ArticlePDF Available

Abstract and Figures

Second screening politics is an emerging communication practice for engaging with public affairs content. Scholars are increasingly interested in exploring pro-democratic effects of dual screening during news events and election cycles. This paper examines the potential for second screening practices to develop social capital on social media platforms through online and offline political discussion: a key component of maintaining social resources. More specifically, this manuscript focuses on the development of community-related social capital. Relying on two waves of panel data from 19 countries, the results suggest that dual screening contributes to the proliferation of building social capital on social media over time. This relationship is also partially mediated through online political discussion. Moreover, the between-country variation in the relationship between second screening and social media social capital is related to country-level freedom of expression indicators.
Content may be subject to copyright.
Human Communication Research ISSN 0360-3989
ORIGINAL ARTICLE
Effects of Second Screening: Building Social
Media Social Capital through Dual Screen Use
Brigitte Huber
1
, Homero Gil de Zúñiga
1,2
, Trevor Diehl
3
, & James Liu
4
1 Department of Communication, University of Vienna, A-1090 Vienna, Austria
2 Facultad de Comunicacíon y Letras, Universidad Diego Portales, CP 8320000, Santiago, Chile
3 School of Broadcast and Cinematic Arts, Central Michigan University, Mt. Pleasant, MI 48859, USA
4 School of Psychology, Massey University, North Shore Auckland 0745, New Zealand
Second screening politics is an emerging communication practice for engaging with
public aairs content. Scholars are increasingly interested in exploring pro-democratic
eects of dual screening during news events and election cycles. This paper examines
the potential for second screening practices to develop social capital on social media
platforms through online and oine political discussion: a key component of maintain-
ing social resources. More specically, this manuscript focuses on the development of
community-related social capital. Relying on two waves of panel data from 19 coun-
tries, the results suggest that dual screening contributes to the proliferation of building
social capital on social media over time. This relationship is also partially mediated
through online political discussion. Moreover, the between-country variation in the
relationship between second screening and social media social capital is related to
country-level freedom of expression indicators.
Keywords: Social Capital, Social Media, Second Screening, Dual Screen Use, Multi-Screen
Use, Complementary Simultaneous Media Use, Online Political Discussion, Freedom of
Expression.
doi:10.1093/hcr/hqz004
Second screening is an emerging phenomenon. The practice of using a comple-
mentary, or dual, screen when watching the news or political events on television
(or another device) has received considerable research attention. Scholars are
beginning to investigate the eects of second screening politics on a variety of dem-
ocratic behaviors (Gil de Zúñiga, Garcia-Perdomo, & McGregor, 2015;Lin &
Chiang, 2017;Liu, Kim, & Kim, 2017;McGregor & Mourão, 2017). It may be, for
example, that people are employing second-screening practices to bypass elite insti-
tutions (Gillespie & OLoughlin, 2015), while other scholars suggest that second-
screening is an important indicator of collective action (Chadwick, Dennis, &
Smith, 2016). Given the wide range of social and communicative aordances dual
Corresponding author: Brigitte Huber; e-mail: brigitte.huber@univie.ac.at
1Human Communication Research 00 (2019) 132 © The Author(s) 2019. Published by Oxford University Press on behalf of
International Communication Association. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
Downloaded from https://academic.oup.com/hcr/advance-article-abstract/doi/10.1093/hcr/hqz004/5382473 by ICA Member Access user on 01 April 2019
screening implies, understanding exactly how network communication technolo-
gies are being leveraged for political activity is still an open research question. In
particular, the connective and networked nature of second screening has implica-
tions for the development of social resources online. This paper examines the
potential inuence of second screening on developing online forms of social capi-
tal, a central element in spurring civic participation (Kim, 2007). What, if any, is
the relationship between second screening and the spread of digital forms of social
capital in society?
This study examines whether second screening has the potential to foster social
media social capital: that is, the social capital created in social media environments.
Current literature in this area oers several theoretical reasons to expect that sec-
ond screening will be positively related to social media social capital. First, infor-
mational media use is a positive predictor of social capital oine (Geber, Scherer,
& Hefner, 2016). One of the main motivations for citizens to engage in second
screening behaviors is to seek additional information (Chadwick, OLoughlin, &
Vaccari, 2017) and, therefore, this communicative practice may indirectly foster
social capital. A second important motivation for second screening, especially for
political uses, is discussing with others the information people come across through
the mainstream media (Liu et al., 2017). Second screeners interact with people in
their own network, as well as with people they dont necessarily know, in platforms
like Twitter or Facebook (McGregor, Mourão, Neto, Straubhaar, & Angeluci,
2017). These types of interactions may increase social capital online.
Despite the growth of research in this area, one ongoing problem is the empiri-
cal verication of theoretical expectations. Since second screening involves both
information and discussion, conceptual models of media eects that incorporate
cognitive responses to media content may be useful for understanding how second
screening leads to other behaviors. The communication mediation model (McLeod,
Scheufele, & Moy, 1999), the cognitive mediation model (Eveland, 2001,2004),
and the O-S-R-O-R model (Cho et al., 2009) are associated theoretical accounts of
media eects that suggest that the path from media use to its outcomes is not a
direct one. For example, prior research has shown that political discussion med-
iates the relationship between news use and social capital (Ardèvol-Abreu, Diehl, &
Gil de Zúñiga, 2018). Accordingly, we applied path modelling to test whether this
mediation process also exists for the relationship between second screening and
social media social capital.
Relying on two waves of panel data, collected in 19 countries, this study tested
the relationship between second screening politics and public aairs and social
media social capital in dierent social contexts. Additionally, we tested whether
this relationship is mediated through online political discussion. Finally, we investi-
gated the role macro-level variables (Internet connectivity and the democratic
expression index) might play in shaping dierences in this relationship across
countries.
2Human Communication Research 00 (2019) 132
Effects of Second Screening B. Huber et al.
Downloaded from https://academic.oup.com/hcr/advance-article-abstract/doi/10.1093/hcr/hqz004/5382473 by ICA Member Access user on 01 April 2019
Second screening effects
Second screening is a communicative practice in which individuals watching tele-
vision use an additional electronic device or screento access the Internet or social
networking sites to obtain more information about the program or event they are
watching or to discuss it in real time(Gil de Zúñiga et al., 2015, p. 795). The addi-
tional electronic device, or screen, can be a smartphone, tablet, or laptop. In the lit-
erature, this communication practice is also referred to as dual screen use,
complementary, simultaneous media use,or multiscreen use.Sometimes, sec-
ond screening may entail two dierent devices without a TV set, such as a laptop
and a smartphone. Whatever the specic case may be, there are dierent motiva-
tions for using a second screen: seeking further information, sharing information,
discussing with others, or inuencing others (Chadwick et al., 2017;Gil de Zúñiga
et al., 2015). In general, younger people tend to use a second screen more than old-
er people do (Gil de Zúñiga et al., 2015;Gil de Zúñiga & Liu, 2017). People who
dont have a great deal of trust in traditional media also rely on a second screen
more (Gil de Zúñiga et al., 2015). Additionally, second screening practices vary
from country to country. For example, in Turkey, China, and Brazil, second
screening is very common, but it is relatively less so in Germany, the United
Kingdom, and the United States (Gil de Zúñiga & Liu, 2017). This trend of dier-
ential use was also established by McGregor et al. (2017), such that higher levels of
second screening were found in Brazil over the United States.
When it comes to the eects of second screening, it is relevant to have a look at
multitasking literature, as second screening can be seen as one form of media mul-
titasking (Gottfried, Hardy, Holbert, Winneg, & Jamieson, 2017). Multitasking is
not a new phenomenon, but it has become an increasingly prominent topic in light
of media saturation and convergent technologies (Wang & Tchernev, 2012).
Multiple resource theory (Wickens, 2002) and thread cognition (Salvucci &
Taatgen, 2008) help explain what happens when people split their attention across
tasks simultaneously. The basic assumption of these theoretical approaches is that
peoples cognitive capacity to process information is restricted. However, as long as
two tasks do not require the same pool of mental resources, multiple tasks can be
handled eectively (Wickens, 2002).
Research on media multitasking suggests that people seem to have an intuitive
grasp of their own cognitive limits and adjust their behaviors accordingly when
interacting with this changing media landscape(Wang, Irwin, Cooper, &
Srivastava, 2015, p. 122). A large body of literature has found negative eects of
media multitasking on various outcomes, such as work performance and learning
(Armstrong, Boiarsky, & Mares, 1991;Bowman, Levine, Waite, & Gendron, 2010;
Pool, Koolstra, & van der Voort, 2003). Accordingly, second screening could be
expected to have negative eects, as it may distract viewers from following the
debate, news, election coverage, and so forth on TV. However, Houston,
Hawthorne, Spialek, Greenwood, and McKinney (2013) found that people who
3Human Communication Research 00 (2019) 132
B. Huber et al. Effects of Second Screening
Downloaded from https://academic.oup.com/hcr/advance-article-abstract/doi/10.1093/hcr/hqz004/5382473 by ICA Member Access user on 01 April 2019
used Twitter while watching a political debate paid more attention to the debate.
While Gottfried et al. (2017) showed that the positive eect of watching a debate
on political knowledge is dulled when simultaneously engaging in social media
multitasking, other studies found a positive eect on political knowledge (Houston,
McKinney, Hawthorne, & Spialek, 2013;Liu et al., 2017). Scholars investigating the
eects of second screening also found positive eects on other outcome variables,
such as political participation (Gil de Zúñiga et al., 2015;Gil de Zúñiga & Liu,
2017;Lin & Chiang, 2017;Liu et al., 2017;McGregor & Mourão, 2017), reconsider-
ing ones political views (Barnidge, Gil de Zúñiga, & Diehl, 2017), and political dis-
cussion (Liu et al., 2017).
Second screening and political discussion
The bourgeoning eld of scholarship in this area has, ironically, passed over the
central role of political discussion in second screening eects. The positive relation-
ship between second screening and political discussion might be explained by the
information-seeking motivation of dual screening; prior research has suggested
that media use for informational purposes (Jung, Kim, & Gil de Zúñiga, 2011;
Kim, Wyatt, & Katz, 1999) and online information seeking (Shah, Cho, Eveland, &
Kwak, 2005;Xenos & Moy, 2007) positively predict political discussion. When
using media for informational purposes, people get new ideas, facts, and opinions
that can spur political discussions (Beaudoin & Thorson, 2006). They encounter
dissimilar political views that they would not be exposed to through interpersonal
political discussants alone (Mutz & Martin, 2001). This might be especially true for
second screeners, as social media has the potential to diversify and expand news
and information networks.
1
For example, research has shown that friendships on
social media cut across ideological aliations (Bakshy, Messing, & Adamic, 2015);
most social media users are embedded in ideologically diverse networks (Barberá,
2014). News use on social media has also been found to diversify political commu-
nication (Barnidge, 2015). Accordingly, dual screeners who seek news on social
media may get exposed to a wider range of information and views.
Moreover, exposure to real-time conversations online while watching politics
on TV may provide social cues and other information that may mobilize indivi-
duals to engage in conversation, at least online. According to our denition, second
screening only entails discussions related to the political content people are watch-
ing on TV. Second screening may then further stimulate various forms of political
discussion; that is, conversations over and above talk related to the specic content
watched. For example, people watching a political talk show dealing with the
results of elections on TV may then further discuss the candidate of the winning
party on Twitter. This discussion might, subsequently, lead to conversations about
other topics in other discursive contexts, both online and o. Thus, second screen-
ing has the potential to stimulate further discussions online.
However, it is not clear that these practices actually inuence oine discussions
as well. One scenario that is very plausible is that online conversations taking place
4Human Communication Research 00 (2019) 132
Effects of Second Screening B. Huber et al.
Downloaded from https://academic.oup.com/hcr/advance-article-abstract/doi/10.1093/hcr/hqz004/5382473 by ICA Member Access user on 01 April 2019
during second screening activities, as well as content looked up, might stimulate
conversations oine. For example, after discussing the candidate of the winning
party with other users on Twitter while watching the election coverage on TV, one
might afterwards start a conversation about this discussion oine while having
dinner: One user on Twitter argued that this candidate only succeeded because of
his connections to inuential people. I dont agree; I think that he is simply smart
and people liked his agenda. What do you think?Or one might look up informa-
tion about the candidate online while watching the election coverage on TV, and
talk about this information with others face-to-face afterwards. In this case, the sec-
ond screening activity might have the potential to spur oine political discussion
as well. However, another possible scenario might be that the opportunity to dis-
cuss with others about the election coverage online while watching TV deters peo-
ple from actively seeking to have these discussions oine. For example, two people
might watch the election coverage together (e.g., in a shared apartment or in a bar)
and both engage, separately, in conversations about the election with their own
friends via WhatsApp. In this case, second screening might not foster oine politi-
cal discussion.
Thus, we expected second screening to positively predict political discussions
online. Second, we aimed to explore how second screening is related to political
discussion oine.
H1: Second screening is positively associated with online political discussion.
RQ1: How does second screening relate to oine political discussion?
Social capital and political discussion
Since the early 90s, scholars from dierent disciplines have used the concept of
social capital to explain why communities succeed or fail in resolving collective
problems (Coleman, 1990;Fukuyama, 1995;Putnam, 1993). Social capital can be
described as the resources embedded in peoples social networks (Lin, 2008). These
resources of information, norms, and social relations enable people to coordinate
collective action and achieve common goals (Shah & Gil de Zúñiga, 2008).
According to Wellman, Haase, Witte, and Hampton (2001), three forms of social
capital can be distinguished: (a) network capital, (b) participatory capital, and (c)
community commitment. Network capital comprises the relationships with friends,
neighbors, relatives, and workmates, and provides companionship, emotional sup-
port, goods and services, information, and a sense of belonging. Participatory capi-
tal is the involvement of people in politics and voluntary organizations, which
gives the chance to bond and articulate demands and desires. Community commit-
ment reects a strong attitude toward community:thatis,amotivated,responsible
sense of belonging. As we are interested in social capital as a basis for a strong and
active civil society, and drawing on previous research along these lines (Ardèvol-Abreu
5Human Communication Research 00 (2019) 132
B. Huber et al. Effects of Second Screening
Downloaded from https://academic.oup.com/hcr/advance-article-abstract/doi/10.1093/hcr/hqz004/5382473 by ICA Member Access user on 01 April 2019
et al., 2018), our operationalization of social capital is closer to the latter concept.
Hence, this study focused on community-related social capital.
Oine, online, and social media social capital
Social capital has been discussed by a range of scholars across disciplines. While
the denition of social capital diers to some extent from researcher to researcher,
there is conformity that social capital is derived from relations with other people in
a social structure(Sum, Mathews, Pourghasem, & Hughes, 2008, p. 204). With
the rise of the Internet and social media platforms, new possibilities to build and
maintain relationships with other people emerged. In 2001, Resnick spoke of the
opportunities of sociotechnical capital,highlighting that information and com-
munication technologies have the potential to remove barriers to interaction,
expand interaction networks, or maintain histories of previous interactions
(Resnick, 2001). Hence, social capital can also be distinguished according to the
setting or channel it is generated and accessed through (Abbas & Mesch, 2018;
Ellison, Steineld, & Lampe, 2007;Gil de Zúñiga, Barnidge, & Scherman, 2017;
Williams, 2006): (a) face-to-facegenerated social capital (i.e., oine), (b) online-
generated social capital, and (c) social mediagenerated social capital. While oine
social capital refers to the social ties and resources that are accessed in face-to-face
encounters, online social capital refers to connections and resources that are
accessed using the Internet (Abbas & Mesch, 2018). Scholars also further dierenti-
ate between bonding and bridging social capital online (e.g., Internet Social Capital
Scales, see Williams, 2006; for a discussion on the validity of dierent social capital
measures, see Appel et al., 2014). Additionally, the term social media social capi-
talwas introduced to describe a third form of social capital (Gil de Zúñiga et al.,
2017). The term is used to refer to connections and resources accessed in social
media settings. In this paper, we use the term social media for social network sites
and micro-blogging sites (e.g., Facebook, Twitter, Google+, Pinterest, Instagram,
Tumblr, Reddit, LinkedIn). Accordingly, we use the term social media social capital
for all forms of connections and resources accessed using these and similar
platforms.
Gil de Zúñiga et al. (2017, p. 45) argue that social media change the structure
and nature of social connection, and therefore they may alter the distribution and
nature of social capital embedded in those social relationships.More specically,
social media social capital is characterized by the latent distribution of social values
and resources, diversied processes of recognizing and developing value in social
relationships, and new possibilities to convert those latent social values and
resources into individual or collective benets (Gil de Zúñiga et al., 2017). Hence,
while in terms of access, online social capital and social media social capital might
be very similar, as the social media environment is a part of the broader concept of
online media environments, the dierences between the two concepts becomes
apparent when looking at their attributes. Social media have unique attributes,
such as the level of interactions, content, context, connections, network structure,
6Human Communication Research 00 (2019) 132
Effects of Second Screening B. Huber et al.
Downloaded from https://academic.oup.com/hcr/advance-article-abstract/doi/10.1093/hcr/hqz004/5382473 by ICA Member Access user on 01 April 2019
platform, and so forth. These attributes are distinct from more general online
experiences and behaviors. For example, an anonymous online forum oers dier-
ent possibilities for building social capital than a platform like Facebook or Twitter,
where a wide range of users register with their real names and prole pictures. The
prole does some of the work of showing common interests and helping users nd
like-minded others (Ellison & Vitak, 2015). What makes social network sites
unique is not that they allow individuals to meet strangers, but rather that they
enable users to articulate and make visible their social networks. This can result in
connections between individuals that would not otherwise be made(boyd &
Ellison, 2007, p. 211).
Scholars are increasingly interested in investigating how these dierent types of
social capital are related to each other (Dunbar, 2016;Dunbar, Arnaboldi, Conti, &
Passarella, 2015;Williams, 2006). There are two competing theoretical perspectives:
the rich get richermodel (Kraut et al., 2002) and social compensation theory
(McKenna & Bargh, 1998). The rich get richer model (Kraut et al., 2002) assumes
that people who are highly sociable (extraverted people) and who already have
existing social support oine will receive more social benets from using the
Internet. They do so by adding people to their social networks and/or by fostering
existing connections. In contrast, social compensation theory (McKenna & Bargh,
1998) expects that people with low oine social resources will benet from meet-
ing new people online. Hence, less extraverted people might prot from the new
possibilities to connect with others, without being forced to engage in face-to-face
situations.
Empirical results in this area paint a mixed picture. On the one hand, research
suggests that social media do not necessarily boost the reach of peoples social net-
works, as the size and range of online egocentric networks was found to be similar
to that of oine face-to-face networks (Dunbar, 2016; Dunbar et al., 2015). On the
other hand, research shows that peoples stock of oine social capital is not
enough to predict online social capital (Abbas & Mesch, 2018). In the same vein,
the social capital created on social media was found to be a stronger predictor of
oine social capital than the other way around (Gil de Zúñiga et al., 2017).
Because the current study is concerned with the nature of networked communica-
tion behaviors, and based on recent work on social capital in digital spaces, we
focused on social media social capital.
Since social capital is, in part, a by-product of the discursive interactions
between citizens (La Due Lake & Huckfeldt, 1998), we expected that social media
social capital would be related to online political discussion as well. In other words,
discussion seems to be a fundamental resource that allows for the further develop-
ment of social and community ties. Prior research has shown a positive relation-
ship between political discussion and social capital oine (Ardèvol-Abreu et al.,
2018). There is little reason to assume that this function does not take place online
and through social media as well. However, the empirical connections between o-
line discussions and social media social capital are less developed, and therefore a
7Human Communication Research 00 (2019) 132
B. Huber et al. Effects of Second Screening
Downloaded from https://academic.oup.com/hcr/advance-article-abstract/doi/10.1093/hcr/hqz004/5382473 by ICA Member Access user on 01 April 2019
research question was proposed. Hence, we formulated the following hypothesis
and research question:
H2: Online political discussion is positively related to social media social capital.
RQ2: How is oine political discussion related to social media social capital?
Second screening and social media social capital
The relationship between media use and social capital has attracted increased
scholarly attention since Putnam (1995a,1995b) blamed television for destroying
civic engagement in the United States. He argued that television is privatizing the
use of leisure time and disrupting opportunities for building social capital (time
displacement hypothesis). Additionally, and based on cultivation theory, Putnam
assumed heavy television viewing to lead to more skepticism about the benevolence
of other people and to pessimism about human nature (the mean world syndrome;
Gerbner, Gross, Morgan, & Signorielli, 1980). These assumptions were called into
question by empirical research indicating that time spent with television did not
aect civic engagement through perceptions of time pressure eects (Moy,
Scheufele, & Holbert, 1999), and that eects are determined by exposure to specic
content rather than by overall television use (Shah, McLeod, & Yoon, 2001). For
example, watching a social drama had a positive eect on respondentscivic partic-
ipation, but watching a comedy program had the opposite eect (Shah et al., 2001).
Similarly, the Internet initially was assumed to have negative eects on social
capital, due to displacement of social activities by solitary activities (Graber, Bimber,
Bennett, Davis, & Norris, 2004). Diverse work in the area, based on meta-analyses,
rebutted these negative eects (Boulianne, 2009,2015,2017;Skoric, Zhu, Goh, &
Pang, 2016), suggesting a positive relationship between Internet use and civic engage-
ment and between social media use and civic engagement. Increased access to a
large, diverse set of political information may help stimulate civic engagement. In
other words, the Internet may reduce the costs of participation (time, eort) by
increasing the availability of information(Boulianne, 2009,p.205).
Recent research has suggested that Internet use generally and informational
media use specically both increase social capital (Geber et al., 2016). Gil de
Zúñiga, Jung, and Valenzuela (2012) found that social media use for news is a posi-
tive predictor of social capital. As one common practice of second screening is
seeking news on social media while watching television, this result is especially rel-
evant when theorizing about the eect of second screening on social capital. Social
media diversify and expand news and information networks (Bakshy et al., 2015;
Barnidge, 2015;Kim, 2011). Moreover, news on social media are accompanied by
social recommendations (Bode, 2012,2016;Thorson & Wells, 2015), which aect
how users engage with news. For example, individuals tend to engage with news
and information posted by people that are similar to them (Bonchi, Castillo, &
Ienco, 2013;Ma, Lee, & Goh, 2013). Similarly, feeling close to the person and
8Human Communication Research 00 (2019) 132
Effects of Second Screening B. Huber et al.
Downloaded from https://academic.oup.com/hcr/advance-article-abstract/doi/10.1093/hcr/hqz004/5382473 by ICA Member Access user on 01 April 2019
trusting the person that posted the information enhances ones willingness to
engage with the content (Ganley & Lampe, 2009;Media Insight Project, 2017;
Wang & Mark, 2013).
Second screeners consuming news on social media may prot from the fact
that social media enable communication with other users and help bring together
those with shared interests, common values, and goals (Ellison, Steineld, &
Lampe, 2011). McGregor et al. (2017) showed that dual screeners interact with a
wide range of people: people who are already part of their own network, in plat-
forms that are not accessible to the general public (e.g., text messaging and email),
as well as with people they dont necessarily know on platforms like Twitter or
Facebook. For these reasons, it was hypothesized that second screening would be
positively related to social media social capital across 19 countries.
H3: Second screening will be positively related to social media social capital.
Second screening enables dierent groups to actively participate in discussions
(Bentivegna & Marchetti, 2015), has the potential to add new ideas and content to
discussions (Anstead & OLoughlin, 2011), and provides space to express and form
opinions (Giglietto & Selva, 2014). As elaborated above, prior research, based on
the O-S-R-O-R framework, showed that the relationship between news use and
social capital is mediated through political discussion (Ardèvol-Abreu et al., 2018).
Hence, we proposed political discussion as a mediator between second screening
and social media social capital.
H4: Second screening is indirectly associated with social media social capital
through online political discussion.
Macro-level variables: Freedom of expression and Internet connectivity
While there is a rich body of research investigating the relationship between media use
and social capital at the individual level, systematic eorts to specify the role of context-
level predictors have been relatively lacking (Kang & Kwak, 2003). Why is it fruitful to
compare countries and consider contextual conditions? According to Hallin and
Mancini (2004), comparative research is valuable because it makes us aware of variations
and similarities, which can help to form and rene concepts. This is important, given
that most research is highly ethnocentric, in the sense that it refers only to the experi-
ence of a single country, yet is written in general terms(Hallin & Mancini, 2004,p.2).
What makes comparative research especially fruitful is its recognition of the relevance of
contextual conditions: that is, the attempt to link macro-level system conditions and
micro-level variables (Esser, 2013).Ourstudycontributestothiskindofresearchby
investigating the relationship between a specic type of media use (second screening)
and a specic type of social capital (community-related social capital) in 19 countries,
and by including macro-variables to explain dierences between countries.
It stands to reason that media eects on building social capital vary by country.
For example, democratic media structures were found to moderate the impact of
9Human Communication Research 00 (2019) 132
B. Huber et al. Effects of Second Screening
Downloaded from https://academic.oup.com/hcr/advance-article-abstract/doi/10.1093/hcr/hqz004/5382473 by ICA Member Access user on 01 April 2019
Internet use on social capital. Geber et al. (2016) showed that the eect is stronger
in societies with democratic media systems than in countries with weak democratic
structures, and concluded that societies may benet from an open, free, and plural-
istic media system, which provides a forum for discussing with others and promot-
ing civil engagement. A fruitful approach to measuring how free and open media
are, as well as how individuals can express themselves, is the V-dem freedom of
expression index. The index is provided by the Varieties of Democracy Institute, an
independent research institute based at the Department of Political Science at the
University of Gothenburg in Sweden. The freedom of expression index is based on
9 indicators: namely, government censorship eorts in media; harassment of jour-
nalists; media self-censorship; freedom of discussion for men; freedom of discussion
for women; freedom of academic and cultural expression; media bias; print/broad-
cast media critical; and print/broadcast media perspectives (Coppedge et al., 2016).
Additionally, cross-national research has suggested that broadband penetration
of a country may serve as a moderator when looking at digital media eects in dif-
ferent countries. Recent ndings indicated that the positive relationship between
political media use and political participation is stronger in countries with higher
broadband penetration (Vettehen, Troost, Boerboom, Steijaert, & Scheepers, 2017).
This might also be true for the relationship between second screening and social
capital. As discussed above, Internet use on the individual level is positively related
to social capital. Therefore, the arguments discussed above should also apply on
the contextual level. Hence, we included Internet connectivity as the second
macro-variable in our analysis. The percentages of Internet users and users with
broadband access, xed broadband subscriptions, and mobile phones may aect
the relationship between second screening and social media social capital.
Democratic expression and Internet connectivity were, therefore, included in the
comparative analysis. These indices, while limited, may account for how access to and
free use of media alter the eects of political behaviors. The extent to which second
screening matters may depend on systemic inuences on media use. As we were inter-
ested in testing whether second screening may have dierent eects contingent upon
these dierent macro-level variables, we posed the following research questions:
RQ3: How do (a) Internet connectivity and (b) freedom of expression relate to
social media social capital?
RQ4: Are there cross-level interactions between second screening and macro-level
variables(a) Internet connectivity and (b) freedom of expressionthat aect
social media social capital?
Methods
Sample and data
This study used two waves of panel data, gathered in 19 countries worldwide (see
Table 2). A large group of participating scholars translated the items for each
10 Human Communication Research 00 (2019) 132
Effects of Second Screening B. Huber et al.
Downloaded from https://academic.oup.com/hcr/advance-article-abstract/doi/10.1093/hcr/hqz004/5382473 by ICA Member Access user on 01 April 2019
country, and responses were back-translated into English by research teams in
Europe and New Zealand (Behling & Law, 2000). The rst wave of the study was
elded online between 14 and 24 September 2015, and the second wave of the
study between 22 March and 17 April 2016. The online survey was performed by
Nielsen, drawing on more than 10 million potential participants. Stratied quota
sampling techniques were applied to create samples whose demographics closely
matched those reported by ocial census agencies in each country (Callegaro et al.,
2014). The sample sizes were 20,361 in Wave 1 (W1)
2
and 8,708 in Wave 2 (W2).
The overall cooperation rate was relatively high, averaging 77% across the panel
(AAPOR, 2011). Because Nielsen partners with companies that employ a combina-
tion of panel- and probability-based sampling methods, the limitations of web-only
survey designs were minimized in this case (Bosnjak, Das, & Lynn, 2016).
However, some parameters of the panel invitations are unknown and, therefore,
traditional response rates should not be calculated (AAPOR, 2011). For a demo-
graphic breakdown by country, see Supplementary Appendix Tables A1A3.
Individual-level measures
Social media social capital
The dependent variable in the analysis is social media social capital. This variable
focused on the social connectedness aspect of social capital (Williams, 2006;Zhang
& Chia, 2006): that is, it captured community-related social capital. Based on prior
research (Gil de Zúñiga et al., 2017) this variable relied on four questionnaire items
that asked respondents how often they use social media (1 =never; 7 =all the
time) to nd people to solve problems in their community, to connect community
members to each other, to encourage conversation about solving community pro-
blems, and to foster community values. The nal variable took the average of the
four scores (W1: α=.96, M=3.26, SD =1.70; W2: α=.96, M=2.83, SD =1.66;
Table 1).
Oine social capital
Also based on prior research (Gil de Zúñiga et al., 2017), people were asked how
much they agreed or disagreed (1 =never; 7 =all the time) with the following
statements: people in my community feel like family to me,”“I think people in
my community share values,”“in my community, we talk to each other about com-
munity problems,”“I think people in my community feel connected to each other,
and in my community, people help each other when there is a problem.The ve
items were averaged to create the nal variable (W1: α=.93, M=3.10, SD =1.21).
Online political discussion
The two political discussion variables (online and oine) were operationalized
based on previous literature (Eveland & Hively, 2009;Gil de Zúñiga, Valenzuela, &
Weeks, 2016;Kwak, Williams, Wang, & Lee, 2005), and focus on the frequency of
discussions about politics and public aairs with both weak and strong ties. Online
political discussion was assessed by asking respondents how often they talk about
11Human Communication Research 00 (2019) 132
B. Huber et al. Effects of Second Screening
Downloaded from https://academic.oup.com/hcr/advance-article-abstract/doi/10.1093/hcr/hqz004/5382473 by ICA Member Access user on 01 April 2019
politics or public aairs (1 =never; 7 =all the time) online with (a) a spouse or
partner; (b) family, relatives, or friends; (c) acquaintances; and (d) strangers. The
four items were averaged to create the nal variable (W1: α=.88, M=2.45, SD =
1.46; W2: α=.89, M=2.26, SD =1.43).
Oine political discussion
Oine political discussion was measured by asking respondents how often they
talk about politics or public aairs (1 =never; 7 =all the time) face-to-face or over
the phone with (a) a spouse or partner; (b) family, relatives, or friends; (c) acquain-
tances; and (d) strangers. These four items formed a reliable scale (W1: α=.82, M
=3.38, SD =1.38; W2: α=.82, M=3.34, SD =1.39).
Second screening
Based on prior work (Gil de Zúñiga et al., 2015), second screening was assessed by
measuring how often people use an additional electronic device to access the
Internet while watching the news on TV. The exact wording was as follows:
When watching television, how often do you use an additional electronic device
to access the Internet or a social media to get more information or to talk about
the program or event youre viewingan activity sometimes called second screen-
ing?Then, we asked how often they second screen while watching (a) political
speeches or debates, (b) the news, and (c) election coverage. The three items, which
were measured on a 7-point scale (1 =never; 7 =all the time), were averaged to
create the nal variable (W1: α=.92, M=3.10, SD =1.73).
Table 1 Conrmatory Factor Analysis
Items Estimate
Oine social capital
I think people in my community feel connected to each other. .92*
In my community, people help each other when there is a problem. .85*
I think people in my community share values. .84*
In my community, we talk to each other about community problems. .88*
People in my community feel like family to me. .79*
Social media social capital
a
I use social media to encourage conversation about solving community
problems.
.96*
I use social media to connect community members to each other. .91*
I use social media to foster community values. .93*
I use social media to nd people to solve problems in my community. .89*
Note: N =20,328. CI =condence interval.
a
The question wording was: Please tell us how often you use social media for the following
activities: to Cell entries are standardized factor loading estimates for the twolatent
variable model.
*p<.001.
12 Human Communication Research 00 (2019) 132
Effects of Second Screening B. Huber et al.
Downloaded from https://academic.oup.com/hcr/advance-article-abstract/doi/10.1093/hcr/hqz004/5382473 by ICA Member Access user on 01 April 2019
Social media news use
Also based on prior research (Gil de Zúñiga et al., 2012;Valenzuela, Arriagada, &
Scherman, 2012), four questionnaire items asked how often respondents use social
media to (a) get news, (b) stay informed about current events and public aairs,
(c) get news about their local communities, and (d) get news about current events
from mainstream media (e.g., professional news services). These four items, which
were measured on a 7-point scale (1 =never, 7 =all the time), formed a reliable
scale (α=.87, M=4.30, SD =1.50).
Traditional news use
Three items, measured on a 7-point scale (1 =never, 7 =all the time), asked
respondents how often they get news from television news (cable or local network
news), printed newspapers, and radio (α=.60, M=4.52, SD =1.32).
Political interest
Two survey items asked respondents (a) how closely they pay attention to and (b)
how interested they are in information about what is going on in politics and pub-
lic aairs. The two scores were averaged to create the nal variable (Spearman-
Brown Coecient =.94, M=4.52, SD =1.45).
Internal political ecacy
Respondents were asked how much they agree or disagree (1 =never, 7 =all the
time) with the following statements about public life: people like me can inuence
governmentand I consider myself well qualied to participate in politics.The
two items were averaged to create the nal score (Spearman-Brown Coecient =.71,
M=3.93, SD =1.16).
Strength of partisanship
The strength of partisanship was measured using three items that asked respon-
dents to place themselves on the partisan spectrum in terms of party identication
(0 =strongly liberal, 10 =strongly conservative) on (a) political issues, (b) eco-
nomic issues, and (c) social issues. These three items were averaged and then
folded in the following way: scores farther away from the midpoint (5) took higher
values and those closer to the midpoint took smaller values (α=.91, M=2.82, SD
=1.53).
Media trust
We asked respondents how much they trust (a) news from mainstream news
media (e.g., newspapers, TV), (b) news from alternative media (e.g., blogs, citizen
journalism), and (c) news from social media. The three items were averaged to cre-
ate the nal variable (α=.77, M=3.47, SD =1.11).
Frequency of social media use
Building on prior research (Correa, Bachmann, Hinsley, & Gil de Zúñiga, 2013),
the frequency of social media use was measured by asking respondents how much
they use social media and instant messaging (1 =never, 7 =all the time). The two
13Human Communication Research 00 (2019) 132
B. Huber et al. Effects of Second Screening
Downloaded from https://academic.oup.com/hcr/advance-article-abstract/doi/10.1093/hcr/hqz004/5382473 by ICA Member Access user on 01 April 2019
items were averaged to build the nal variable (Spearman-Brown coecient =.63;
M=4.82; SD =1.53).
Demographics
The analyses account for the following demographic variables: age (M=41, SD =
14.64), gender (51% female), education (measured on an 8-point scale where 1 =
none and 7 =post-graduate degree; M=4.34, SD =1.30), income (annual house-
hold income, range of scale: 1 =010 percentile; 5 =91100 percentile; M=2.94,
SD =1.10), and ethnicity or race (coded as majority =1; for detailed information
per country, see Supplementary Appendix Table A1).
Country-level measures
Freedom of expression
The freedom of expression index is based on 9 indicators (government censorship
eorts in media; harassment of journalists; media self-censorship; freedom of dis-
cussion for men; freedom of discussion for women; freedom of academic and cul-
tural expression; media bias; print/broadcast media critical; and print/broadcast
media perspectives). These values were taken directly from data reported on www.
v-dem.net (see Coppedge et al., 2016). The index (M=.82, SD =.22) ranges from
.25 (China) to .99 (United Kingdom).
Internet connectivity
The Internet connectivity index comprised the percentage of Internet users, per-
centage of users with broadband access, percentage of users with xed broadband
subscriptions, and percentage of users with mobile phone subscriptions. These four
Internet connectivity statistics were collected from www.webworldwide.io. The
index (M=74.35, SD =13.26) ranges from 45.15% (Philippines) to 88.45%
(Estonia; for more details on the macro-level variables per country, see
Supplementary Table A5 and Supplementary Figure A1).
Analysis
First, a conrmatory factor analysis was performed to test whether social media
social capital is empirically distinct from oine social capital. Following the
instructions of Holbert and Grill (2015), we compared a twolatent variable model
against a onelatent variable model. We used R (package lavaan). Second, one-
sample t-tests provided a description for how countries dier from the grand
means of social media social capital. Third, we employed hierarchical ordinary least
squares regressions to test our rst set of hypotheses and answer our rst set of
research questions. Since there were multiple measurements over time, the regres-
sion models included cross-sectional, lagged, and autoregressive models, where
appropriate. In the rst instance (cross-sectional), we used the W1 data set; in the
second instance (lagged), we tested the W1 predictors on the W2 outcome variable;
and in the third model (autoregressive), we tested the W1 predictors on the W2
outcome variable, controlling for the outcome variable in W1. While the cross-
14 Human Communication Research 00 (2019) 132
Effects of Second Screening B. Huber et al.
Downloaded from https://academic.oup.com/hcr/advance-article-abstract/doi/10.1093/hcr/hqz004/5382473 by ICA Member Access user on 01 April 2019
sectional data do not allow for causal inferences, this study includes respondents
prior scores on the outcome variable in an autoregressive association, in order to
better deal with issues of endogeneity and causal inference (for more details see,
e.g., Greenberg, 2008;Kleinnijenhuis, 2016). To test our hypothesis on indirect
eects, we ran mediation models using PROCESS (Hayes, 2013). Finally, multi-
level models and cross-level interactions were analyzed using R, and Figure 1was
created with the visregpackage (Breheny & Burchett, 2017).
Results
Before testing the hypotheses, we checked whether social media social capital is a
distinct concept from oine social capital. We compared a twolatent variable
model against a onelatent variable model (see Holbert & Grill, 2015; see Table 1).
The singlelatent variable model produced the following t statistics: a conrma-
tory t index (CFI) of .508, a root mean square error of approximation (RMSEA)
of .398 (90% condence interval .396.400), and a standardized root mean squared
residual (SRMR) of .242. None of the three t estimates met the recommended cut-
ovalues (CFI, .95 or higher; RMSEA, .06 or lower; SRMR, .09 or lower). The
twolatent variable model resulted in better model t statistics (CFI =.984;
RMSEA =.07, 90% condence interval .069.074; SRMR =.03). When comparing
the two models, Δχ2(df =1) =87,0492,776 =84,273, statistically signicant at
the p<.001 level. The twolatent variable model was shown to be an improvement
over the onelatent variable model. Hence, our results suggest that social media
social capital is empirically distinct from oine social capital.
Next, one-sample t-tests were conducted to assess each countrysdierence
with the overall sample in terms of mean levels of social media social capital (M=
3.26, SD =1.70). The results are summarized in Table 2.
3
The highest test statistics
(indicating country means greater than the grand mean) were seen in Indonesia
(23.95), the Philippines (22.07), and Turkey (15.48). Meanwhile, the lowest test sta-
tistics were seen in Japan (23.50), Ukraine (22.10), and Estonia (19.97).
Finally, non-signicant test statistics (indicating a country mean close to the grand
mean) were observed in Argentina (1.35), China (1.44), South Korea (1.60),
Poland (1.17), and Spain (.14).
H1 predicted a positive relationship between second screening and online polit-
ical discussion. We ran autoregressive regression models testing the eect of our
control variables and second screening on online and oine political discussion.
Table 3shows that second screening at Time 1 is a positive predictor of online
political discussion at Time 2 =.060, p<.001). These results support H1, and
show that holding all other predictors constant, each SD increase in the frequency
of second screening results in a .06 increase of a SD in political discussions online.
RQ1 asked how second screening relates to oine political discussion. Results in
Table 3indicate that second screening at Time 1 is not a predictor of oine
15Human Communication Research 00 (2019) 132
B. Huber et al. Effects of Second Screening
Downloaded from https://academic.oup.com/hcr/advance-article-abstract/doi/10.1093/hcr/hqz004/5382473 by ICA Member Access user on 01 April 2019
political discussion at Time 2. That is, those who second screen more are not
engaging more in political discussions face-to-face or over the phone.
H2 predicted a positive relationship between online political discussion and
social media social capital. Consistent with our expectation, the results in Table 4
show that online political discussion at Time 1 is positively related to social media
social capital in all three models: in the cross-sectional model, higher scores on a
scale of online political discussion resulted in about a fth of a SD higher score
in social media social capital (ß =.223, p<.001). The results are similar for the
lagged model =.258, p<.001) and the autoregressive model (ß =.161, p<.001).
Those who engage more in political discussions online also build more social capital in
social media environments. RQ2 asked how oine political discussion relates to social
media social capital. Results in Table 4suggest that people engaging in more oine
discussions about politics are less likely to build social media social capital.
Table 2 Tests of Mean Dierences between Country Mean and the Grand Mean for Social
Media Social Capital
Social Media Social Capital
Country M(SD)t(df)
Argentina 3.33 (1.69) 1.35 (1143)
Brazil 4.03 (1.70) +14.88 (1083)*
China 3.36 (1.37) 1.44 (1002)
Estonia 2.48 (1.34) −−19.97 (1161)*
Germany 2.62 (1.67) −−12.44 (1047)*
Indonesia 4.32 (1.45) +23.95 (1074)*
Italy 3.50 (1.72) +4.49 (1036)*
Japan 2.21 (1.39) −−23.50 (973)*
South Korea 3.18 (1.53) 1.60 (942)
New Zealand 2.47 (1.46) −−18.44 (1156)*
Philippines 4.26 (1.47) +22.07 (1047)*
Poland 3.32 (1.58) 1.17 (1057)
Russia 3.48 (1.66) +4.41 (1143)*
Spain 3.27 (1.69) .14 (1018)
Taiwan 3.83 (1.32) +13.68 (1003)*
Turkey 4.07 (1.62) +15.48 (955)*
Ukraine 2.21 (1.55) −−22.10 (1062)*
United Kingdom 3.13 (1.58) −−2.82 (1217)
United States 2.38 (1.53) −−19.64 (1159)*
Notes: Cell entries are means (M), standard deviations (SD), test statistics (t) and degrees of
freedom (df) from one-sample t-tests assessing the dierence between each country mean
and the grand mean for social media social capital (M=3.26, SD =1.70). The +or signs
denote whether the dierence with the grand mean is a positive or a negative one.
*p<.001 (two-tailed tests).
16 Human Communication Research 00 (2019) 132
Effects of Second Screening B. Huber et al.
Downloaded from https://academic.oup.com/hcr/advance-article-abstract/doi/10.1093/hcr/hqz004/5382473 by ICA Member Access user on 01 April 2019
H3 predicted a positive relationship between second screening and social media
social capital. Results in Table 4conrmed this assumption: second screening at
Time 1 positively predicts social media social capital in the cross-sectional model
=.095, p<.001), the lagged model =.090, p <.001), and the autoregressive
model (ß =.048, p <.001).
4
These results suggest that individualshabits to second
screen while watching political speeches or debates, the news, or election coverage
foster creating social capital, even after controlling for demographics, social orien-
tations, news use, political discussion, and oine social capital. Among all the
demographics introduced in the autoregressive model, age =.053, p<.001)
negatively predicted social media social capital. That is, younger individuals tend to
create more social media social capital than older individuals do. Gender, educa-
tion, income, and race do not predict social media social capital. Among all social
orientations included in the autoregressive model, media trust (ß =.037, p<.001)
positively predicted social media social capital. Additionally, social media news use
entered in the model as control positively predicted social media social capital.
People who consume news via social media tend to build more social media social
capital (ß =.081, p<.001).
H4 predicted that the relationship between second screening and social media
social capital is partially mediated through online political discussion. To test H4,
we carried out mediation models (Table 5;Hayes, 2013) testing the indirect eects
of second screening at Time 1 on social media social capital through online politi-
cal discussion (at Time 1). As presented in Table 5, we found signicant indirect
eects on social media social capital in the cross-sectional, lagged, and
autoregressive models. Online political discussion positively mediates the relation-
ship between second screening and social media social capital. Those who second
screen more often tend to discuss politics and public aairs with others online
more often, which increases social capital in social media environments.
Comparative results
The last set of research questions asked how Internet connectivity (RQ3a) and free-
dom of expression (RQ3b) relate to social media social capital, and whether they
can explain country dierences in the eect of second screening on social media
social capital (RQ4ab). Figure A1 in the Supplementary Appendix plots the
Internet connectivity and freedom of expression index scores by country (also see
Table A5 in the Supplementary Appendix). The highest-scoring countries on
Internet connectivity included Estonia, the United Kingdom, South Korea, Russia,
Japan, and Germany. The lowest-scoring countries were the Philippines, Indonesia,
China, Brazil, and Turkey. Those countries scoring high on the freedom of expres-
sion index (indicating societies freer to express themselves) included the United
Kingdom, Estonia, Germany, Brazil, Spain, New Zealand, and Italy. The lowest
scores were seen in China, Russia, Turkey, Ukraine, and South Korea.
Turning to the random component models, and in agreement with Table 2,a
multi-level model with a random intercept (null model, no predictors) was a better
17Human Communication Research 00 (2019) 132
B. Huber et al. Effects of Second Screening
Downloaded from https://academic.oup.com/hcr/advance-article-abstract/doi/10.1093/hcr/hqz004/5382473 by ICA Member Access user on 01 April 2019
t (log likelihood =15,442.58) than a model with a xed intercept (pooled sam-
ple, log likelihood =16,275.39; p<.001). That is, about 20% of the variance in
social media social capital is due to country dierences (intraclass correlation ICC
=.202). Table 6shows the full multi-level model, including individual-level
Table 3 Autoregressive Regression Model Predicting Political Discussion (19 countries)
Online Political
Discussion
W2
Oine Political
Discussion
W2
Block 1: Demographics
Age .081* .026
Gender (female =1) .034* .016
Education .011 .001
Income .021 .060*
Race (majority =1) .014 .012
ΔR
2
.074* .044*
Block 2: Social orientation
Political interest
W1
.027 .112*
Internal political ecacy
W1
.028 .018
Strength of partisanship
W1
.032* .039*
Media trust
W1
.026 .027
ΔR
2
.115* .193*
Block 3: Media use
Social media news use
W1
.092* .030
Traditional news use
W1
.048* .108*
Frequency of social media
use
W1
.071* .027
ΔR
2
.110* .054*
Block 4: Political discussion
Online political discussion
W1
.053*
Oine political discussion
W1
.045*
ΔR
2
.046* .047*
Block 5: Autoregressive term
Online political discussion
W1
.463*
Oine political discussion
W1
.477*
ΔR
2
.121* .119*
Block 6: Second screening
Second screening
W1
.060* .008
ΔR
2
.003* .000
Total R
2
.467* .455*
Note:N=6,828. Cell entries are nal-entry ordinary least squares standardized coecients
(β). The autoregressive model predictor is W1 and the outcome variable is W2, controlling
for the autoregressive term (online political discussion/oine political discussion in W1).
W1 =Wave 1; W2 =Wave 2.
*p<.001.
18 Human Communication Research 00 (2019) 132
Effects of Second Screening B. Huber et al.
Downloaded from https://academic.oup.com/hcr/advance-article-abstract/doi/10.1093/hcr/hqz004/5382473 by ICA Member Access user on 01 April 2019
Table 4 Cross-Sectional, Lagged, and Autoregressive Regression Model Predicting Social
Media Social Capital
SM
Social Capital
W1
(cross-sectional)
SM
Social Capital
W2
(lagged)
SM
Social Capital
W2
(autoregressive)
Block 1: Demographics
Age .004 .057* .053*
Gender (female =1) .056* .050* .033
Education .004 .000 .008
Income .012 .002 .006
Race (majority =1) .010 .017 .007
ΔR
2
.062* .077* .077*
Block 2: Social orientation
Political interest
W1
.081* .042* .014
Political ecacy
W1
.035* .035* .017
Strength of
partisanship
W1
.011 .003 .011
Media trust
W1
.013 .043* .037*
ΔR
2
.100* .090* .090*
Block 3: Media use
SM news use
W1
.527* .345* .081*
Traditional news
use
W1
.016 .025 .027
Frequency of SM
use
W1
.045* .047* .063*
ΔR
2
.314* .219* .219*
Block 4: Political
discussion
Online discussion
W1
.223* .258* .161*
Oine discussion
W1
.019 .048* .043*
ΔR
2
.053* .055* .055*
Block 5: Oine social
capital
Oine social
capital
W1
.152* .113* .048*
ΔR
2
.019* .010* .010*
Block 6: Autoregressive
term
SM social capital
W1
.467*
ΔR
2
.093*
(Continued)
19Human Communication Research 00 (2019) 132
B. Huber et al. Effects of Second Screening
Downloaded from https://academic.oup.com/hcr/advance-article-abstract/doi/10.1093/hcr/hqz004/5382473 by ICA Member Access user on 01 April 2019
variables, macro-level variables, and the cross-level interactions. The average eect
for the model (intercept) is .261 (SE =.054), and that eect varies between the 19
countries, with a between-country variance of .01 (SD =.11). The within-country
variance is much higher (1.23, SD =1.11), suggesting that individual-level factors
are more explanatory.
Table 4 Continued
SM
Social Capital
W1
(cross-sectional)
SM
Social Capital
W2
(lagged)
SM
Social Capital
W2
(autoregressive)
Block 7: Second screening
Second screening
W1
.095* .090* .048*
ΔR
2
.007* .006* .002*
Total R
2
.554* .456* .545*
Note. Cell entries are nal-entry ordinary least squares standardized coecients (β). N
cross
=
16,138; N
lagged/auto
=6,915. In the cross-sectional model, the predictor was W1 and the out-
come variable was W1. In the lagged model, the predictor was W1 and the outcome variable
was W2. In the autoregressive model, the predictor was W1 and the outcome variable was
W2, controlling for the autoregressive term of social media social capital in W1. SM =social
media; W1 =Wave 1; W2 =Wave 2.
*p<.001.
Table 5 Indirect Eect Tests of Second Screening over Social Media Social Capital via
Online Political Discussion (cross-sectional, lagged, and autoregressive; 19 countries)
Indirect Eects
Point
Estimate 95% CI
Second screening (W1) - >Online political discussion (W1) ->
Social media social capital (W1)
.0375 .0336 to .0416*
Second screening (W1) ->Online political discussion (W1) ->
Social media social capital (W2)
.0419 .0355 to .0494*
Second screening (W1) ->Online political discussion (W1) ->
Social media social capital (W2)
.0216 .0170 to .0267*
Notes: Table reports unstandardized coecients. The indirect eect is based on bootstrap-
ping to 5,000 samples with bias-corrected CIs. The eect of demographic variables (age,
gender, education, income, and race), social orientations (political interest, political ecacy,
strength of partisanship, and media trust), media use (social media news use, traditional
news use, and frequency of general social media use), oine political discussion, and oine
social capital were included as control variables. N
cross
=16,138; N
lagged/auto
=6,915. CI =
condence interval; W1 =Wave 1; W2 =Wave 2.
*p<.001.
20 Human Communication Research 00 (2019) 132
Effects of Second Screening B. Huber et al.
Downloaded from https://academic.oup.com/hcr/advance-article-abstract/doi/10.1093/hcr/hqz004/5382473 by ICA Member Access user on 01 April 2019
Table 6indicates that, for the country-level predictors, only Internet connectiv-
ity matters. Countries with higher Internet connectivity rate create less social media
social capital. Freedom of expression does not predict social media social capital.
When it comes to the interaction eects, Internet connectivity does not explain the
country-level dierences in social media social capital, but freedom of expression
does. There is a positive, statistically signicant, cross-level interaction eect
between the freedom of expression index and second screening (b=.092,
SE =.041; p=.03). This eect is plotted in Figure 1. The eect of second screening
at Time 1 on social media social capital at Time 2 is stronger in countries with
higher levels of freedom of expression than in countries with lower levels of free-
dom of expression. One can observe a substantial dierence in the slope (eect
size) between the solid (low freedom of expression) and the dashed (high freedom
of expression) lines.
Discussion
The aim of this study was to test whether second screening positively inuences
political discussion and, in turn, social media social capital in dierent countries.
Our results show a positive relationship between dual screening and social media
social capital across dierent countries. The eect size found in our study is similar
to that found in previous research on the eect of social media news use on social
capital (Gil de Zúñiga et al., 2012, p. 326). Our ndings also revealed that online
Low Dem. Expression High Dem. Expression
−0.25
−0.20
−0.15
−0.10
−0.05
0.00
0.05
0.10
−2 −1 0 1 2 3 4
SM Social Capital W2
Second Screening W1
Figure 1 The plot shows the cross-level interaction between the second screening (W1)
and the Freedom of Democratic Expression Index (Table 5) on social media social capital
(W2). The moderator was probed at the 10
th
(low) and 90
th
(high) percentiles. W1 =Wave
1; W2 =Wave 2.
21Human Communication Research 00 (2019) 132
B. Huber et al. Effects of Second Screening
Downloaded from https://academic.oup.com/hcr/advance-article-abstract/doi/10.1093/hcr/hqz004/5382473 by ICA Member Access user on 01 April 2019
political discussion partially mediates the relationship between second screening
and social media social capital. Those who second screen more are more likely to
engage in political discussions online, which increases their social capital in social
media environments.
Table 6 Autoregressive, Multi-Level Models and Cross-Level Interactions Predicting Social
Media Social Capital
W2
SM Social Capital
W2
b(SE)
Fixed eects
Age .004 (.001)** .004 (.001)***
Gender (female =1) .090 (.028)* .091 (.028)**
Education .018 (.012) .018 (.012)
Income .004 (.013) .005 (.013)
Race (majority =1) .002 (.048) .001 (.048)
Political interest
W1
.019 (.012) .019 (.012)
Internal political ecacy
W1
.028 (.013)** .028 (.013)*
Strength of partisanship
W1
.013 (.010) .013 (.01)
Media trust
W1
.064 (.014)*** .064 (.014)***
SM news use
W1
.071 (.015)*** .070 (.015)***
Traditional news use
W1
.041 (.012)*** .042 (.012)***
Frequency of SM use .053 (.013)*** .053 (.013)***
Online political discussion
W1
.173 (.014)*** .173 (.014)***
Oine political discussion
W1
.045 (.014)*** .045 (.014)**
Oine social capital
W1
.065 (.013)*** .066 (.013)***
Second screening
W1
.050 (.009)*** .049 (.009)***
SM social capital
W1
.437 (.013)*** .437 (.013)***
Country-level predictors
Internet connectivity .011 (.003)*** .011 (.002)***
Democratic expression index .189 (.144) .194 (.143)
Cross-level interactions
Second screening
W1
×Democratic expression .092 (.041)*
Second screening
W1
×Internet
connectivity
.001 (.001)
Random eects (SD)
Variance within country 1.23 (1.11) 1.23 (1.11)
Variance between country .01 (.11) .01 (.11)
Log likelihood 10,537.34 10,534.81
Intercept b (SE) .261 (.054) .261 (.054)
Note: N =6,915. Predictors and dependent variables centered to the grand mean. In the
autoregressive model, the predictor was W1 and the outcome variable was W2, controlling
for the autoregressive term of social media social capital in W1. SM =social media; W1 =
Wave 1; W2 =Wave 2.
*p<.05; **p<.01; ***p<.001.
22 Human Communication Research 00 (2019) 132
Effects of Second Screening B. Huber et al.
Downloaded from https://academic.oup.com/hcr/advance-article-abstract/doi/10.1093/hcr/hqz004/5382473 by ICA Member Access user on 01 April 2019
There are several reasons why dual screening might be directly related to social
media social capital. First, those who already cultivate social capital oine are also
likely to be more involved in politics and, thus, use emerging communication prac-
tices to further increase their social and political resources. Second, those that sec-
ond screen politics tend to be motivated news and information consumers. Thus,
the increased opportunities to access and share information are likely to expand
their online networks. In a similar vein, the indirect eects on building social
media social capital through discussion help solidify current contacts, expose users
to larger discussion networks, and perhaps provide information necessary for facili-
tating social capital through those discussions. Further research should investigate
who might prot most from second screening activity: extraverted people who
already have social support oine (the rich get richer model; see Kraut et al., 2002)
or introverted people with low oine social resources, for whom the online setting
oers new possibilities to connect with others (social compensation theory; see
McKenna & Bargh, 1998).
People who second screen seem to benet from the potential of second screen-
ing to transform individual domestic consumption (privatization) into socializing
processes (online social interaction; Dheer & Courtois, 2016). Additionally, second
screening turns the relatively passive, information-receptionoriented practice of
watching TV into a complex mix of passive practices (lean-back) and active prac-
tices (lean-forward;Vaccari, Chadwick, & OLoughlin, 2015). To what extent
these active practices might entail, facilitate, or foster various forms of civic activi-
ties are open questions for further research.
Finally, our ndings revealed that the eect of second screening on social media
social capital diers between countries. The eect of second screening on social
media social capital is stronger in countries that score high in a freedom of expres-
sion index than in countries with low freedom of expression. This nding is consis-
tent with prior research on the eect of Internet use on social capital in democratic
media systems (Geber et al., 2016), and shows that, in societies that are freer to
express themselves in the public sphere, second screening can contribute to the
development of social capital on social media platforms. There is also a downside
of this nding. In societies where freedom of expression is reduced, second screen-
ing does not reach its potential to foster social capital.
These conclusions are limited in some ways. First, one should consider that the
concept of social media social capital was introduced recently and, hence, it needs
further renement. For example, while the items used to capture oine social capi-
tal refer to weand people,the social media social capital items make stronger
connections to the individual. This could be a potential confound. Accordingly,
future studies should adjust the wording. One should also consider that our mea-
surement of social media social capital centered on a specic type of social capital:
namely, the community aspect. Future studies may aim to expand the concept of
social media social capital by, for example, taking into account participatory capital
(see Wellman et al., 2001). Moreover, there is a need for future studies to
23Human Communication Research 00 (2019) 132
B. Huber et al. Effects of Second Screening
Downloaded from https://academic.oup.com/hcr/advance-article-abstract/doi/10.1093/hcr/hqz004/5382473 by ICA Member Access user on 01 April 2019
distinguish between social media social capital and online social capital, not only
theoretically, as we did via a literature review, but also empirically, as Gil de
Zúñiga et al. (2017) distinguished empirically between social media social capital
and oine social capital. Related to this point, a key question when investigating
social capital is to what extent it is fungible (see Bourdieu, 1991;Coleman, 1990).
According to Coleman (1990, p. 302), social capital is inherent in the structure of
relationships and only fungible towards specic activities: a given form of social
capital that is valuable in facilitating certain actions may be useless or even harmful
for others.So to what extent is the social capital generated in social media settings
useful for our oine existence? And to what extent is the social capital generated
face-to-face benecial for actions in the digital environment? For example, recent
research has suggested that both oine-generated social capital and social media
generated social capital are positively associated with online political participation,
but only the social capital generated in social media relates to oine political par-
ticipation (Gil de Zúñiga et al., 2017). Hence, more research is needed to investi-
gate the relationship between dierent forms of social capital and various types of
online and oine behavior.
Second, we did not distinguish between second screening for information seek-
ing and second screening for discussions with others. Similarly, we did not distin-
guish between the dierent platforms used for second screening activities (Twitter,
Facebook, etc.). Accordingly, future studies could dierentiate between these and
further motives for second screening, as well as the use of dierent platforms for
second screen activities, and explore the eects of dierent forms of second screen-
ing more in detail. Second, conversations taking place during second screening and
what we called online political discussion in this study might overlap to some
extent. Similarly, the social media social capital measure included communicative
activities. More specically, it measured, among other activities, to what extent
people use social media to encourage conversations about solving community pro-
blems. Hence, a key challenge for communication scholars is to nd a way to better
deal with this measurement issue, by explicitly excluding conversations taking place
during second screening activities from other types of online conversations.
Third, the mediation models were based on pooled sample coecients. There
are likely to be variations in the mediation eect across sub-samples. Moreover, the
eect sizes for the cross-level interactions, are very small. Cross-level interactions
in multi-level regression are notoriously dicult to detect (see, e.g., Mathieu,
Aguinis, Culpepper, & Chen, 2012), because doing so requires at least 15 second-
level groups (in our case, countries) to have enough statistical power. Hence, we
would argue that detecting any cross-level interaction is a noteworthy nding.
However, readers should interpret these eect sizes with caution. In addition, since
the results are based on self-report measures, we do not have access to the content
of the discussions, nor do we know the specic content accessed during second
screening activities. Future studies should test for content eects with dierent
designs. Additionally, an interesting point for future studies could be to see how
24 Human Communication Research 00 (2019) 132
Effects of Second Screening B. Huber et al.
Downloaded from https://academic.oup.com/hcr/advance-article-abstract/doi/10.1093/hcr/hqz004/5382473 by ICA Member Access user on 01 April 2019
second screening activities can vary in terms of their normative values. Although
in our study we focused on pro-civic eects, second screening can also lead to
unhealthy engagement.
Overall, our study shows evidence, across 19 countries, for a positive relation-
ship between second screening and social media social capital. Additionally, it
points out the mediating role of online political discussion in the relationship
between second screening and social media social capital. Finally, the eect of sec-
ond screening on social media social capital is stronger in countries with more
freedom of expression than in countries with less freedom of expression.
Supplementary Material
Supplementary material is available at Human Communication Research online.
Please note: Oxford University Press is not responsible for the content or function-
ality of any supplementary materials supplied by the authors. Any queries (other than
missing material) should be directed to the corresponding author for the article.
Acknowledgments
This research was supported by Grant FA2386-15-1-0003 from the Asian Oce of
Aerospace Research and Development. Responsibility for the information and
views set out in this study lies entirely with the authors.
Notes
1 For an overview on the debate of whether social media increase or decrease exposure to
diverse perspectives, see Flaxman, Goel, and Rao (2016).
2 Individual-country sample sizes: Argentina (1146), Brazil (1086), China (1004), Estonia
(1168), Germany (1054), Indonesia (1080), Italy (1041), Japan (975), South Korea (944),
New Zealand (1157), Philippines (1064), Poland (1060), Russia (1145), Spain (1020),
Taiwan (1008), Turkey (961), the United Kingdom (1064), Ukraine (1223), and the
United States (1161).
3 As we had a large sample size, we used a stricter alpha level and do not report p<.05
and p<.01 in the tables (see Holbert et al., 2018), except in Table 6because there, N=
19 (countries).
4 We also tested the model by using the traditional news use items separately (TV, radio,
newspaper) instead of as a combined measure. The eect of second screening on social
media social capital still held. The results show no signicant eect of TV news (.005)
and radio news (.011), only a signicant eect of printed newspapers (.041).
References
AAPOR American Association for Public Opinion Research (2011). Standard denitions:
Final dispositions of case codes and outcome rates for surveys. Retrieved from http://
aapor.org.
25Human Communication Research 00 (2019) 132
B. Huber et al. Effects of Second Screening
Downloaded from https://academic.oup.com/hcr/advance-article-abstract/doi/10.1093/hcr/hqz004/5382473 by ICA Member Access user on 01 April 2019
Abbas, R., & Mesch, G. (2018). Do rich teens get richer? Facebook use and the link between
oine and online social capital among Palestinian youth in Israel. Information,
Communication & Society, 2,6379. doi:10.1080/1369118x.2016.1261168
Anstead, N., & OLoughlin, B. (2011). The emerging viewertariat and BBC Question Time.
Television debate and real-time commenting online. The International Journal of Press/
Politics, 16, 440462. doi:10.1177/1940161211415519
Appel, L., Dadlani, P., Dwyer, M., Hampton, K., Kitzie, V., Matni, Z. A., ... Teodoro, R.
(2014). Testing the validity of social capital measures in the study of information and
communication technologies. Information, Communication & Society, 17, 398416.
doi:10.1080/1369118×.2014.884612
Ardèvol-Abreu, A., Diehl, T., & Gil de Zúñiga, H. (2018). Building social capital. How the
news and the strength of the ties in the political discussion foster reciprocity. Revista
Internacional de Sociología,76(1), e083. doi:10.3989/ris.2018.76.1.16.147
Armstrong, G. B., Boiarsky, G. A., & Mares, M.-L. (1991). Background television and
reading performance. Communication Monographs, 58, 235253. doi:10.1080/
03637759109376228
Bakshy, E., Messing, S., & Adamic, L. (2015). Exposure to ideologically diverse news and
opinion on Facebook. Science (New York, N.Y.), 348, 11301132. doi:10.1126/science.
aaa1160
Barberá, P. (2014). How social media reduces mass political polarization: Evidence from
Germany, Spain, and the United States. Unpublished manuscript. Retrieved from http://
pablobarbera.com/static/barbera_polarization_APSA.pdf
Barnidge, M. (2015). The role of news in promoting political disagreement on social media.
Computers in Human Behavior, 52, 211218. doi:10.1016/j.chb.2015.06.011
Barnidge, M., Gil de Zúñiga, H., & Diehl, T. (2017) Second screening and political
persuasion on social media. Journal of Broadcasting & Electronic Media, 61, 309331.
doi:10.1080/08838151.2017.1309416
Beaudoin, C. E., & Thorson, E. (2006). The social capital of Blacks and Whites: Diering
eects of the mass media in the United States. Human Communication Research, 32,
157177. doi:10.1111/j.1468-2958.2006.00007.x
Behling, O., & Law, K. S. (2000). Translating questionnaires and other research instruments:
Problems and solutions. Thousand Oaks, CA: Sage.
Bentivegna, S., & Marchetti, R. (2015). Live tweeting a political debate: The case of the
Italia bene comune.European Journal of Communication, 30, 631647. doi:10.1177/
0267323115595526
Bode, L. (2012). Facebooking it to the polls: A study in online social networking and
political behavior. Journal of Information Technology & Politics, 9, 352369. doi:10.
1080/19331681.2012.709045
Bode, L. (2016). Political news in the news feed: Learning about politics from social media.
Mass Communication and Society, 19,2448. doi:10.1080/15205436.2015.1045149
Bonchi, F., Castillo, C., & Ienco, D. (2013). Meme ranking to maximize post vitality in
microblogging platforms. Journal of Intelligent Information Systems, 40, 211231.
doi:10.1007/s10844-011-0181-4
Bosnjak, M., Das, M., & Lynn, P. (2016). Methods for probability-based online and mixed-
model panels selected: Recent trends and future perspectives. Social Science Computer
Review, 34,37. doi:10.1177/0894439315579246
26 Human Communication Research 00 (2019) 132
Effects of Second Screening B. Huber et al.
Downloaded from https://academic.oup.com/hcr/advance-article-abstract/doi/10.1093/hcr/hqz004/5382473 by ICA Member Access user on 01 April 2019
Boulianne, S. (2009). Does Internet use aect engagement? A meta-analysis of research.
Political Communication, 26, 193211. doi:10.1080/10584600902854363
Boulianne, S. (2015). Social media use and participation: a meta-analysis of current
research. Information, Communication & Society, 18, 524538. doi:10.1080/1369118x.
2015.1008542
Boulianne, S. (2017). Revolution in the making? Social media eects across the globe.
Information, Communication & Society,116. doi:10.1080/1369118x.2017.1353641
Bourdieu, P. (1991). Sozialer raum und klassen. Zwei vorlesungen [Social space and classes.
Two lectures]. Frankfurt, Germany: Suhrkamp.
Bowman, L. L., Levine, L. E., Waite, B. M., & Gendron, M. (2010). Can students really
multitask? An experimental study of instant messaging while reading. Computers &
Education, 54, 927931. doi:10.1016/j.compedu.2009.09.024
boyd, d. m., & Ellison, N. B. (2007). Social network sites: Denition, history, and
scholarship. Journal of Computer-Mediated Communication, 13, 210230. doi:10.1111/j.
1083-6101.2007.00393.x
Breheny, P., & Burchett, W. (2017). Visreg: Visualization of regression models. R package
version 2.4-1. Retrieved from https://CRAN.R-project.org/package=visreg
Callegaro, M., Baker, R. P., Bethlehem, J., Goritz, A. S., Krosnick, J. A., & Lavrakas, P. J.
(Eds.). (2014). Online panel research: A data quality perspective. Sussex, England: John
Wiley & Sons.
Chadwick, A., Dennis, J., & Smith, A. P. (2016). Politics in the age of hybrid media: Power,
systems and media logics. In A. Bruns, G. Enli, E. Skogerbo, A. O. Larsson, C.
Christiansen (Eds.), The Routledge companion to social media and politics (pp. 722).
New York, NY: Routledge.
Chadwick, A., OLoughlin, B., & Vaccari, C. (2017). Why people dual screen political
debates and why it matters for democratic engagement. Journal of Broadcasting &
Electronic Media, 61, 220239. doi:10.1080/08838151.2017.1309415
Cho, J., Shah, D. V., McLeod, J. M., McLeod, D. M., Scholl, R. M., & Gotlieb, M. R. (2009).
Campaigns, reection, and deliberation: Advancing an O-S-R-O-R model of
communication eects. Communication Theory, 19,6688. doi:10.1111/j.1468-2885.
2008.01333.x
Coleman, J. (1990). Foundations of social theory. Cambridge, MA: Harvard Univ. Press.
Coppedge, M., Gerring, J., Lindberg, S. I., Skaaning, S.-E., Teorell, J., Altman, D.,
Zimmerman, B. (2016). V-Dem codebook v6. Varieties of democracy (V-Dem) project.
Retrieved from www.v-dem.org.
Correa, T., Bachmann, I., Hinsley, A., & Gil de Zúñiga, H. (2013). Personality and Social
Media Use. In Li, E., Loh, S., Evans, C., and Lorenzi, F. (Eds.) Organizations and Social
Networks: Utilizing Social media to Engage Consumers. (pp. 4161). Hershey, PA: IGI
Global.
Dheer, E., & Courtois, C. (2016). The changing dynamics of television consumption in the
multimedia living room. Convergence: The International Journal of Research into New
Media Technologies, 22,317. doi:10.1177/1354856514543451
Dunbar RIM. 2016 Do online social media cut through the constraints that limit the size of
oine social networks? R. Soc. open sci. 3: 150292. http://dx.doi.org/10.1098/rsos.
150292
27Human Communication Research 00 (2019) 132
B. Huber et al. Effects of Second Screening
Downloaded from https://academic.oup.com/hcr/advance-article-abstract/doi/10.1093/hcr/hqz004/5382473 by ICA Member Access user on 01 April 2019
Dunbar, R. I., Arnaboldi, V., Conti, M., & Passarella, A. (2015). The structure of online
social networks mirrors those in the oine world. Social Networks, 43,3947. doi:10.
1016/j.socnet.2015.04.005
Ellison, N. B., Steineld, C., & Lampe, C. (2007). The benets of Facebook friends:Social
capital and college studentsuse of online social network sites. Journal of Computer-
Mediated Communication, 12, 11431168. doi:10.1111/j.1083-6101.2007.00367.x
Ellison, N. B., Steineld, C., & Lampe, C. (2011). Connection strategies: Social capital
implications of Facebook-enabled communication practices. New Media & Society, 13,
873892. doi:10.1177/1461444810385389
Ellison, N., & Vitak, J. (2015). Social media aordances and their relationship to social
capital processes. In S. Sundar (Ed.), The handbook of psychology of communication
technology (pp. 205227). Boston, MA: Wiley-Blackwell.
Esser, F. (2013). The emerging paradigm of comparative communication enquiry:
Advancing cross-national research in times of globalization. International Journal of
Communication, 7, 113128. doi:10.5167/uzh-91247
Eveland, W. P., Jr. (2001). The cognitive mediation model of learning from the news.
Evidence from nonelection, o-year election, and presidential election contexts.
Communication Research, 28, 571601. doi:10.1177/009365001028005001
Eveland, W. P., Jr. (2004). The eect of political discussion in producing informed citizens:
The roles of information, motivation, and elaboration. Political Communication, 21,
177193. doi:10.1080/10584600490443877
Eveland, W. P., & Hively, M. H. (2009). Political discussion frequency, network size, and
heterogeneityof discussion as predictors of political knowledge and participation.
Journal of Communication, 59, 205224. doi:10.1111/j.1460-2466.2009.01412.x
Flaxman, S. R., Goel, S., & Rao, J. M. (2016). Filter bubbles, echo chambers, and online
news consumption. Public Opinion Quarterly, 80(S1), 298320. doi:10.1093/poq/nfw006
Fukuyama, F. (1995). Trust: The social virtues and the creation of prosperity. New York,
NY: The Free Press.
Ganley, D., & Lampe, C. (2009). The ties that bind: Social network principles in online
communities. Decision Support Systems, 47, 266274. doi:10.1016/j.dss.2009.02.013
Geber, S., Scherer, H., & Hefner, D. (2016). Social capital in media societies. The impact of
media use and media structures on social capital. The International Communication
Gazette, 78, 493513. doi:10.1177/1748048516640211
Gerbner, G., Gross, L., Morgan, M., & Signorielli, N. (1980). The mainstreamingof
America: Violence prole no. 11. Journal of Communication, 30,1029. doi:10.1111/j.
1460-2466.1980.tb01987.x
Giglietto, F., & Selva, D. (2014). Second screen and participation. A content analysis on a
full season dataset of tweets. Journal of Communication, 64, 260277. doi:10.1111/jcom.
12085
Gil de Zúñiga, H., Barnidge, M., & Scherman, A. (2017). Social media social capital, oine
social capital, and citizenship: Exploring asymmetrical social capital eects. Political
Communication, 34,4468. doi:10.1080/10584609.2016.1227000
Gil de Zúñiga, H., Garcia-Perdomo, V., & McGregor, S. (2015). What is second screening?
Exploring motivations of second screen use and its eects on online political
participation. Journal of Communication, 65, 793815. doi:10.1111/jcom.12174
28 Human Communication Research 00 (2019) 132
Effects of Second Screening B. Huber et al.
Downloaded from https://academic.oup.com/hcr/advance-article-abstract/doi/10.1093/hcr/hqz004/5382473 by ICA Member Access user on 01 April 2019
Gil de Zúñiga, H., Jung, N., & Valenzuela, S. (2012). Social media news for news and
individuals social capital, civic engagement and political participation. Journal of
Computer-Mediated Communication, 17,319336. doi:10.1111/j.1083-6101.2012.01574.x
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, 193219. doi:10.1080/
08838151.2017.1309420
Gil de Zúñiga, H., Valenzuela, S., & Weeks, B. (2016). Motivations for political discussion:
Antecedents and consequences on civic engagement. Human Communication Research,
42, 533552. doi:10.1111/hcre.12086
Gillespie, M., & OLoughlin, B. (2015). International news, social media and soft power:
The London and Sochi Olympics as global media events. Participations: Journal of
Audience & Reception Studies, 12, 388412. Retrieved from http://www.participations.
org/Volume%2012/Issue%201/24.pdf
Gottfried, J., Hardy, B. W., Holbert, R. L., Winneg, K., & Jamieson, K. H. (2017). The
changing nature of political debate consumption: Social media, multitasking, and
knowledge generation. Political Communication, 34, 172199. doi:10.1080/10584609.
2016.1154120
Graber, D. A., Bimber, B., Bennett, W. L., Davis, R., & Norris, P. (2004). The Internet and
politics: Emerging perspectives. In H. Nissenbaum & M. E. Price (Eds.), Academy & the
Internet (pp. 90119). New York, NY: Peter Lang Publishing.
Greenberg, D. V. (2008). Causal analysis with nonexperimental panel data. In M. Scott
(Ed.), Handbook of longitudinal research: Design, measurement, and analysis (pp.
259278). New York, NY: Elsevier.
Hallin, D. C., & Mancini, P. (2004). Comparing media systems: Three models of media and
politics. Cambridge, England: Cambridge University Press.
Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional process
analysis: A regression-based approach. New York, NY: The Guilford Press.
Holbert, L., & Grill, C. (2015). Clarifying and expanding the use of conrmatory factor
analysis in journalism and mass communication research. Journalism and Mass
Communication Quarterly, 92(2), 292319. doi:10.1177/1077699015583718
Holbert, R. L., Hardy, B. W., Park, E., Robinson, N. W., Jung, H., Zeng, C., ... Sweeney, K.
(2018). Addressing a statistical power- alpha level blind spot in political- and health-
related media research: discontinuous criterion power analyses. Annals of the
International Communication Association, 42,7592. doi:10.1080/23808985.2018.
1459198
Houston, J. B., Hawthorne, J., Spialek, M. L., Greenwood, M., & McKinney, M. S. (2013).
Tweeting during presidential debates: Eect on candidate evaluations and debate
attitudes. Argumentation and Advocacy, 49, 301311. doi:10.1080/00028533.2013.
11821804
Houston, J. B., McKinney, M. S., Hawthorne, J., & Spialek, M. L. (2013). Frequency of
tweeting during presidential debates: Eect on debate attitudes and knowledge.
Communication Studies, 64, 548560. doi:10.1080/10510974.2013.832693
Jung, N., Kim, Y., & Gil de Zúñiga, H. (2011). The mediating role of knowledge and
ecacy in the eects of communication on political participation. Mass Communication
and Society, 14, 407430. doi:10.1080/15205436.2010.496135
29Human Communication Research 00 (2019) 132
B. Huber et al. Effects of Second Screening
Downloaded from https://academic.oup.com/hcr/advance-article-abstract/doi/10.1093/hcr/hqz004/5382473 by ICA Member Access user on 01 April 2019
Kang, N., & Kwak, N. (2003). A multilevel approach to civic participation. Individual
length of residence, neighborhood residential stability, and their interactive eects with
media use. Communication Research, 30,80106. doi:10.1177/0093650202239028
Kim, J., Wyatt, R. O., & Katz, E. (1999). News, talk, opinion, participation: The part played
by conversation in deliberative democracy. Political Communication, 16, 361385.
doi:10.1080/105846099198541
Kim, S.-H. (2007). Media use, social capital, and civic participation in South Korea.
Journalism & Mass Communication Quarterly, 84, 477494. doi:10.1177/
107769900708400305
Kim, Y. (2011). The contribution of social network sites to exposure to political dierence:
The relationships among SNSs, online political messaging, and exposure to cross-
cutting perspectives. Computers in Human Behavior, 27, 971977. doi:10.1016/j.chb.
2010.12.001
Kleinnijenhuis, J. (2016). Chapter 22: Multilevel regression analysis. In H. Keman & J. J.
Woldendorp (Eds.), Handbook of research methods and applications in political science
(pp. 323340). Cheltenham, UK: Edgar Elward Publishing.
Kraut, R., Kiesler, S., Boneva, B., Cummings, J., Helgeson, V., & Crawford, A. (2002).
Internet paradox revised. Journal of Social Issues, 58,4974. doi:10.1111/1540-4560.
00248
Kwak, N., Williams, A. E., Wang, X., & Lee, H. (2005). Talking politics and engaging
politics: An examination of the interactive relationships between structural features of
political talk and discussion engagement. Communication Research, 32,87111. doi:10.
1177/0093650204271400
La Due Lake, R., & Huckfeldt, R. (1998). Social capital, social networks, and political
participation. Political Psychology, 19, 567584. https://www.jstor.org/stable/3792178
Lin, N. (2008). A network theory of social capital. In D. Castiglione, J. W. Van Deth, & G.
Wolleb (Eds.), The handbook of social capital (pp. 5069). New York, NY: Oxford
University Press.
Lin, T. T. C., & Chiang, Y.-H. (2017). Dual screening: Examining social predictors and
impact on online and oine political participation among Taiwanese Internet users.
Journal of Broadcasting & Electronic Media, 61, 240263. doi:10.1080/08838151.2017.
1309419
Liu, Y., Kim, B., & Kim, Y. (2017, August). Toward engaged citizens: Inuences of second
screening on college studentspolitical knowledge and participation. Paper presented at
the annual conference of the Association for Education in Journalism and Mass
Communication, Chicago, IL.
Ma, L., Lee, C. S., & Goh, D. H. (2013, August). Understanding news sharing in social
media from the diusion of innovations perspective. Presented at the International
Conference on Cyber, Physical, and Social Computing of the Institute of Electrical and
Electronics Engineers (IEEE), Beijing, China. doi:10.1109/GreenCom-iThings-CPSCom.
2013.173.
Mathieu, J. E., Aguinis, H., Culpepper, S. A., & Chen, G. (2012). Understanding and
estimating the power to detect cross-level interaction eects in multi-level modeling.
Journal of Applied Psychology, 97, 951966. Retrieved from http://hermanaguinis.com/
JAP2012.pdf.
30 Human Communication Research 00 (2019) 132
Effects of Second Screening B. Huber et al.
Downloaded from https://academic.oup.com/hcr/advance-article-abstract/doi/10.1093/hcr/hqz004/5382473 by ICA Member Access user on 01 April 2019
McGregor, S. C., & Mourão, R. R. (2017). Second screening Donald Trump: Conditional
indirect eects on political participation. Journal of Broadcasting & Electronic Media,
61, 264290. doi:10.1080/08838151.2017.1309418
McGregor, S. C., Mourão, R. R., Neto, I., Straubhaar, J. D., & Angeluci, A. (2017). Second
screening as convergence in Brazil and the United States. Journal of Broadcasting &
Electronic Media, 61, 163181. doi:10.1080/08838151.2016.1273924
McKenna, K. Y. A., & Bargh, J. (1998). Coming out in the age of the Internet: Identity
demarginalizationthrough virtual group participation. Journal of Personality and
Social Psychology, 75, 681694. Retrieved from http://citeseerx.ist.psu.edu/viewdoc/
summary?doi=10.1.1.477.419
McLeod, J. M., Scheufele, D. A., & Moy, P. (1999). Community, communication, and
participation: The role of mass media and interpersonal discussion in local political
participation. Political Communication, 16(3), 315336, doi:10.1080/105846099198659
Media Insight Project. (2017). Who shared it?: How Americans decide what news to trust
on social media. Retrieved from https://www.americanpressinstitute.org/publications/
reports/survey-research/trust-social-media/.
Moy, P., Scheufele, D. A., & Holbert, R. L. (1999). Television use and social capital: Testing
putnams time displacement hypothesis. Mass Communication and Society, 2,2745.
doi:10.1080/15205436.1999.9677860
Mutz, D. C., & Martin, P. S. (2001). Facilitating communication across lines of political
dierence: The role of mass media. American Political Science Review, 95,97114.
Retrieved from https://www.jstor.org/stable/3117631
Pool, M. M., Koolstra, C. M., & van der Voort, T. H. A. (2003). The impact of background
radio and television on high school studentshomework performance. Journal of
Communication, 53,7487. doi:10.1111/j.1460-2466.2003.tb03006.x
Putnam, R. D. (1993). Making democracy work: Civic traditions in modern Italy. Princeton,
NJ: Princeton University Press.
Putnam, R. D. (1995a). Bowling alone: Americas declining social capital. Journal of
Democracy, 6,6578. doi:10.1353/jod.1995.0002
Putnam, R. D. (1995b). Tuning in, tuning out: The strange disappearance of social capital
in America. Political Science and Politics, 28, 664683. doi:10.2307/420517
Resnick, P. (2001). Beyond bowling together: Sociotechnical capital. In J. M. Carroll (Ed.),
HCI in the New Millenium (pp. 247272). Boston, MA: Addison-Wesley.
Salvucci, D., & Taatgen, N. (2008). Threaded cognition: An integrated theory of concurrent
multitasking. Psychological Review, 115, 101130. doi:10.1037/0033-295x. 115.1.101
Shah, D. V., Cho, J., Eveland, W. P., Kwak, N. (2005). Information and expression in a
digital age. Modeling Internet eects on civic participation. Communication Research,
32, 531565. doi:10.1177/0093650205279209
Shah, D. V., & Gil de Zúñiga, H. (2008). Social capital. In P. J. Lavrakas (Ed.), Encyclopedia
of survey research methods (pp. 824825). Thousand Oaks, CA: Sage Publications.
Shah, D. V., McLeod, J. M., & Yoon, S. H. (2001). Communication, context and
community: An exploration of print, broadcast and Internet inuences. Communication
Research, 28, 464506. doi:10.1177/009365001028004005
Skoric, M. M., Zhu, Q., Goh, D., & Pang, N. (2016). Social media and citizen engagement:
A meta-analytic review. New Media & Society, 18, 18171839. doi:10.1177/
1461444815616221
31Human Communication Research 00 (2019) 132
B. Huber et al. Effects of Second Screening
Downloaded from https://academic.oup.com/hcr/advance-article-abstract/doi/10.1093/hcr/hqz004/5382473 by ICA Member Access user on 01 April 2019
Sum, S., Mathews, M. R., Pourghasem, M., & Hughes, I. (2008). Internet technology and
social capital: How the Internet aects seniorssocial capital and wellbeing. Journal of
Computer-Mediated Communication, 14,202220. doi:10.1111/j.1083-6101.2008.01437.x
Thorson, K., & Wells, C. (2015). Curated ows: A framework for mapping media exposure
in the digital age. Communication Theory, 26, 309328. doi:10.1111/comt.12087
Vaccari, C., Chadwick, A., & OLoughlin, B. (2015). Dual screening the political: Media
events, social media, and citizen engagement. Journal of Communication, 65,
10411061. doi:10.1111/jcom.12187
Valenzuela, S., Arriagada, A., & Scherman, A. (2012). The social media basis of youth
protest behavior: The case of Chile. Journal of Communication, 62(2), 299314. doi:10.
1111/j.1460-2466.2012.01635.x
Vettehen, P. H., Troost, J., Boerboom, L., Steijaert, M., & Scheepers, P. (2017). The
relationship between media content preferences and political participation in 25
European countries: The moderating role of broadband penetration and broadband
access. Communication Research,127. doi:10.1177/0093650217701033
Wang, Y., & Mark, G. (2013). Trust in online news: Comprising social media and ocial
media use by Chinese citizens. CSCQ 13. San Antonio, TX: Association for Computing
Machinery. doi:10.1145/2441776.2441843
Wang, Z., Irwin, M., Cooper, C., & Srivastava, J. (2015). Multidimensions of media
multitasking and adaptive media selection. Human Communication Research, 41,
102127. doi:10.1111/hcre.2015.41.issue-1765
Wang, Z., & Tchernev, J. M. (2012). The mythof media multitasking: Reciprocal
dynamics of media multitasking, personal needs, and gratications. Journal of
Communication, 62, 493513. doi:10.1111/j.1460-2466.2012.01641.x
Wellman, B., Haase, A. Q., Witte, J., & Hampton, K. (2001). Does the Internet increase,
decrease, or supplement social capital? Social networks, participation, and community
commitment. American Behavioral Scientist, 45, 436455. doi:10.1177/
00027640121957286
Wickens, C. D. (2002). Multiple resources and performance prediction. Theoretical Issues in
Ergonomics Science, 3, 159177. doi:10.1080/14639220210123806
Williams, D. (2006). On and othe net: Scales for social capital in an online era. Journal of
Computer-Mediated Communication, 11, 593628. doi:10.1111/j.1083-6101.2006.00029.x
Xenos, M., & Moy, P. (2007). Direct and dierential eects of the Internet on political and
civic engagement. Journal of Communication, 57, 704718. doi:10.1111/j.1460-2466.
2007.00364.x
Zhang, W., & Chia, S. C. (2006). The eects of mass media use and social capital on civic
and political participation. Communication Studies, 57, 277297. doi:10.1080/105109-
70600666974
32 Human Communication Research 00 (2019) 132
Effects of Second Screening B. Huber et al.
Downloaded from https://academic.oup.com/hcr/advance-article-abstract/doi/10.1093/hcr/hqz004/5382473 by ICA Member Access user on 01 April 2019
... Beyond the field of personality and political discussion, research shows that online discussion is predicted by age (Brundidge 2010;Evans and Ulbig 2012;Huber et al. 2019;Kim and Baek 2018;Stromer-Galley 2002) and gender (Huber et al. 2019;Evans and Ulbig 2012;Stromer-Galley 2002). Political interest is a predictor of political discussion online and offline (Evans and Ulbig, 2012;Stromer-Galley, 2002). ...
... Beyond the field of personality and political discussion, research shows that online discussion is predicted by age (Brundidge 2010;Evans and Ulbig 2012;Huber et al. 2019;Kim and Baek 2018;Stromer-Galley 2002) and gender (Huber et al. 2019;Evans and Ulbig 2012;Stromer-Galley 2002). Political interest is a predictor of political discussion online and offline (Evans and Ulbig, 2012;Stromer-Galley, 2002). ...
... Combining these modes would hide these ideological and age differences in patterns of participation. Age is a consistent predictor of online political discussion (Brundidge 2010;Evans and Ulbig 2012;Huber et al. 2019;Kim and Baek 2018;Stromer-Galley 2002). Finally, females are more likely to participate in offline political talk, but less likely to talk on social media (also see: Evans and Ulbig 2012;Huber et al. 2019;Stromer-Galley 2002). ...
Full-text available
Article
Political discussion is a key mechanism for the development of reasoned opinions and political knowledge, but online political discussion has been characterized as uncivil, intolerant, and/or ideologically homogeneous, which is detrimental to this development. In this paper, we examine the role of personality in various forms of political talk—online and offline—as well as like-minded discussion. Based on a 2017 survey conducted in the United Kingdom, United States, and France, we find that people who are open-minded and extraverted are more likely to engage in political talk but less likely to engage in like-minded discussion. Individuals who are older, less educated, introverted, and conscientious are more likely to find themselves in like-minded discussions, both online and on social media. Like-minded discussion is rare; personality, rather than ideology, predicts whether people engage in this form of political talk in online and offline modes. Our findings challenge the role of social media in the creation of like-minded discussion. Instead, we should look to the role of individual attributes, such as personality traits, which create a disposition that motivates the use of social media (and offline networks) to cultivate like-minded discussion.
... The widespread use of dual screens in the networked society raises important political concerns, as an informed citizenry is a foundation of a healthy functioning democracy and political knowledge is a key indicator of democratic citizenship competence (Ran & Yamamoto, 2019). As a result, a strand of literature argues that second screening is a significant predictor of online political participation -that acts as a key link between television news and political engagement (Gil De Zúñiga et al., 2015;Huber, De Zúñiga, Diehl, & Liu, 2019). Studies demonstrate that practices like second screening allow audiences enhanced abilities to shape public narratives, alongside journalistic organizations and political elites (Anstead & O'Loughlin, 2011), for example through commenting on a news event or engaging in conversations about it (Chen, 2019). ...
... The widespread use of dual screens in the networked society raises important political concerns, as an informed citizenry is a foundation of a healthy functioning democracy and political knowledge is a key indicator of democratic citizenship competence (Ran & Yamamoto, 2019). As a result, a strand of literature argues that second screening is a significant predictor of online political participation -that acts as a key link between television news and political engagement (Gil De Zúñiga et al., 2015;Huber, De Zúñiga, Diehl, & Liu, 2019). Studies demonstrate that practices like second screening allow audiences enhanced abilities to shape public narratives, alongside journalistic organizations and political elites (Anstead & O'Loughlin, 2011), for example through commenting on a news event or engaging in conversations about it (Chen, 2019). ...
Full-text available
Article
Contemporary audiences have abundant access to news anywhere, anytime and through a wide range of media. Due to the short length of news items, they do not require prolonged attentional focus. Consequently, the genre is consequential for performing simultaneous activities, which are related (i.e. second screening) or unrelated (i.e. media multitasking) to the content of the news items. Despite the potential of second screen activities to increase the political participation of young people, and the fact that adolescents are heavy users of digital media, there is almost no data on the extent to which they perform such activities. Also, there are almost no comparative data about the rates of news video consumption of adolescents vs. adults. This descriptive study analyzes how adolescents and adults consume news video content in an era of second screens. For four consecutive days, subjects reported their viewing behaviors using a dedicated mobile application. Results demonstrate that adolescents consume significantly less news video content compared to adults; smartphones occupy a more prominent role in their news consumption behavior compared to adults; and they are significantly more involved in Second Screening activities. Implications of the findings for understanding adolescents’ contemporary media ecology are examined. Impact Summary • (a) Prior State of Knowledge Young people are the primary age group that consumes news content mainly on digital platforms, and simultaneously performs multitasking and second screening activities. • (b) Novel Contributions Adolescents’ consumption habits are significantly different from those of adults: they consume less news, the smartphone occupies a more prominent role in their news consumption behavior, and they are significantly more involved in second screening activities. • (c) Practical implications The question of how adolescents consume news content has sparked significant interest of scholars and parents alike. As adolescents are in a developmental stage in which they socialize and acquire habits that will affect them later in life, an understanding of their media consumption habits can have significant consequences contemporarily, and possibly at later stages in their lives. The paper provides novel insights about adolescents’ news consumption patterns.
... Gil de Zúñiga et al. (2014) also found that social media use is positively associated with political participation via political expression. Huber et al. (2019) revealed that those who conduct second screening practices more are more likely to engage in political discussions online, which increases their social capital on social media. As to the cognitive outcomes, one important effect of deliberative interpersonal conversation is that it increases members' intrapersonal self-reflecting on the issue (Pingree, 2007), which significantly mediates the effects of media use on audiences' learning (Eveland, 2001). ...
Full-text available
Article
Commonly conducted alongside political TV viewing, second screening is a new media use behavior that merits explorations at diverse levels. But the research on second screening is still in its infancy. To deeply understand the complicated mechanism of second screening and its influence in a political milieu, this study focuses on the second screening activities aroused by televised political campaign news and programs, and (1) identifies specific categories of second screening activities during television campaign exposure, and (2) establishes the indirect influences of each type of second screening activity on individuals’ cognition (political knowledge) and behavior (political participation) through internal (internal political efficacy) and external (political discussion) factors.
... Más recientemente, se ha asociado esta conducta mediática con un incremento del capital social, gracias a la práctica de la discusión política, lo que se hace más evidente en países con mayores indicadores de libertad de expresión (Huber, Gil de Zúñiga, Diehl, & Liu, 2019). ...
Full-text available
Article
Watching television and tweeting at the same time changes the nature of the media experience; thus, this quantitative research applies a frequency analysis and the seminality index to describe the evolution of this convergence between 2013 and 2017 in Chilean political shows. The comparison showed an increase in tweeting rates in second screening, while TV shows stopped managing their official accounts. This behavior shows a growing autonomy of both screens, consistent with Jenkins' convergency concept: making connections among dispersed media contents.
Article
This study explores how Presidential candidates compete for Twitter attention during a televised debate using two datasets of tweets posted during the first 2012 and 2016 U.S. Presidential debates. Using a time series analysis, we find that both debates exhibited zero-sum attention dynamics such that when one candidate gained attention, the other lost attention. However, our error correction models also revealed that the attention dynamics of Romney and Obama in 2012 produced an equilibrium that neither candidate could break from, whereas in 2016, Twitter attention to Trump overwhelmed Clinton. We therefore conclude that social media attention competition in during the 2012 U.S. presidential debate was more tightly contested compared to 2016.
Article
Media use–social capital research has studied traditional and social media use and associated social capital. Still, little is known about whether social capital would be cultivated or damaged by the use of conversational artificial intelligence (AI). This study explores the associations between conversational AI use and various measures of social capital using a territory-wide survey of an online representative sample in Hong Kong ( n = 1022). The results showed that conversational AI users ( n = 398) were more likely to have more offline and online bonding and bridging social capital, social trust, and civic participation than non-users ( n = 624). For the conversational AI users, intensity and frequency of conversational AI use were the positive predictors of the social capital measures. The findings demonstrated larger effect sizes for online bonding and bridging social capital than offline bonding and bridging social capital.
Full-text available
Article
A generalized climate of distrust in political institutions is not functional to healthy democracies. With the advent of social media, recent scholarly efforts attempt to better understand people's conspiracy theory beliefs in inhibiting institutional trust. This study contributes to this literature by considering the direct antecedent effects of uncertainty avoidance and the moderating role of active social media use—SMU (i.e., interactional SMU, informational SMU, and political expressive SMU). The former is theorized to enable conspiracy theories to thrive, while the latter should cushion the negative effects of conspiracy beliefs on institutional trust. Relying on diverse survey data across different cultures from Europe, the Americas, and New Zealand (N = 11,958) and applying structural equation modeling, findings supported the hypothesized model. In high uncertainty‐avoidance societies, where less well‐known situations are perceived as uncomfortable or downright threatening, conspiracy beliefs proliferate and negatively impact institutional trust. Active SMU attenuates these effects. Via social media, citizens have the ability to strengthen social relationships (interactional SMU), keep themselves informed about the community (informational SMU), and engage in political self‐expression (political expressive SMU), which mitigate conspiracy‐belief negative effects on institutional trust. Future research implications and key limitations of the study are all discussed.
Full-text available
Article
Multiscreening (using a device like a smartphone while watching TV) is pervasive and may have beneficial and detrimental consequences for informed citizenship. This national survey (N = 847) examines how multiscreening during debates and TV news during the 2016 U.S. presidential campaign was related to political learning. Multiscreening was associated with higher levels of thinking about politics (elaboration) and greater confidence about knowledge (information efficacy) but lower actual knowledge. The positive relationship with efficacy yet negative relationship with knowledge raises the possibility that multiscreening contributes to overconfidence. Implications for cognitive resource theories are discussed.
Full-text available
Article
Commonly conducted alongside political TV viewing, second screening is a new media use behavior that merits explorations at diverse levels. But the research on second screening is still in its infancy. To deeply understand the complicated mechanism of second screening and its influence in a political milieu, this study focuses on the second screening activities aroused by televised political campaign news and programs, and (1) identifies specific categories of second screening activities during television campaign exposure, and (2) establishes the indirect influences of each type of second screening activity on individuals’ cognition (political knowledge) and behavior (political participation) through internal (internal political efficacy) and external (political discussion) factors.
Full-text available
Article
This article seeks to explain political persuasion in relation to second screening—people’s use of a second screen (i.e., smartphone/laptop) while watching television to access further information or discuss TV programs. Employing a two-wave-panel survey in the United States, results show this emergent practice makes people more open to changing their political opinions, particularly among those who habitually use social media for news or frequently interact with others in social media contexts.
Full-text available
Article
The pervasive use of multiple technological tools to engage with media and political content (i.e., TV sets, laptops, tablets, smartphones, etc.) has deeply altered the way citizens around the world consume information and discuss public affair issues. Many are using 2, or even several “screens” at the same time to do so, a phenomenon known as second or dual screening. The goal of this article is twofold. First, it introduces a set of novel studies published as a special section devoted to second screening. Second, based on nationally representative original survey data collected in twenty societies (N = 22.033), the study depicts a snapshot of second screening habits for news and politics around the world. Findings reveal that young people tend to second screen more than older counterparts. Similarly, there are also differences in political behaviors between groups of high and low frequency second screen users. More intensive users tend to politically express themselves in social media, and participate more often in offline political activities. On the other hand, results indicate little or no differences between these two groups in terms of their voting behavior.
Full-text available
Article
This study examined the idea that the rise of broadband Internet has contributed to an aggravation of the divide between those who are politically active and those who are not. It was hypothesized that both access to broadband Internet (as an individual-level characteristic) and broadband penetration (as a country-level characteristic) would strengthen the positive relation between relative preferences for political media content and political participation. Analyses were based on data that were collected in the 2010 wave of the European Social Survey (N = 40,582; 25 countries). Political participation was measured both using a voter turnout variable and using a more general political participation scale. Findings from multilevel analyses provide support to the moderating role of both broadband access and broadband penetration, but only when using the general political participation scale.
Full-text available
Article
This study investigates how social factors affect Taiwanese Internet users’ dual screening use and its impact on online and offline political participation. The Web survey recruits 961 dual screen users. Structural equation modeling results show that bridging social capital and perceived social presence are significantly associated with dual screening use. Dual screening is positively related to online and offline political participation. Dual screening use strongly influences offline political participation when mediated by online political participation. Moreover, alternative media trust is positively related to online political participation, whereas mainstream media trust shows a negative association.
Article
A content analysis of 900 political- and health-related media articles published in 11 outlets from 2010 through 2015 reveals a complete disconnect between discussions of statistical power and alpha levels. This study proposes the use of discontinuous criterion power analyses to address this power-alpha blind spot. Additional analyses indicate a sizeable percentage (41.1%) of published works in these areas retaining sufficient statistical power (i.e. .95 or greater) to warrant the use of a stricter alpha level than the traditional α = .05 to detect small effects. The surgical use of more stringent alpha levels to guard against Type I error will instill greater confidence in the empirical findings generated by media research conducted in political and health contexts.
Article
Social networking sites are popular tools to engage citizens in political campaigns, social movements, and civic life. However, are the effects of social media on civic and political participation revolutionary? How do these effects differ across political contexts? Using 133 cross-sectional studies with 631 estimated coefficients, I examine the relationship between social media use and engagement in civic and political life. The effects of social media use on participation are larger for political expression and smaller for informational uses, but the magnitude of these effects depends on political context. The effects of informational uses of social media on participation are smaller in countries like the United States, with a free and independent press. If there is a social media revolution, it relates to the expression of political views on social networking sites, where the average effect size is comparable to the effects of education on participation.
Article
As second screening becomes more widespread, this study addresses its mediating role on the impact of TV news in political participation online and offline, and how this impact varies across groups. We expand the existing line of research by assessing the moderating role of support for Donald Trump on the established mediated model. Through a cross-lagged autoregressive panel survey design applied to the communication mediation model, our results support the link between second screening and political participation—but the mediating role of second screening is contingent upon attitudes towards Trump. For those who do not view Trump favorably, second screening during news leads to a decrease in political participation, both online and offline. As such, this article adds to the communication mediation model by suggesting that discussion and elaboration may not always be positive antecedents to political participation. When individuals disagree with the message dominating TV news and social media, deliberation via second screening leads to political disengagement.
Article
Dual screening during televised election debates is a new domain in which political elites and journalists seek to influence audience attitudes and behavior. But to what extent do non-elite dual screeners seek to influence others, particularly their social media followers, social media users in general, and even politicians and journalists? And how does this behavior affect short- and longer-term engagement with election campaigns? Using unique, event-based, panel survey data from the main 2015 UK general election debate (Wave 1 = 2,351; Wave 2 = 1,168) we reveal the conditions under which people experience agency, empowerment, and engagement now that social media have reconfigured broadcast political television.
Article
Second screening is widespread worldwide, particularly in younger populations. We analyze a survey of college students in Brazil and the United States to compare second screening frequency, types, platforms, and motivations between the two countries. Despite lower Internet penetration, Brazilians second screen significantly more than Americans, a result of the country’s tradition of interacting with producers of television. In both countries, those who use the interactive affordances of social media are more likely to second screen. As such, we posit this unique audience-driven act works to bridge Web-connected devices and television to create a converged atmosphere.