14 new media & society 00(0)
informed choices, or to actually make any choice whatsoever. Similarly, the linear rela-
tionships might be explained by the idea that NFM impacts political “drive.” That is,
NFM perceptions lead to a decline in the political energy and fuel to get mobilized.
There are at least two explanations for these negative effects. First, reliance on the
social networks for political information makes individuals increasingly reliant on the
political profile of their particular social network. Unless politics are central to the iden-
tity of the network, it is unlikely that algorithmic and self-curated information filters will
provide an opportunity for rich and diverse exposure to political information. Conversely,
people may be more likely to believe information or sources that are otherwise false. In
addition, politics are rarely a topic of central concern within social networks (Bisgin
et al., 2012). This is partly because the demands of maintaining social ties might over-
ride political concerns. It also may be because other activities, like entertainment or
social bonding, are more important to users.
NFM perceptions imply a shift in the long-held assumptions about how surveillance
uses of mass media operate. Based on the correlation results (Table 1), younger adults,
women, non-Whites, and those who report lower incomes are more likely to hold NFM
perceptions. In both the resources and civic norms paradigms, historically marginalized
groups tend to participate in politics at lower levels. NFM perceptions may help explain
the participation gap. If NFM perceptions lead to a false sense of awareness of one’s
civic sophistication, increased reliance on personal filters for information is likely to
disproportionately favor those who are already involved.
This study has some limitations. First, self-reported survey measures can be prone to
measurement error. Second, the two items measuring voting are limited in important
ways. First, no major election took place between waves. Thus, we were unable to take
full advantage of the panel design (i.e. autoregressive test for voting) and make no claims
of causal relationships. Second, some scholars correctly note that the social desirability
of voting items inflate reports. However, the social desirability bias in voting items is
debatable (see Krosnick, 1999). Based on the two-item average used in this study, 55%
of respondents reported voting often in national and local elections. This is roughly con-
sistent with national voter turnout rates for presidential elections (United States Election
Project, 2014). This study was also conducted before widespread knowledge of disinfor-
mation campaigns. NFM may have a role in the trust in or effects of fake news or propa-
ganda campaigns. Future work should explore the connections between following
disinformation campaigns and the NFM perception.
We also lack a deeper understanding of what drives the NFM perception. We know
that these individuals tend to be younger and rely more on social media for news.
However, it is unclear what the media content preferences are for these groups. What
kinds of news content (e.g. popular topical news, in-depth reporting, recommended sto-
ries) leads to the patterns of self-delusion implied by the NFM perception? This cannot
be explained by a simple preference for entertainment. It is more likely that the large
bubble of social media content that mingles news, entertainment, social interaction, and
so on generates an overall perception of being up to date. People are up-to-date with what
happens in the social network, with important issues to them, and ultimately, they believe
it is the same with public affairs news.
Gil de Zúñiga and Diehl 15
The effect of the NFM perception may be particularly relevant for young people. This
group displays higher levels of the NFM perception than their older counterparts. It is pos-
sible that this generation will become increasingly isolated from institutionalized politics.
In fact, voter turnout around the time of data collection (2014 Midterm elections) was the
lowest recorded for adults aged 18–29 at 20% (CIRCLE, 2014)—though initial results
from 2018 midterms suggest this may not be the whole story. More effort needs to be taken,
in classrooms and within the news media, to better socialize youth toward civic norms, like
voting and consuming hard news. This is an enormous challenge, particularly because an
overreliance on social media may act as a barrier to informing the public at large. Voting is
one of the most important diagnostics for the health of a democracy. Media has had an
historical role in informing the public about political issues, and this role fuels voting
behavior and participation in civic life (for an example on multiple societies, see Gil de
Zúñiga et al., 2018). Perceptions that the news will find us without effort may alter this
relationship, for the worse. Thus, future studies should explore how this new reality is
shaping citizenship. This study was a modest but important step in that direction.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this
article.
Notes
1. Twenty-two cases were deleted for patterns (Little’s MCAR test: Chi-square = 1555.246;
df = 1511; p = .209). Missing values were imputed using the Expectation–Maximization (EM)
algorithm to 25 iterations.
2. The items used in this study are similar to the Pew (2012) Research Center Election Questions,
and results here mimic national turnout numbers (55% of respondents in our survey reported
voting “all the time,” compared to the national rate of 58% in the 2012 presidential election;
United States Election Project 2014).
3. Panel design was limited to a 3-month period in the United States. Although respondents were
asked to report how often they vote in state and federal elections in both waves, no election
took place during that time. Therefore, we felt that any measurement difference between the
two waves would be due to measurement error. Statistical tests confirm these assumptions
(inter-item correlation, r = .90, p = .000). Therefore, the authors were limited in their ability to
include autoregressive terms for the dependent variable.
4. The authors ran several alternative path models, where the causal order is reversed (political
interest and political knowledge predicting news finds me [NFM] perceptions). Alternative
models yielded no statistically significant indirect effects voting through NFM.
ORCID iD
Homero Gil de Zúñiga https://orcid.org/0000-0002-4187-3604
References
Aldrich JH, Gibson RK, Cantijoch M, et al. (2016) Getting out the vote in the social media era:
are digital tools changing the extent, nature and impact of party contacting in elections? Party
Politics 22(2): 165–178.
16 new media & society 00(0)
American Association for Public Opinion Research (2016) Standard definitions: final dispositions
of case codes and outcome rates for surveys. Available at: https://www.aapor.org/AAPOR_
Main/media/publications/Standard-Definitions20169theditionfinal.pdf
Atkinson JW (1964) An Introduction to Motivation. Oxford: Van Nostrand.
Bachmann I and Gil de Zúñiga H (2013) News platform preference as a predictor of political and
civic participation. Convergence 19(4): 496–512.
Bandura A (2001) Social cognitive theory: an agentic perspective. Annual Review of Psychology
52: 1–26.
Bennett WL, Segerberg A and Yang Y (2018) The strength of peripheral networks: negotiating atten-
tion and meaning in complex media ecologies. Journal of Communication 68(4): 659–684.
Bimber B (2003) Information and American Democracy: Technology in the Evolution of Political
Power. Cambridge: Cambridge University Press.
Bimber B, Cantijoch Cunill M, Copeland L, et al. (2015) Digital media and political participation:
the moderating role of political interest across acts and over time. Social Science Computer
Review 33(1): 21–42.
Bisgin H, Agarwal N and Xu X (2012) A study of homophily on social media. World Wide Web
15(2): 213–232.
Boczkowski PJ, Matassi M and Mitchelstein E (2018) How young users deal with multiple
platforms: the role of meaning-making in social media repertoires. Journal of Computer-
Mediated Communication 23(5): 245–259.
Bosnjak M, Das M and Lynn P (2016) Methods for probability-based online and mixed-mode panels:
selected recent trends and future perspectives. Social Science Computer Review 34(1): 3–7.
Boulianne S (2011) Stimulating or reinforcing political interest: using panel data to examine
reciprocal effects between news media and political interest. Political Communication 28(2):
147–162.
Callegaro M, Baker RP, Bethlehem J, et al. (eds) (2014) Online Panel Research: A Data Quality
Perspective. West Sussex, UK: John Wiley.
Campbell DE (2006) Why We Vote: How Schools and Communities Shape Our Civic Life.
Princeton, NJ: Princeton University Press.
Carmines EG and Stimson JA (1980) The two faces of issue voting. American Political Science
Review 74(1): 78–91.
Chadwick A, Vaccari C and O’Loughlin B (2018) Do tabloids poison the well of social media?
Explaining democratically dysfunctional news sharing. New Media & Society 20: 4255–4274.
Chaffee S and Frank S (1996) How Americans get political information: print versus broadcast
news. The Annals of the American Academy of Political and Social Science 546: 48–58.
Chambers S (2003) Deliberative democratic theory. Annual Review of Political Science 6: 307–326.
CIRCLE (2014) 2014 youth turnout and youth registration rates lowest ever recorded; changes
essential in 2016. Available at: http://www.civicyouth.org/2014-youth-turnout-and-youth-
registration-rates-lowest-ever-recorded-changes-essential-in-2016/
David C (2009) Learning political information from the news: a closer look at the role of motiva-
tion. Journal of Communication 59(2): 243–261.
Delli Carpini MX and Keeter S (1996) What Americans Know about Politics and Why It Matters.
New Haven, CT: Yale University Press.
DeVito MA (2017) From editors to algorithms: a values-based approach to understanding story
selection in the Facebook news feed. Digital Journalism 5(6): 753–773.
Diehl T, Barnidge M and Gil de Zúñiga H (2018) Multi-platform news use and political par-
ticipation across age groups: toward a valid metric of platform diversity and its effects.
Journalism & Mass Communication Quarterly. Epub ahead of print 20 September 2018.
DOI: 10.1177/1077699018783960.
Gil de Zúñiga and Diehl 17
Elenbaas M, de Vreese C, Schuck A, et al. (2014) Reconciling passive and motivated learning:
the saturation-conditional impact of media coverage and motivation on political information.
Communication Research 41(4): 481–504.
Eveland WP (2001) The cognitive mediation model of learning from the news evidence from
nonelection, off-year election, and presidential election contexts. Communication Research
28(5): 571–601.
Eveland WP, Shah DV and Kwak N (2003) Assessing causality in the cognitive mediation model:
a panel study of motivations, information processing, and learning during campaign 2000.
Communication Research 30(4): 359–386.
Gastil J and Xenos M (2010) Of attitudes and engagement: clarifying the reciprocal relationship
between civic attitudes and political participation. Journal of Communication 60(2): 318–343.
Gil de Zúñiga HG (2012) Modeling the process of political participation in the EU. In: Thiel
M (ed), European identity and culture: Narratives of transnational belonging. New York:
Routledge, pp. 75-95.
Gil de Zúñiga H (2017) Attributes of interpersonal political discussion as antecedents of cog-
nitive elaboration. Revista Española de Investigaciones Sociológicas [Spanish Journal of
Sociological Research] 157: 65-84. DOI: 10.5477/cis/reis.157.65
Gil de Zúñiga H, Diehl T and Ardèvol-Abreu A (2018) Assessing civic participation around the
world: how evaluations of journalists’ performance leads to news use and civic participation
across 22 countries. American Behavioral Scientist 62(8): 1116–1137.
Gil de Zúñiga H, Weeks BE and Ardévol-Abreu A (2017) Effects of the news-finds-me percep-
tion in communication: social media use implications for news seeking and learning about
politics. Journal of Computer-Mediated Communication 22(3): 105–123.
Haim M, Graefe A and Brosius HB (2018) Burst of the filter bubble? Digital Journalism 6(3):
330–343.
Hayes AF (2013) Introduction to Mediation, Moderation, and Conditional Process Analysis: A
Regression-based Approach. New York: Guilford Press.
Hermida A (2010) Twittering the news: the emergence of ambient journalism. Journalism Practice
4(3): 297–308.
Hopmann DN, Wonneberger A, Shehata A, et al. (2016) Selective media exposure and increasing
knowledge gaps in Swiss referendum campaigns. International Journal of Public Opinion
Research 28(1): 73–95.
Kaiser J, Keller TR and Kleinen-von Königslöw K (2018) Incidental news exposure on
Facebook as a social experience: the influence of recommender and media cues on
news selection. Communication Research. Epub ahead of print 23 October 2018. DOI:
10.1177/0093650218803529.
Kim JW, Chadha M and Gil de Zúñiga H (2018) News media use and cognitive elaboration: the
mediating role of media efficacy. Revista Latina de Comunicación Social 73: 168–183.
Krosnick JA (1999) Survey research. Annual Review of Psychology 50(1): 537–567.
Lee JK and Kim E (2017) Incidental exposure to news: predictors in the social media setting and
effects on information gain online. Computers in Human Behavior 74: 1008–1015.
Lee S and Xenos M (2019) Social distraction? Social media use and political knowledge in two
U.S. Presidential elections. Computers in Human Behavior 90: 18–25.
Lemert JB (1981) Does Mass Communication Change Public Opinion after All? A New Approach
to Effects Analysis. Chicago: Nelson-Hall.
Levordashka A and Utz S (2016) Ambient awareness: from random noise to digital closeness in
online social networks. Computers in Human Behavior 60: 147–154.
18 new media & society 00(0)
Lu Y and Lee JK (2018) Stumbling upon the other side: incidental learning of counter-attitudinal
political information on Facebook. New Media & Society. Epub ahead of print 16 August
2018. DOI: 10.1177/1461444818793421.
Lupia A and Philpot TS (2005) Views from inside the net: how websites affect young adults’
political interest. The Journal of Politics 67(4): 1122–1142.
Matsa EK and Shearer E (2018) News use across social media platforms 2018. Pew Research
Center. Available at: http://www.journalism.org/2018/09/10/news-use-across-social-media-
platforms-2018/
Molyneux L (2018) Mobile news consumption: a habit of snacking. Digital Journalism 6(5):
634–650.
Newman N (2018) News in social media and messaging apps. The Digital News Project. Available
at: http://www.digitalnewsreport.org/publications/
Norris P (1996) Does television erode social capital? A reply to Putnam. PS: Political Science &
Politics 29(3): 474–480.
Pew Research Center. (2012). Pew research center election questions. Available at: http:// www.
people-press.org/files/2014/10/Pew-Research-2012-Election-Questions.pdf
Prior M (2005) News vs. entertainment: how increasing media choice widens gaps in political
knowledge and turnout. American Journal of Political Science 49(3): 577–592.
Prior M (2007) Post-broadcast Democracy: How Media Choice Increases Inequality in Political
Involvement and Polarizes Elections. New York: Cambridge University Press.
Schudson M (1998) The Good Citizen: A History of American Civic Life. New York: Free Press.
Shoemaker PJ (1996) Hardwired for news: using biological and cultural evolution to explain the
surveillance function. Journal of Communication 46(3): 32–47.
Sunstein CR (2018) #Republic: Divided Democracy in the Age of Social Media. Princeton, NJ:
Princeton University Press.
Trilling D and Schoenbach K (2013) Skipping current affairs: the non-users of online and offline
news. European Journal of Communication 28(1): 35–51.
Tufekci Z and Wilson C (2012) Social media and the decision to participate in political protest:
observations from Tahrir Square. Journal of Communication 62(2): 363–379.
United States Election Project (2014) 2014 November general election turnout rates. Available at:
http://www.electproject.org/2014g
Verba S and Nie NH (1987) Participation in America: Political Democracy and Social Equality.
Chicago, IL: The University of Chicago Press.
Verba S, Schlozman KL and Brady HE (1995) Voice and Equality: Civic Voluntarism in American
Politics. Vol. 4. Cambridge, MA: Harvard University Press.
Williams BA and Delli Carpini MX (2011) After Broadcast News: Media Regimes, Democracy,
and the New Information Environment. New York: Cambridge University Press.
Wright CR (1960) Functional analysis and mass communication. Public Opinion Quarterly 24:
605–620.
Xenos M and Moy P (2007) Direct and differential effects of the Internet on political and civic
engagement. Journal of Communication 57(4): 704–718.
Yamamoto M, Kushin MJ and Dalisay F (2018) How informed are messaging app users about pol-
itics? A linkage of messaging app use and political knowledge and participation. Telematics
and Informatics 35(8): 2376–2386.
Author biographies
Homero Gil de Zúñiga (PhD, University of Wisconsin–Madison) holds the Medienwandel
Professorship at University of Vienna, where he leads the Media Innovation Lab (MiLab). He also
Gil de Zúñiga and Diehl 19
serves as a research fellow at Universidad Diego Portales in Chile. His research addresses the
influence of new technologies and digital media on democratic processes.
Trevor Diehl (PhD, University of Vienna) is an assistant professor at Central Michigan University
at the School of Broadcast and Cinematic Arts. His research interests include social media and
politics, multi-platform news, and the effects of emerging media technologies on society.