Beaufort & Seethaler: Citizen Participation, Social Media Use, and the Shift in Political Trust
Citizen Participation, Social Media Use, and the Shift in Political
Trust: Results from a European Survey
Paper presented at the
WAPOR regional conference on
“Political Trust in Contemporary Representative Democracies”
Barcelona, November 24-25, 2016
Maren Beaufort
Josef Seethaler
Austrian Academy of Sciences
Institute for Comparative Media and Communication Studies (CMC)
1010 Wien, Postgasse 7
Austria
PHONE: +43-1-51581-3110
FAX: +43-1-51581-3120
E-MAIL: [email protected]
E-MAIL: [email protected]
WWW: http://www.oeaw.ac.at/cmc/cmc/staff/maren-beaufort/
http://www.oeaw.ac.at/cmc/cmc/staff/josef-seethaler/
Please do not quote without the authors’ permission.
1
1 Beaufort & Seethaler: Citizen Participation, Social Media Use, and the Shift in Political Trust
Citizen Participation, Social Media Use, and the Shift in Political Trust:
Results from a European Survey
The most common measures whether democratic citizenship is in decline have been voter
turnout, membership in political and societal organizations, and trust in the leading political
and social institutions such as the government, the parliament, the Supreme Court, or the
military (Schudson, 1998; Putnam, 2000). All these three indicators are prominent issues in
public discourse when it comes to discussing what is often called a growing “democratic
deficit” (Seidle, 2004) or even a “crisis of democracy” (Zittel & Fuchs, 2007). It is commonly
accepted that the only way to take steps against this crisis is to increase the level of political
participation, thus revealing a particular relationship between the aforementioned three
indicators in that institutional trust appears as a prerequisite of the other two indicators, voter
turnout and membership in political and societal organizations, which both can be regarded as
measures of political participation – at least in a traditional sense.
Therefore, it is not surprising that besides suggestions to reform the institutions of
democracy in order to get people more involved in politics there is growing demand throughout
the Western world that the forms of liberal-representative democracy have to be expanded to
include participatory approaches that aim at empowering as many citizens as possible to raise
their voice and to take part in the “common work” of keeping an eye on social problems,
heading them off or introducing them into the political process. While some scholars take a
more skeptical stance arguing that representative democracy as commonly enshrined in the
constitution of most democracies does not allow the implementation of approaches aiming at
comprehensive mass participation (Fuchs, 2007), others recognize the critical role voluntary
civic participation play in helping promote active citizenship, thus representing a first necessary
step to ensure acceptance and functioning of the democratic system (Zukin et al., 1996;
Inglehart & Welzel, 2005).
2
2 Beaufort & Seethaler: Citizen Participation, Social Media Use, and the Shift in Political Trust
In any case, there is widespread agreement that the contemporary erosion of political
participation depends to a certain degree on the growing use of digital communication
technologies – as political participation was always triggered by changes to media
communication forms (Norris, 2000). However, opinions are as divided with regard to the
impact social media might have on political participation, as they are with regard to the impact
participatory approaches might exert on democratic processes. While the more skeptical view
expects an exacerbation of the trend toward increasing political apathy and declining voter
turnout, as the displacement thesis would suggest (Putnam, 2000), more optimistic assumptions
(usually associated with the mobilization thesis) point toward the fostering of political
engagement through the use of online social networks. For example, Gil de Zúñiga, Jung and
Valenzuela (2012), Kim, Hsu and Gil de Zúñiga (2013), and Gil de Zúñiga, Copeland and
Bimber (2014) have shown that, under certain preconditions, use of the Internet or social media
can have a positive impact on social capital and political participation, and, on an interpersonal
level, on inclusion in a heterogeneous range of debates. To shed light on these contradictory
assessments of the political consequences of social media use, we suggest taking a closer look
at the role social media use might play with regard to the various forms of political
participation, how it differs from that of the use of traditional mass media, and, how it is linked
to institutional trust as a major prerequisite of participation.
From “Collective” to “Connective” Action
More recent theoretical considerations point to an ongoing process of societal change in which
established ideas about politics, participation, and the role media plays in society are
superseded (Highfield, 2016; Loader et al., 2014). For example, Bennett and Segerberg’s
(2013) “logic of connective action” approach draws, on the one hand, on the rapidly changing
notions of democracy, but conceptualizes them in accordance with the nature of digitally
3
3 Beaufort & Seethaler: Citizen Participation, Social Media Use, and the Shift in Political Trust
mediated social action, and in contrast with the traditional dynamics of organization-centered
and brokered “collective action” (as formulated by Olson in 1965) with its emphasis on voting
and party membership (see Table 1). From this point of view, “public connections” (Couldry
& Markham, 2006), arising in trustworthy media-based interactions, are assumed to replace
former notions of the importance of groups and group coercion for achieving the common
good, and to contribute to the development of shared, issue-based objectives and the
willingness to take an interest in them.
The different logics of these two dynamics are based on a changed concept of political
participation. The connective dynamic implies a fundamental move away from the established
representative system, along with its group-based identities and interests and its
institutionalized forms of participation, toward more individualized, informal, and spontaneous
forms of participation. The latter are intrinsically motivated by autonomously formulated
concerns and aspirations, which correspond to personal ideas about what constitutes a “good
life”, and are flexibly aimed at various actors not only in the political sphere, but also in the
economic sphere (Ward, 2014): “Taking public action or contributing to a common good
becomes an act of personal expression.” (Bennett & Segerberg, 2012, p. 752) This change has
its origins in increasing individualization with its rise of self-expression values, which weakens
pressures for group conformity (Inglehart & Welzel, 2005), but also in dissatisfaction with the
current status of democratic government and a loss of trust in the established political
institutions (Patterson, 2002; Dahlgren, 2009; Perednea et al., 2012).
Just as traditional mass media corresponds to “collective” behavior in that it acts as a
communication agent that provides the information citizens need to make political decisions,
the structural features characteristic of social network services (SNS) make them an ideal
platform for implementing the specific logic of connective action. They offer individuals the
opportunity to autonomously express, share and discuss their own concerns and ideas in their
4
4 Beaufort & Seethaler: Citizen Participation, Social Media Use, and the Shift in Political Trust
own characteristic way within familiar relationship networks under flexible conditions (i.e.
anytime, anywhere) and at dramatically reduced communication costs. Like offline networks,
online networks are characterized by “weak ties”. In other words, their strength lies in loose
organizational ties and easily personalized operational frameworks, which allow individuals to
contribute their own identities and (hi)stories instead of adapting to collective interpretative
frameworks (Granovetter, 1973). Online communication via SNS, therefore, typically
combines the public/formal sphere and the private/informal sphere (Bimber et al., 2005). The
personal and the political are no longer mutually exclusive, “and separating the two is both
impossible and impractical: they are closely interlinked, encouraged by the conventions and
norms of social media.” (Highfield, 2016, p. 15) Politics is thus drawn out of the arcane realm,
which the public could generally only access indirectly through elections or party membership,
if at all, and is integrated into people’s lives by means of a broad spectrum of social activities
outside of institutional politics, as is typical for connective action. Under the given flexible
conditions of access to content that is of relevance to the public, this can increase people’s
willingness to participate in politics (Bennett et al., 2012; Weinstein, 2014) not least by
overcoming the “freeloader effect” that is typical of collective action (Olson, 1965; Bimber et
al., 2005). Considering all of this, SNS change the overall social context in which people act
with regards to public affairs (Bimber et al., 2012).
Taking into account this process of change from a traditional collective dynamic to a
new connective one, it can be argued that the aforementioned contradictory assessments of the
impact of social media use on political behavior depend, at least to some extent, on different
definitions of participation, which themselves reflect changing notions of democratic life
(Bimber, 2012; Dahlgren, 2013). In liberal-representative democracies, as codified in the
constitutions of many Western countries, political participation (in a “collective” sense) is
regarded as an “activity that has the intent or effect of influencing government action – either
5
5 Beaufort & Seethaler: Citizen Participation, Social Media Use, and the Shift in Political Trust
directly by affecting the making or implementation of public policy or indirectly by influencing
the selection of people who make those policies.” (Verba, Schlozman, & Brady, 1995, p. 38)
In any case, citizens delegate responsibility to political organizations (especially parties) and
their representatives. In doing so, they support the performance of the system, i.e. the functions
of political organizations, executive and legislative organs on various levels, and the various
intermediary actors.
In the context of participatory approaches to democracy, political participation – here
often termed “civic engagement” – has a different and broader frame of reference that refers to
every “voluntary activity focused on problem solving and helping others, […] undertaken alone
or in concert with others to effect change.” (Zukin et al. 2006, p. 7).1 People do not only want
to delegate their sovereign powers, they also want to get involved in political life; they not only
want to monitor the decision making process, they also want to feed it with their own personal
concerns and ideas. Participation in this “connective” sense means a shift from institutional,
duty-based forms to more cause- and civic-oriented forms of engagement (Dalton, 2008), from
being only of instrumental value to the people involved (satisfying extrinsic goals subject to
economic assessment) to being of intrinsic value, which is derived from fostering civic
responsibility and a sense of belonging (Putnam, 2003). Thus, it can be interpreted as a sign of
support for the system as such.
Relating these different notions of participation to the ongoing media change, only the
latter can be expected to be accompanied by shared content production within social networks
and their environment of blurred boundaries between the public and the private sphere, i.e. by
a general use of social media, while the former seems to be more associated with the
consumption of mass media for political-information-seeking purposes (Bennett et al., 2012).
To test the assumption that possible effects of social media use on political participation depend
on how social media is used and how participation is viewed and practiced, and that the kind
6
6 Beaufort & Seethaler: Citizen Participation, Social Media Use, and the Shift in Political Trust
of use and the form of participation differ from the use of traditional mass media and its
consequences, the study focuses on two hypotheses, which represent two sides of a coin. As
Xenos and Moy (2007) have argued, when change in media use patterns has slowed down, at
least to a certain degree, such tests seem to be very fruitful. These two hypotheses read like
this:
H1a: Regular use of online social networks (namely as a general tool for
communicating and sharing with others) is more closely associated with civic
engagement than it is with traditional political involvement.
And vice versa:
H1b: Regular use of mainstream media (namely as a tool for seeking political
information) is more closely associated with traditional political involvement than it is
with civic engagement.
Although a longitudinal study is needed to measure changes in the relationship between media
use and participation, confirmation of the two related hypotheses can be seen as an indication
of the co-existence of two different dynamics of political behavior, one of which is a more
recent development. Both hypothesized effects should be given, even when controlled for the
other respective media use variables, the most important predictors of political participation,
such as informedness, political efficacy, satisfaction with democracy and satisfaction with
personal life (Inglehart, 1977; Verba et al., 1995; Hooghe & Marien, 2013) as well as basic
social stratification markers like gender, age, education, financial situation, community type,
and self-assessed socioeconomic status. “Informedness” refers to the self-assessed state of
being informed about political matters (e.g., Oppenhuis, 1995; Morales, 2009); “political
efficacy” is about the citizens’ feelings “that they have some power to influence the actions of
their government” (Wright, 1981, p. 69); and “life satisfaction” and “satisfaction with
7
7 Beaufort & Seethaler: Citizen Participation, Social Media Use, and the Shift in Political Trust
democracy” are overall measures which are “relatively autonomous” from each other
(Inglehart, 1977, p. 129), with the latter more referring to approval of democratic political
institutions and their work than to support of democracy as an idea. That means, for example,
that people who are disappointed about the way democracy works are not necessarily against
democracy in principle, but wish that it worked better in their country (Schaefer, 2013).
While the previous (two-fold) hypothesis is concerned with the relationship between
media use and participation, the second hypothesis aims at analyzing the role that institutional
trust might play with regard to political participation and in interplay with changing patterns
of media use, thus again referring to a process of transition from a collective to a connective
dynamic. Albeit trust in the democratic institutions can be considered as a core concept in
political participation (e.g. Inglehart, 1977; Hooghe & Marien, 2013), in a connective dynamic,
as already mentioned, institutional loyalties are under question (Bennett et al., 2012; Inglehart
& Welzel, 2005)). Some scholars even have argued that the lower institutional trust the higher
the probability of engaging in non‐institutionalized participation (Kaase, 1999; Hooghe &
Marien, 2013). On the other hand, SNS are becoming some kind of loci of power, which partly
replace, or at least supplement, work done by hierarchical political organizations, because they
provide the necessary structure not only to facilitate expressing, sharing and coordinating
individually motivated concerns and ideas, but to interact with others, to link people to larger
collectives, and thereby to organize politics – even when it must not be understood in terms of
traditional, but of some kind of new “personalized politics” (Bennett, 2012, p. 28). Not
comparable to mass-mediated communication, communication via online social networks
“becomes an organizational process that goes well beyond the exchange of messages” (Bennett,
2012, p. 28), thus turning SNS into a structuring organizational principle in which people gain
trust and confidence (Benkler, 2006; Bennett et al., 2012; Bennett & Segerberg, 2013;
Sweetser, 2014). Given this close relationship between use of and trust in the organizational
structure of SNS, it has to be assumed that the effect of social media use on civic engagement
8
8 Beaufort & Seethaler: Citizen Participation, Social Media Use, and the Shift in Political Trust
(as predicted by H1a) depends on the amount of trust people have in SNS for ensuring the
desired interpersonal connectivity. This constitutes the last hypothesis:
H2: The effect of social network use on civic engagement will be stronger for those
people who have trust in social network services than for those who have not, while
institutional trust loses its importance in predicting civic engagement (as compared to
predicting traditional political involvement).
Data and Method
Considering the global nature of media change, the present study covers new ground in its
attempt to investigate the effects of media use and trust on the various forms of political
participation by applying a transnational design that has been contextualized in a model of
connective action. The study is based on comparative data provided by the Eurobarometer, a
survey carried out on behalf of the European Commission twice a year in all EU countries. It
is based on a multistage, stratified sample of around 1,000 EU citizens aged 15 and older per
country (exceptions being Luxembourg, Malta and Cyprus, with only about 500 participants,
and Germany and the UK, where separate surveys were conducted in West and East Germany
and in Great Britain and Northern Ireland, respectively, resulting in a larger number of
interviews). The random route procedure that was applied begins with a random number of
sampling points, which representatively reflect the structure of the population in relation to its
regional distribution. Despite some weaknesses, the Eurobarometer surveys are an important
source of data for Europe-wide comparative studies, and, not least due to the continual
refinement of their methods and the scientific quality control procedures applied, they serve as
a model for numerous multinational surveys (Kaase & Saris, 1997).
The Eurobarometer offers a unique corpus of data for the present inquiry. One survey
in its 76th round in autumn 2011 was devoted to the topic of media use in the EU. In comparison
9
9 Beaufort & Seethaler: Citizen Participation, Social Media Use, and the Shift in Political Trust
to earlier surveys, this focused strongly on the position of online social networks in political
communication, while a second set of questions surveyed political attitudes and behaviors.
Such a broad repertoire of questions on media use has not been repeated in subsequent surveys,
and so far an analysis of the relationship between the two sets of variables has neither been
carried out in the European Commission’s own Eurobarometer reports (European Commission,
2012) nor in academic literature.
The following study focuses on the EU-15 countries, in other words the member states
before the eastward enlargement of the European Union in January 2004. This restriction was
chosen in the interest of a more homogeneous basis for the study since contextual factors
relating to both the political and media system are seen as having a considerable but as yet
underexplored influence on the matter under investigation here (Perea et al., 2012). The
Eurobarometer 76.3 data was gathered in November 2011 by TNS Opinion & Social using
personal (face-to-face) interviews in the respondents’ national languages (N = 15.498). The
data was then weighted in line with the national population sizes to calculate the aggregate
values and mean values for all EU citizens; the oversampling for Germany’s new federal states
(i.e. former East Germany) and Northern Ireland was also corrected by a weighting factor.
To clarify possible connections between political participation, media use and trust, all
these variables are measured in different ways according to the assumptions about “collective”
and “connective” action. The impact of media use and trust is controlled for the most important
conventional predictors of participation, such as informedness, efficacy, and (dis)satisfaction
with democracy and personal life, as well as for the most basic social stratification markers like
age, sex, education, financial situation, community type, and self-assessed socioeconomic
status (Couldry, Livingstone, & Markham, 2006; Hooghe & Quintelier, 2013).
10
10 Beaufort & Seethaler: Citizen Participation, Social Media Use, and the Shift in Political Trust
Different Forms of Political Participation
The Eurobarometer questions used to operationalize various forms of political participation are
slightly revised versions of the two questions of Inglehart’s “cognitive mobilization indicator”
(Inglehart, 1977; 1990; Inglehart & Rabier, 1979), which aim at measuring what Inglehart,
drawing on Barnes and Kaase (1979), called “unconventional” forms of participation in
contrast to more conventional types that “may require little or no cognitive response to current
issues” (Inglehart, 1990, p. 340). However, in light of the recent changes in participation
behavior outlined here, one question points more to connective, and the other to collective,
action.
This becomes obvious when they are viewed in the context of Ekman and Amnå’s
(2012) typology of latent and manifest political participation, which combines representative
and participatory approaches in democracy theory and thus integrates conventional and newer
forms of participation into one single model (see Table 2). Focusing on “involvement” and
“civic engagement” as the two latent forms, “civic engagement” concerns all “activities based
on personal interest” (Ekman & Amnå, 2012, p. 292; emphasis added), among them the
willingness to engage in personal conversations to try to convince the people in one’s
professional and private life about the things that one considers right and important – a
definition that comes very close to Bennett and Segerberg’s (2013) description of “connective
action.” The relevant “cognitive mobilization” question is: “Do you sometimes try to persuade
your friends, relatives or colleagues to share an opinion that is very important to you? Does
this happen often/from time to time/seldom/never?” (Eurobarometer 76.3, QA3) Defining the
indicator in this way explicitly anchors civic engagement offline, in real-life behavior, and
therefore beyond a purely virtual manifestation of “point-and-click democracy” (similar:
Boulianne, 2009).
11
11 Beaufort & Seethaler: Citizen Participation, Social Media Use, and the Shift in Political Trust
On the other hand, “involvement” can be defined as “a basic curiosity” about political
and societal matters that “precedes both ‘civic’ and ‘political’ activities” (Ekman & Amnå,
2012, p. 293). Thus it depends on how involvement is defined and whether it can be related to
connective or collective behavior (Martín & van Deth, 2007). The wording of the appropriate
Eurobarometer question is limited to purely political matters: “When you are with friends or
relatives, would you say that you discuss national/European political matters often,
occasionally, or never?” (Eurobarometer 76.3, QA2) Thus it can be interpreted as a
precondition for collective action because, as long-term studies have shown, people who are
interested in institutional politics – here operationalized as taking part in political discussions
about European and national matters – are more likely to vote (e.g., Berelson, Lazarsfeld, &
MacPhee, 1963; Verba et al., 1995; van Deth & Elff, 2004). When constructing this dependent
variable, we merged the two aforementioned items into one scale (α = .82) while excluding a
third item mentioned by the Eurobarometer questionnaire – interest in local matters – since
local matters are more likely to be of personal concern and are thus more closely related to
civic engagement (Crouch, 2004). Because connective action is explicitly anchored in offline
behavior, and is thus more “traditional”, the measurement of collective action goes beyond
voting as the “easiest and most widespread form of participation” (Inglehart, 1977, p. 300), and
refers to individual predispositions to attend to politics. Thus avoiding a black-white
framework, the results of the analyses can be expected to be more valid.
In order to avoid over-interpreting the information of the Eurobarometer’s variables,
both indices were dichotomized (see Inglehart, 1990). The first two answer categories (referred
to as “strong” and “medium” in the Eurobarometer questionnaire) were combined, as were the
last two (“low” and “not at all”), yielding the result that 58.9% of respondents are interested in
institutional politics. In a similar manner, the different answers to the question about the
12
12 Beaufort & Seethaler: Citizen Participation, Social Media Use, and the Shift in Political Trust
readiness to try to persuade others were put together into two categories: “often” and “from
time to time” as well as “seldom” and “never” (high level of readiness: 47.1%).
Media Use
As explained previously, the key explanatory variables for the exercise of specific participation
options are the intensity and motivation of media use. When operationalizing these variables,
it should be borne in mind that, in online communication, the boundaries between public and
private spheres are likely to be blurred (Bimber et al., 2005). For the empirical study, this means
an expansion of previous models (which often only include an isolated operationalization of
the “news search” via SNS) toward an operationalization of the intentional and a general use
of online social networks. The general, non-intentional use (Eurobarometer 76.3, QD3.6), is
measured on a six-point scale from “never” to “every day” (M = 2.72; SD = 2.11), and the
intentional use of these networks as a source of political information is represented by an index
based on three “yes/no” questions concerning the deliberate search for political information
about national and European matters as well as for the usefulness of this information
(Eurobarometer 76.3, QD6.3, QD7.3, QD12.3). The reliability analysis of the index yielded a
highly acceptable value of the Kuder-Richardson (KR-20) formula for dichotomous items of
.80 (scale from 0 to 3; M = 1.19; SD = .58).2
The importance of a more general use of SNS for connective political action is
counterbalanced by the significance of the intentional use of traditional mass news media for
seeking political information in collective action. This supposed opposition is based on the
assumption that a change in information and communication technologies leads to changes in
forms of communication, thus transforming the structures of the political public sphere
(Seethaler, 2013). For a long time (and particularly driven by television as the first and only
mainstream medium), keeping oneself politically informed via the media had become a sort of
13
13 Beaufort & Seethaler: Citizen Participation, Social Media Use, and the Shift in Political Trust
social norm and was the prerequisite for qualified political participation within the terms of a
collective dynamic. For the present study, a political media use index has been constructed to
examine the role of the intentional use of mass media for seeking political information. The
additive index is based on several dichotomous questions about the offline and online use of
television, radio, and print media as means of seeking political information on national and
European issues (Eurobarometer 76.3, QD4t.1-3, QD5t.1-3, QD6.2, QD7.2, QD9.3-6, QD12.2;
scale from 0 to 9; M = 4.97; SD = 2.16). The reliability index value (KR-20 = .67) is not
especially high, but acceptable. In addition, the general use of television, radio and printed
media (Eurobarometer 76.3, QD3.1+2,3,4) was each measured on a six-point scale from
“never” to “every day” in the Eurobarometer and combined into a media use index (α = .79; M
= 4.92; SD = 1.05).
Trust
Considering differences in trust levels between national and international institutions, two
additive indices have been constructed to measure trust in political institutions of the nation
states and in the most important institutions of the European Union, comprising two and three
dichotomous items, respectively. The two questions that make up the “national trust index” are
aimed at the respective national government and national parliament (KR-20 = .80; scale from
0 to 2; M = .54; SD = .81; Eurobarometer 76.3, QA10.6-7); the “EU trust index” consists of
three questions which measure trust in the European Commission, the European Parliament,
and the European Council (KR-20 = .89; scale from 0 to 3; M = .99; SD = 1.27; Eurobarometer
76.3, QA14.1-3).
The index for trust in SNS is based on agreement (measured on a four-point scale from
“agree completely” to “disagree completely”) with the following statements in the
Eurobarometer questionnaire: that social networks “offer a modern way to keep abreast of
14
14 Beaufort & Seethaler: Citizen Participation, Social Media Use, and the Shift in Political Trust
political affairs,” “get people interested in political affairs,” and “offer a good way to have your
say on political issues.” A fourth available item was excluded after reliability analysis (α = .83;
scale from 1 to 4; M = 2.70; SD = 0.76). For all the questions about trust in relation to online
social networks there was – in contrast to all the other items cited here – a large number of
“don’t know” answers (from 27.3 to 28.2%). In view of the sensitivity of these items, in the
statistical analyses the missing answers were not included.
Control Variables: Political Attitudes and Sociodemographic Characteristics
The control variables selected serve to test the connections observed between media use and
political participation in connective and collective structures. They encompass political
attitudes like informedness, efficacy, and satisfaction with democracy and personal life, which
are usually considered as strong predictors of participation, as well as some of the most basic
and important social stratification markers: age, sex, education, financial situation, community
type, and socioeconomic status (SES).
The Eurobarometer provides an “informedness” index on a four-point scale, ranging
from “very well” to “very badly”, based on the respondents’ subjective assessment of how well
informed they felt they were about European matters (Eurobarometer 76.3, QD2; M = 2.81;
SD = .73). The highly disputed (Arzheimer, 2008) operationalization of political efficacy
follows the example of several studies (e.g., Bennett, 1997; Anderson & Tverdova, 2001;
Bachmann et al., 2010; Gil de Zúñiga et al., 2014) in utilizing a single measure, such as whether
“my voice counts in the EU” (Eurobarometer 76.3, QA19a.3: 26.7%). Satisfaction with
democracy was based on two questions asking whether the respondent is satisfied with the way
democracy works in her or his country as well as in the European Union (measured on a fourpoint scale from “very satisfied” to “not at all satisfied”; Eurobarometer 76.3, QA18.a,b; α =
.73; M = 2.64; SD = .74). For the “personal satisfaction” variable, the Eurobarometer
15
15 Beaufort & Seethaler: Citizen Participation, Social Media Use, and the Shift in Political Trust
questionnaire provides the question (QA1): “On the whole, are you very satisfied, fairly
satisfied, not very satisfied or not at all satisfied with the life you lead?”. Again, answers were
measured on a four-point scale (M = 2.04; SD = .77).
The sociodemographic variables (Eurobarometer 76.3, D8, D10, D11, D25, D61, QC2)
include age, gender (female: 51.5%), the level of education (evaluated using a four-point scale:
less than high school, high school graduate, some college training, college graduate; the median
was high school graduate), the type of community the respondent lives in (rural village =
32.8%; small or middle sized town = 43.2%; large town = 24.0%), and the socioeconomic
status (self-assessment on a scale from 1 to 10; M = 5.60; SD = 1.53). The financial situation
of the household is reflected by three categories describing the current situation as not allowing
the respondent “to make any plan for the future” (34.3%), to make plans for “the next six
months” (30.7%), to have “a long-term perspective of what [their] household will be in the next
one or two years.”3 Finally, because the data set encompasses 15 European countries, all
regressions were controlled for possible country effects.
Results
The first hypothesis expects the use of online social networks – especially as a general tool for
communicating and sharing – to be more strongly associated with civic engagement than with
involvement in traditional institutional politics. We tested this expectation in two ways: firstly
with simple and partial correlations between various forms of media use and participation, and
secondly by applying logistic regression models (because of the dichotomous nature of the
dependent variables) in order to control for the influence of other factors, such as political
attitudes and sociodemographic characteristics, as well as to examine the odds ratios for the
various effects.4 Because of the different scales of measurement, a z-transformation of all
interval-scaled predictor variables was carried out to create a common metric.
16
16 Beaufort & Seethaler: Citizen Participation, Social Media Use, and the Shift in Political Trust
Use of social media and use of mainstream media were, to a great extent, revealed as
different dimensions since seeking political information in mainstream media and seeking
political information in social media were not significantly correlated (r = .015, p = .065) and
seeking political information in social networks was negatively correlated with general media
use (r = -.029, p < .001). Conversely, seeking political information in mainstream media was
negatively correlated with general social media use (r = -.022, p = .006). Only general use of
both mainstream and social media was slightly correlated (r = .025, p = .020). However, for
both kinds of media, general and intentional use were strongly related to each other (social
media: r = .363, p < .001; mainstream media: r = .475, p < .001). We therefore ran partial pointbiserial correlation analyses to test the association between the various forms of media use and
participation. (Point-biserial correlation analysis is employed because one variable is expressed
as interval data and the other variable is represented by a dichotomous nominal scale.)
Consistent with our expectation, the partial point-biserial correlation (with control for
intentional use of SNS) is 0.126 (p < .001) between general use of SNS and civic engagement,
and 0.002 (p = .804) between general use of SNS and political involvement (see columns 1 and
2 on Table 3). With regard to the relationship between the various types of media use and civic
engagement, all other correlation coefficients are smaller than .126, and compared to the
correlation of general use of SNS and civic engagement, the differences in the strength of the
correlations are statistically significant, according to Steiger’s (1980) z-test for overlapping
dependent correlations (see upper part of Table 4). Concerning the comparison of each pair of
correlations between a particular type of media use and both forms of participation,
standardized scores show that most of these differences are statistically significant (see column
3 on Table 3). Only general use of SNS, however, is more closely associated with civic
engagement than it is with traditional political involvement (z = 12.708, p < .000). All other zscores point in the opposite direction, with intentional use of social networks for seeking
17
17 Beaufort & Seethaler: Citizen Participation, Social Media Use, and the Shift in Political Trust
information playing a comparatively minor and rather similar role for both forms of
participation (z = -1.675, p = .052). These tests support H1a.
On the other hand, the partial point-biserial correlation (with control for general
mainstream media use) is .246 (p < .001) between intentional mainstream media use and
political involvement, and .102 (p = < .001) between intentional mainstream media use and
civic engagement. Accordingly, the intentional use mainstream media is more closely
associated with traditional political involvement than it is with civic engagement (z = 15.035,
p < .000). Although the latter correlation is also positive, it is not as strong as the one between
general SNS use and civic engagement (z = 2.078 in favor of general use of SNS, p = .019).
However, regarding political involvement, correlation with intentional mainstream media use
is by far the strongest, and compared to the correlation between involvement and all other types
of media use, the differences in the strength of the correlations are statistically significant (see
the results of Steiger’s z-test in the lower part of Table 4). H1b is therefore also supported.
To explore these relationships further, we constructed two logistic regression models:
one for each dependent variable (see Table 5, Models 1a and 2). Originally, the independent
variables were entered in several separate blocks (in order to determine the amount of variance
explained by the various sets of predictors), but in this paper only the overall results are
reported. The regression analyses provide additional clear evidence of the assumed
relationships between various kinds of media use and different forms of participation. As
expected (and already substantiated by the results of the correlation analyses), the different
emphases in media use correlate with different forms of political participation. When it comes
to civic engagement (Table 5, Model 1), frequency of general use of SNS is the most important
media use indicator in predicting this kind of participation (as indicated by the odds ratio
values), while their intentional use plays only a minor role. This means that accessing
politically relevant content can incidentally increase people’s willingness to engage in socially
18
18 Beaufort & Seethaler: Citizen Participation, Social Media Use, and the Shift in Political Trust
relevant matters. The specific nature of SNS to combine the public and private spheres clearly
promotes (and does not jeopardize) political participation in a broader sense. Contrarily,
mainstream mass media has to be primarily used intentionally as a source of political
information, but even in this respect there is only a comparatively minor influence on civic
engagement.
On the other hand, the intentional use of mass media as a source of political information
is by far the most decisive media use predictor of political involvement as a more latent form
of participation in the sense of collective action (Table 5, Model 2a). Albeit the general use of
mass media also plays some role, the probability that people become involved in politics
because of intentional use of mass media is about three times higher (Exp(b) = 1.431) than
because of a more general media use (Exp(b) = 1.143). In contrast, SNS do not have any
relevance in the context of traditional political involvement. All together, these results are
further confirmation of both hypotheses H1a and H1b in terms of connective and collective
action.
All media effects are not only controlled for country effects (which are not further
discussed in this paper5), but also for both political attitudes (informedness, efficacy, and
satisfaction with democracy as well as with personal life) and social stratification factors. Most
control variables become less important – such as (high) level of education, (high)
informedness, (high) efficacy, and the type of community (large town) – or lose any importance
– such as (good) financial situation and (high) socioeconomic status – when it comes to civic
engagement. Gender has a similar impact on both measures of participation (with women
playing a less significant role); and with regard to both satisfaction measures results are in line
with previous studies which showed that “post-materialists” are not only more likely to be
dissatisfied with the current status of democracy than more traditionally oriented people, but
they tend to be more inclined to change the status quo (Inglehart, 1977; 1990). On the other
19
19 Beaufort & Seethaler: Citizen Participation, Social Media Use, and the Shift in Political Trust
hand, life satisfaction, which is usually assumed to vary little from one social group to another,
seems to be a general prerequisite for political participation. However, a much more decisive
difference can be observed between the effects of age on the various forms of participation.
While older people show a significantly higher level of traditional political involvement, they
display a significantly lower level of civic engagement. This seems to indicate an ongoing shift
in the forms of political participation.
The second hypothesis (H2) predicts that the effect of social media use on civic
engagement depends on the amount of trust people have in SNS as organizational units for
ensuring the desired connectivity. This effect should be given even when controlled for trust in
conventional political institutions, which is assumed to lose importance for participation (as
compared to predicting traditional political involvement). To test this assumption we examined
the interaction effects of general use of social media with trust in them (Table 5, Model 1b).
The results indicate that trust in SNS positively moderates the relationship between
respondents’ use of SNS and their participation in civic activities, but only to a small extent (p
= .10). This means that when there is high trust in the organizational structure provided by
SNS, an increase in their use as tools for political communication increases the impact on civic
engagement. Low trust, however, has a reverse effect (see Figure 1).
The moderating effect of trust in SNS is controlled for trust in political institutions. In
order to explore the assumed changes in trust, we computed a second model for predicting
traditional political involvement, which includes – besides all media use variables and control
variables – all three trust variables, but does not include the interaction effect (Table 5, Model
2b). As it could have been expected, trust in SNS does not play any role in a collective action
dynamic. At both the national and the European level trust in democratic institutions has a
significant but not particularly high impact on involvement, while no shift in trust from political
institutions to SNS could be observed. As expected, they only serve as organizational units of
connective action. Contrarily, trust in national political institutions loses any importance, when
20
20 Beaufort & Seethaler: Citizen Participation, Social Media Use, and the Shift in Political Trust
it comes to civic engagement, while trust in European institutions retains its significance as a
predictor of participation also in a connective action dynamic. However, its β-value (β = .063,
p < .05) differ significantly from the one in the political involvement model (β = .118, p < .001).
Taken all together, H” can be confirmed, but only to a lesser extent than expected.
Discussion and Conclusion
Decline in trust in democratic institutions has opened the way to considering the role
participatory concepts of democracy and citizenship in fostering popular acceptance of the
democratic system. These concepts unsettle liberal notions of representative government and
rational citizens who engage in politics through traditional political institutions and media
outlets, and they promote voluntary activities aimed at solving social problems as acts of
participation, regardless of whether they were undertaken alone or in concert with others to
effect change. This ongoing transformation of democratic society is accompanied by an ever
more widespread use of online social networks that revolutionize the way people communicate.
They do this not only by providing new opportunities to connect with others, but also – and
perhaps in even more importantly in the realm of political communication – by blurring the
boundaries between the public and private sphere.
To analyze possible connections between the various forms of political participation (as
a dependent variable) and the various kinds of use of SNS (compared to traditional media use),
the present study is based on a theoretical contextualization formulated by Bennett and
Segerberg (2012; 2013). Here, the changing character of participation and the role of the media
are conceptualized within a model of so-called “connective” action (in contrast to traditional
“collective” action). As a new dynamic, connective action is fundamentally different from the
institutionalized forms of collective action, and means a move away from the established
political system and its group-based identities and interests toward individualized, media-
21
21 Beaufort & Seethaler: Citizen Participation, Social Media Use, and the Shift in Political Trust
networked forms of civic engagement. Within this connective environment (in which
particularly young people operate), SNS are not only supposed to bring about politically
relevant media-based interactions (not least due to the blurring of boundaries between public
and private life), but also to function as key organizational units by guaranteeing the desired
connectivity. Therefore, trust in the organizational power of SNS is assumed to take the place
of one of the previous driving forces of political participation: trust in conventional political
institutions. This also implies changes in the function of traditional mass media, which is
gradually superseded as a source of political information.
To capture this fundamental social change (and thereby to clarify some contradictory
assessments of the effects of social media use), participation was measured in two different
ways, reflecting various notions of democracy: as political involvement (according to
“collective” action in representative democracy) and as civic engagement (according to
“connective” action in participatory democracy). The indicators used are based on Inglehart’s
“cognitive mobilization” approach. They measure latent, or “unconventional”, forms of
participation (in order to provide comparable results). Political involvement is evidenced by
interest in institutional politics, and civic engagement by a willingness to persuade others to
share one’s own perspective on social issues that are seen as personally relevant.
Using these indicators, analysis of the EU-15 data from the autumn 2011
Eurobarometer survey clearly shows a co-existence of patterns of collective and connective
action. As assumed in hypotheses H1a and H1b, SNS play a significant role when it comes to
civic engagement while the use of mainstream mass media is closely associated to political
involvement. However, it is not only the frequency, but also the kind of use that matters.
Mainstream mass media retains its position as the prime source of political information –
particularly when somebody is interested in institutional politics. In a “collective”
environment, social media is only able to exert a very small and insignificant influence on
22
22 Beaufort & Seethaler: Citizen Participation, Social Media Use, and the Shift in Political Trust
participation when, like traditional mass media, it is intentionally used for seeking political
information. On the other hand, SNS unfold their potential for fostering civic engagement when
they are used as a general tool for communicating and sharing all kinds of everyday content. It
can be argued that the effect of this general, not intentionally political, use of social media is
attributable to the blurred boundaries between the public and the private spheres (for similar
results regarding the Internet in general, see Cantijoch, 2012; Perea et al., 2012), which allow
personal concerns that might become politically relevant issues to be shared: “The personal is
political on social media.” (Weinstein 2014, 2010) This interpretation is underlined by the
decreasing importance of how well informed one feels about and how much one feels they have
a say in institutional politics, because, in connective structures, conventional political decisionmakers no longer constitute the primary reference points of political and social activities
(Bennett & Segerberg, 2012, p. 752). The odds ratios for the informedness and efficacy
variables show that the expected increase of civic engagement (for a one-unit change in the
independent variable) is more than 43 percent and almost 12 percent, respectively, lower than
the expected increase of political involvement (Exp(b) = 1.218 vs. 1.710; 1.174 vs. 1.293). On
the other hand, satisfaction with democracy has a significantly negative impact only on civic
engagement, not on political involvement (Exp(b) = .936 vs. .962), or, to put it the other way
round, dissatisfaction with how democracy works leads to civic engagement, but it does not
lead to political involvement. Though there is no reason to expect that people are going to
perceive institutional politics as obsolete, there seems to be a trend of perceiving politics
outside of the traditional political realm.
In any case, use of social media coincides with an expansion of a circle of people who
become actively engaged in civic life. Most importantly, age is one of the most decisive
predictors of both forms of participation. As expected, the younger the people, the more they
are inclined to civic engagement, while older people display a significantly higher level of
23
23 Beaufort & Seethaler: Citizen Participation, Social Media Use, and the Shift in Political Trust
traditional political involvement. This can be seen as an indication for the co-existence of two
different dynamics of political action, with the latter as the more recent one. It seems to be
justified to interpret this co-existence as a sign of a shift from collective to connective action.
The findings of our study provide some more evidence of an ongoing change in political
behavior. As the results for the control variables show, the typical representative of traditional
political involvement is a better-educated, financially well-situated, wealthy male who lives in
big cities and finds himself in a good position in society. Most of these characteristics are
becoming obsolete, or at least less important, when it comes to civic engagement. People who
are not involved in “collective” structures because of their low level of education, low income,
rather low socioeconomic status and the rural environment they live in, and who would
therefore be considered politically apathetic, can be more easily reached through connective
structures – even in the context of politically relevant issues.
With exception of an enduring gender gap, the findings presented here suggest that SNS
do have the potential to mobilize people. However, one of the greatest challenges currently
facing politics is to discover what consequences this might have. This seems not to be an easy
task, because trust in the conventional political institutions is on the wane. Even in the context
of traditional political involvement trust in both national and European political institutions is
among the less influential political attitudes variables included in the analysis. Regarding EU
institutions, the expected increase of political involvement (in case of a one-unit change in the
independent variable) is only 11.8 percent – and this low increase is cut in half with regard to
civic engagement. Trust in national political institutions is slightly more important in the
context of involvement (Exp(b) = 1.151), but does not count in predicting engagement. On the
other hand, the results of our analyses provide some small evidence about a shift in trust from
conventional political institutions to SNS as organizational units of political action.
In contrast to traditional mass media SNS does not refer to the content of
communication, but to the organizational structure provided to ensure the connectivity
24
24 Beaufort & Seethaler: Citizen Participation, Social Media Use, and the Shift in Political Trust
necessary for communicating and sharing ideas, building relationships, and engaging in
political and societal matters. Therefore, we expected trust in SNS as key organizational units
of connective action to moderate the relationship between social media use and civic
engagement. This assumption can be confirmed only to a certain extent in that trust in SNS
seems to increase the readiness to use them as a tool that is able to participate in and organize
civic activities. Thus SNS fulfill, at least to some extent, functions usually attributed to political
institutions which are, in a democratic system, obliged to provide “opportunities, mechanisms
and the space within which to take part in society” (Anheier, Stares, & Granier, 2004, p. 98).
The present study breaks new ground in its attempt to base the analysis of the
relationship between media use, trust and participation on a broad data set, representative of
the populations of fifteen European countries. This broad data set (around 15,000 interviews)
makes it possible to reach conclusions with greater validity (Kaase & Saris, 1997).
Nevertheless, the study has several shortcomings. For instance, the Eurobarometer can only be
used to operationalize factors on the micro level, yet the changes to the relationship between
media and participation also have a macro dimension. This is because contextual factors
relating to both the political system and the media system are often seen as having a
considerable but as yet underexplored influence on this relationship (Perea et al., 2012).
Furthermore, survey data that has not been gathered specifically for the purpose of this study
is far from ideal for operationalizing a theoretical approach since the definitions of the variables
are, out of necessity, based on the formulation of the questionnaire. So as not to over-interpret
the results of the regression analyses, the values of the dependent variables have been
dichotomized, the strictest possible standards were applied to the construction of the indices,
and a conservative approach was taken when dealing with missing values and carrying out
statistical testing procedures. Furthermore it must be noted that the findings presented are just
a snapshot from the year 2011 while the processes and developments in a transitional period –
25
25 Beaufort & Seethaler: Citizen Participation, Social Media Use, and the Shift in Political Trust
especially the causal relationships – can only be adequately captured using time series analysis.
Unfortunately, later Eurobarometer surveys do not provide comparable and similarly extensive
data.
Moreover, the analyses presented here leave room for improvement. The findings
presented are based on four models, which were founded on theory and tested using logistic
regression equations. An obvious additional step, however, would be to test path models, which
represent mutual dependencies between the predictors. For example, both satisfaction variables
could be modeled as preconditions for the level of trust in democratic institutions, and interest
in institutional politics (as a latent form of political involvement) could be modeled not just as
a dependent variable, but also as an intervening variable for the explanation of manifest forms
of participation like voting, which are losing ground in Western democracies (cf. e.g. Luskin,
1990; Couldry et al., 2006).
Notes
1
In political science literature, the terms “civil/civic participation” and “engagement” are sometimes used as
synonyms (e.g. Zukin et al. 2006); at times they are differentiated according to whether they focus on behavior
(“participation” in a more narrow sense) or attitudes (“engagement”) (Barrett & Brunton-Smith, 2014; Zani &
Barrett, 2012). Some authors propose “civic engagement” as an umbrella term, encompassing “political” and
“social participation” (Gabriel, 2012).
2
For the construction of the indices, the answers to the individual items were re-coded in reverse order when
necessary.
3
Missing values were set to 0, where 0 marks the value that is not of interest: if, for the example, the question
was concerned with satisfaction in democracy, then it was assumed that missing answers did not indicate
satisfaction. For SES (which was measured on a 10-point scale), missing answers were replaced by the overall
mean. Both procedures are admissible when the missing values constitute up to 5% of the population. This
proportion is not exceeded among the variables used in this analysis.
4
The Hosmer & Lemeshow goodness-of-fit test was carried out for all the regression models described in Table
5. In each case, the standardized residuals were tested, and the existence of influential cases and multicollinearity
was ruled out by approximation using a linear regression.
5
When the regression equations are not controlled for country effects, only two major changes in the b-values
for the independent variables can be observed. With respect to civic engagement, only socioeconomic status
becomes more important (b = .061, Exp(b) = 1.063; Wald χ2 = 10.677, p < .05). Furthermore, the negative impact
of general social media use on political involvement becomes statistically significant (b = -.059, Exp(b) = .942;
Wald χ2 = 4.583, p < .05). These few exceptions aside, the EU-15 countries represent for the most part a rather
politically and culturally homogeneous area. It has to be added that, strictly speaking, the countries of the
respondents represent a second level of analysis. However, an LR-test has revealed that the multilevel null model
does not differ significantly from the individual level null model (Δ deviance = 3.3, p > .05).
26
26 Beaufort & Seethaler: Citizen Participation, Social Media Use, and the Shift in Political Trust
References
Anderson, C. J., & Tverdova, Y. V. (2001). Winners, losers, and attitudes about government in
contemporary democracies. International Political Science Review, 22, 321–338.
Anduiza, E., Jansen, M. J., & Jorba, L. (Eds.) (2012). Digital media and political engagement
worldwide. Cambridge: Cambridge University Press.
Anheier, H. K., Stares, S., & Grenier, P. (2004). Social capital and life satisfaction. In W. Arts
& L. Halman (Eds.), European voters at the turn of the millennium (pp. 57-80). Leiden:
Brill.
Arzheimer, K. (2008). Political efficacy. In L. L. Kaid & C. Holtz-Bacha (Eds.), Encyclopedia
of political communication. London: Sage, 2008.
Bachmann, I., Kaufhold, K., Lewis, S. C., & Gil de Zúñiga, H. (2010). News platform
preference: Advancing the effects of age and media consumption on political participation.
International Journal of Internet Science, 5(1), 34–47.
Barnes, S. H., & Kaase, M. (1979). Political action: Mass participation in five Western
democracies. Beverly Hills, CA: Sage.
Barrett, M., & Brunton-Smith, I. (2014). Political and civic engagement and participation:
Towards an integrative perspective. Journal of Civil Society, 10, 5–28.
Bennett, S. E. (1997). Knowledge of politics and sense of subjective political competence.
American Politics Quarterly, 25, 230–240.
Bennett, W. L. (2012). The personalization of politics: Political identity, social media, and
changing patterns of participation. The Annals of the American Academy of Political and
Social Science, 644 (November), 20–39.
27
27 Beaufort & Seethaler: Citizen Participation, Social Media Use, and the Shift in Political Trust
Bennett, W. L., Freelon, D. G., Hussain, M. M., & Wells, C. (2012). Digital media and youth
engagement. In H. A. Semetko & M. Scammell (Eds.), The SAGE handbook of political
communication (pp. 127–140). Los Angeles, CA: Sage.
Bennett, W. L., & Segerberg, A. (2012). The logic of connective action. Information,
Communication & Society, 15(5), 739–768.
Bennett, W. L., & Segerberg, A. (2013). The logic of connective action: Digital media and the
personalization of contentious politics. Cambridge: Cambridge University Press.
Berelson, B. N., Lazarsfeld, P. F., & MacPhee, W. N. (1963). Voting: A study of opinion
formation in a presidential campaign. Chicago, IL: University of Chicago Press.
Bimber, B. (2012). Digital media and citizenship. In H. A. Semetko & M. Scammell (Eds.),
The SAGE handbook of political communication (pp. 115–126). Los Angeles, CA: Sage.
Bimber, B., Flanagin, A., & Stohl, C. (2005). Reconceptualizing collective action in the
contemporary media environment. Communication Theory, 15(4), 365–388.
Bimber, B., Flanagin, A., & Stohl, C. (2012). Collective action in organizations: Interaction
and engagement in an era of technological change. Cambridge: University Press.
Boulianne, S. (2009). Does Internet use affect engagement? A meta-analysis of research.
Political Communication, 26(2), 193–211.
Boulianne, S. (2015). Social media use and participation: A metaanalysis of current research.
Information, Communication & Society, 18(5), 524–538.
Cantijoch, M. (2012). Digital media and offline participation in Spain. In E. A. Perea, M. J.
Jensen & L. Jorba (Eds.), Digital media and political engagement worldwide. A
comparative study (pp. 118-137). Cambridge: Cambridge University Press.
Couldry, N., Livingstone, S., & Markham, T. (2006). Media consumption and the future of
public connection (Report). London: The London School of Economics and Political
Science.
28
28 Beaufort & Seethaler: Citizen Participation, Social Media Use, and the Shift in Political Trust
Couldry, N., & Markham, T. (2006). Public connection through media consumption: Between
oversocialization and de-socialization? The Annals of the American Academy of Political
and Social Science, 608 (November), 251–269.
Crouch, C. (2004). Post-democracy. Cambridge: Polity Press.
Dahlgren, P. (2009). Media and political engagement: Citizens, communication and
democracy. Cambridge, MA: Cambridge University Press.
Dahlgren, P. (2013). The political web: Media, participation and alternative democracy.
Houndmills: Palgrave-Macmillan.
Dalton, R. (2008). The good citizen: How a younger generation is reshaping American politics.
Washington, CD: CQ Press.
Ekman, J., & Amna, E. (2012). Political participation and civic engagement: Towards a new
typology. Human Affairs, 22, 283–300.
European Commission. (2012). Media use in the European Union (Standard Eurobarometer
Report No. 78). Brussels.
Farrell, H. (2012). The consequences of the Internet for politics. Annual Review of Political
Science, 15, 35–52.
Fuchs, D. (2007). Participatory, liberal and electronic democracy. In T. Zittel & D. Fuchs (Eds.),
Participatory democracy and political participation: Can participatory engineering bring
citizens back in? (pp. 29-54). London: Routledge.
Gabriel, O. W. (2012). Political participation in France and Germany: Traditions, concepts,
measurements, patterns and explanations. In O. W. Gabriel, S. I. Keil & E. Kerrouche
(Eds.), Political participation in France and Germany (pp. 1–32). Colchester: ECPR Press.
Gil de Zúñiga, H., Jung, N., & Valenzuela, S. (2012). Social media use for news and
individuals” social capital, civic engagement and political participation. Journal of
Computer‐Mediated Communication, 17, 319–336.
29
29 Beaufort & Seethaler: Citizen Participation, Social Media Use, and the Shift in Political Trust
Gil de Zúñiga, H., Copeland, L., & Bimber, B. (2014). Political consumerism: Civic
engagement and the social media connection. New Media & Society, 16(3), 488–506.
Granovetter, M. (1973). The strength of weak ties. American Journal of Sociology, 78, 1360–
1380.
Highfield, T. (2016). Social media and everyday politics. Cambridge: Polity.
Hooghe, M, & Marien, S. (2013). A comparative analysis of the relation between political trust
and forms of political participation in Europe. European Societies, 15(1), 131–152.
Hooghe, M, & Quintelier, E. (2013). Political participation in Europe. In S. I. Keil & O. W.
Gabriel (Eds.), Society and democracy in Europe (pp. 220–243). Oxfordshire: Routledge.
Hughes, D. J., Rowe, M., Batey, M., & Lee, A. (2012). A tale of two sites: Twitter vs. Facebook
and the personality predictors of social media usage. Computers in Human Behavior, 28,
561–569.
Inglehart, R. (1977). The silent revolution: Changing values and political styles among Western
publics. Princeton, NJ: Princeton University Press.
Inglehart, R. (1990). Cultural shift in advanced industrial societies. Princeton, NJ: Princeton
University Press.
Inglehart, R., & Abramson, P. (1999). Measuring postmaterialism. American Political Science
Review, 93, 665–677.
Inglehart, R., & Welzel, C. (2005). Modernization, cultural change, and democracy: The
human development sequence. Cambridge: Cambridge University Press.
Inglehart, R., & Rabier, J.-R. (1979). Europe elects a parliament: Cognitive mobilization,
political mobilization and pro-European attitudes as influences on voter turnout.
Government and Opposition, 14(4), 479–507.
Kaase, M. (1999). Interpersonal trust, political trust and non‐institutionalised political
participation in Western Europe. West European Politics, 33(3), 1–21.
30
30 Beaufort & Seethaler: Citizen Participation, Social Media Use, and the Shift in Political Trust
Kaase, M., & Saris, W. (1997). The Eurobarometer – A tool for comparative survey research.
In W. Saris & M. Kaase (Eds.), Eurobarometer – Measurement instrument for opinion in
Europe (pp. 5–31). Mannheim: Zuma.
Kelly, S., & Cook, S. (2011). New technologies, innovative repression: Growing threats to
Internet freedom. In Freedom on the Net 2011: A global assessment of Internet and digital
media. Freedom House.
Kim, Y., Hsu, S-H., & Gil de Zúñiga, H. (2013). Influence of social media use on discussion
network heterogeneity and civic engagement: The moderating role of personality traits.
Journal of Communication, 63, 498–516.
Loader, B. D., & Mercea, D. M. (2012). Social media and democracy: Innovations in
participatory politics. London: Routledge.
Loader, B. D., Vromen, A., Xenos, M. A. (Eds.) (2014). The networked young citizen: Social
media, political participation and civic engagement. New York, NY: Routledge.
Martín, I., & Deth, J. W. van (2007). Political involvement. In J. W. van Deth, J. R. Montero &
A. Westholm (Eds.), Citizenship and involvement in European democracies: A comparative
analysis (pp. 303–333). London: Routledge.
Morales, L. (2009). Institutions, mobilisation and participation in Western democracies.
Colchester: ECPR Press.
Mossberger, K., Tolbert, C., & McNeal R. S. (2008). Digital citizenship: The internet, society,
and participation. Cambridge: MIT Press.
Norris, P. (2000). A virtuous circle: Political communications in postindustrial societies.
Cambridge: Cambridge University Press.
Norris, P., Walgrave, S., & van Aelst, P. (2005). Who demonstrates? Antistate rebels,
conventional participants, or everyone? Comparative Politics, 37(2), 189–206.
Olson, M. (1965). The logic of collective action: Public goods and the theory of groups.
Cambridge: Harvard University Press.
31
31 Beaufort & Seethaler: Citizen Participation, Social Media Use, and the Shift in Political Trust
Oppenhuis, E. (1995). Voting behavior in Europe: A comparative analysis of electoral
participation and party choice. Amsterdam: Het Spinhuis.
Pariser, E. (2011). The filter bubble: What the Internet is hiding from you. London: Viking.
Pasek, J., Kenski, K., Romer, D., & Hall-Jamieson, K. (2006). America’s youth and community
engagement: How use of mass media is related to civic activity and political awareness in
14- to 22-year-olds. Communication Research, 33, 115–135.
Patterson, T. (2002). The vanishing voter: Public involvement in an age of uncertainty. New
York: Knopf.
Perea, E. A., Jensen, M. J., & Jorba, L. (Eds.). (2012). Digital media and political engagement
worldwide: A comparative study. Cambridge: Cambridge University Press.
Putnam, R. D. (2000). Bowling alone: The collapse and revival of American community. New
York: Simon and Schuster.
Putnam, R. D. (2003). Making democracy work: Civic traditions in modern Italy. Princeton:
Princeton University Press.
Schaefer, A. (2013). Affluence, inequality and satisfaction with democracy. In O. W. Gabriel,
S. I. Keil & E. Kerrouche (Eds.), Political participation in France and Germany (pp. 139–
161). Colchester: ECPR Press.
Schudson, M. (1998). The good citizen: A history of American civic life. New York: The Free
Press.
Seethaler, J. (2013). Politics. In P. Simonson, J. Peck, R. T. Craig & J. P. Jackson (Eds.),
Handbook of communication history (ICA handbook series; pp. 302–314). London, New
York: Routledge.
Seidle, F. (2004). Expanding the federal democratic reform agenda. Policy Options, October
2004, 48–53.
32
32 Beaufort & Seethaler: Citizen Participation, Social Media Use, and the Shift in Political Trust
Shah, D. V., Cho, J., Eveland, W. P., & Kwak, N. (2005). Information and expression in a digital
age: Modeling Internet effects on civic participation. Communication Research, 32, 531–
565.
Shah, D. V., Cho, J., Nah, S. Gotlieb, M. R., Hwang, H., Lee, N. J., Scholl, R. M., & McLeod,
D. M. (2007). Campaign ads, online messaging, and participation: Extending the
communication and mediation model. Journal of Communication, 57, 676–703.
Shah, D. V., McLeod, J. M., & Yoon, S-H. (2001). Communication, context and community:
An exploration of print, broadcast, and internet influences. Communication Research, 28,
464–506.
Steiger, J. H. (1980). Tests for comparing elements of a correlation matrix. Psychological
Bulletin, 87, 245–251.
van Deth, J. W., & Elff, M. (2004). Politicisation, economic development and political interest
in Europe. European Journal of Political Research, 43(3), 447–508.
Verba, S., Schlozman, K. L., & Brady, H. E. (1995). Voice and equality: Civic volunteerism in
American politics. Cambridge: Harvard University Press.
Ward, J. (2014). Political consumerism as political participation? In K. Brants & K. Voltmer
(Eds.), Political communication in postmodern democracy: Challenging the primacy of
politics (pp. 167–182). Houndmills: Palgrave Macmillan.
Weinstein, E. (2014). The personal is political on social media: Online civic expression patterns
and pathways among civically engaged youth. International Journal of Communication, 8,
210–233.
Wright, J. D. (1981). Political disaffection. In S. L. Lang (ed.), The handbook of political
behavior, (vol. 4, p. 1–79). New York, NY: PLenum Press.
Xenos, M, & Moy, P. (2007). Direct and differential effects of the Internet on political and civic
engagement. Journal of Communication, 57, 704–718.
33
33 Beaufort & Seethaler: Citizen Participation, Social Media Use, and the Shift in Political Trust
Zani, B., & Barrett, M. (2012). Engaged citizens? Political participation and social engagement
among youth, women, minorities, and migrants. Human Affairs, 22, 273–282.
Zhang, W., & Chia, S. C. (2006). The effects of mass media use and social capital on civic and
political participation. Communication Studies, 57, 277–297.
Zittel, T., & Fuchs, D. (Eds.), Participatory democracy and political participation: Can
participatory engineering bring citizens back in? London: Routledge.
Zukin, C., Keeter, S., Andolina, M., Jenkins, K., & Delli Carpini M. X. (2006). A new
engagement? Political participation, civic life, and the changing American citizen. New
York: Oxford University Press.
34
34 Beaufort & Seethaler: Citizen Participation, Social Media Use, and the Shift in Political Trust
35 Table 1
Collective vs. Connective Action: Some Characteristics
Collective action
Connective action
Group based society
Ideologically framed issues
Society
Mass media as a communication agent
Seeking political relevant information
Media
Formal:
seperation of public & private spheres
Public
sphere
Organizationally brokered, mobilized and
managed
Participation
Age < 30 years
Age
Individualized society
Personal ideas about what constitutes
“good life”
Social media as a communcation
platform
Sharing of ideas through media-based
interactions
Informal:
blurred boundaries between public &
private
Self-motivating and largely selforganized
Age > 30 years
Note. Based on Bennett & Segerberg (2012)
Table 2
Typology of Forms of Political Participation
Latent political participation
Political/social
involvement
Manifest political participation
Civic engagement
Formal
participation
Informal activism
e.g. demonstrations,
strikes, protests, social
movements
through
brokered or
selfenabled by
established organizing
processes
organizations
basic
curiosity
about
matters of
institutional
politics;
sense of
belonging to
a collective
with a
distinct
political
profile
attention to
societal
matters in a
broader
sense; lifestyle
related
politics
(e.g. music,
food,
clothes)
e.g. donating
money, boycott of
firms, persuading
others about issues
of personal concern,
personalized ideas,
plans, and
grievances
e.g. voting;
membership in
political parties, trade
unions etc.
collective
connective
connective
collective
Note. Based on Ekman & Amnå (2012, p. 292)
35
collective
connective
Beaufort & Seethaler: Citizen Participation, Social Media Use, and the Shift in Political Trust
Table 3
Correlations Between Various Forms of Media Use and Participation
Type of media use
Civic
engagement
Political
involvement
.126***
(.145***)
.002
(.020*)
12.708***
.027***
(.078**)
.046***
(.051**)
-1.943
Mass media: general use
.057***
(.100***)
.110***
(.247***)
-5.442***
Mass media: intentional use
.102***
(.125***)
.246***
(.327***)
-15.035***
SNS: general use
SNS: intentional use
Correlations
difference
(z-score)
Note. Cell entries in columns 1 and 2 are two-tailed partial point-biserial correlations rpb (in parentheses: zeroorder point-biserial correlations)
*= p < .05, ** = p < .01, *** = p < .001
Table 4
Pairwise Correlation Differences Between Types of Media Use and Civic Engagement and
Political Involvement Respectively
SNS: intentional
use
Mass media:
general use
Mass media:
intentional use
Civic engagement
SNS: general use
10.961***
SNS: intentional use
6.300***
2.078*
-2.568**
-6.795***
Mass media: general use
-5.488***
Political involvement
SNS: general use
-4.854***
SNS: intentional use
-9.824***
-21.211***
-5.498***
-18.371***
Mass media: general use
-16.863***
Note. Cell entries are results of Steiger’s z-test for overlapping dependent correlations
*= p < .05, ** = p < .01, *** = p < .001
36
36 Beaufort & Seethaler: Citizen Participation, Social Media Use, and the Shift in Political Trust
Table 5
Predicting Civic Engagement and Political Involvement
Civic engagement
Model 1a
b
Countrya
Wald
Political involvement
Model 1b
Exp
(b)
b
152.737***
Wald
Model 2a
Exp
(b)
b
142.928***
-.130
22.107***
.879 -.110
19.639***
.896
Gender: female
vs. male
-.182
19.474***
.834 -.189
20.935***
.828 -.201
.122
25.029***
Financial
situation
medium vs.
bad
good vs. bad
SES
.111
5.315
.097
3.328
.124
4.79*5
.019
.606
Community type
middle sized
town vs.
village
1.130
24.076***
1.118
.268
.271
5.105
1.102
b
3.300
1.102 -.100
1.131
.121
4.538*
1.128
1.019
.017
.501
1.017
Wald
Exp
(b)
307.882***
73.048*** 1.308
.255
81.040***
1.291
.818
-.207
19.994***
.813
91.290*** 1.311
.251
91.085***
1.285
18.912***
19.839***
.097
22.894***
Model 2b
Exp
(b)
331.878***
Age
Level of
education
Wald
.905
-.100
2.953
.904
.159
6.544* 1.173
.141
5.095*
1.152
.074
7.478** 1.077
.065
6.031*
1.067
20.381***
2.945
17.044***
33.085***
30.660***
-.100
4.065*
.905 -.135
5.715*
.873
.140
6.472* 1.151
-.355
30.553***
.701
.149
7.012**
1.161 -.236
20.297***
.790
.367
32.915*** 1.444
-.216
12.960***
.806
Informedness
.248
117.079***
1.281
.230
101.933***
1.259
.537
406.524*** 1.710
.510
370.566***
1.665
Efficacy: yes vs.
no
.160
10.958**
1.174
.107
4.654*
1.113
.257
21.287*** 1.293
.167
8.533**
1.182
-.067
8.352**
.936 -.094
15.314***
.962
-.104
14.796***
.901
Life satisfaction
.094
14.187***
1.099
.085
12.005***
1.089
.058
4.352* 1.059
.043
2.447
1.044
Mass media:
Intentional use
.098
15.759***
1.103
.085
12.610***
1.088
.358
164.985*** 1.431
.333
152.860***
1.395
Mass media:
General use
.076
10.264***
1.079
.076
10.093***
1.079
.134
26.823*** 1.143
.134
26.279***
1.143
SNS: Intentional
use
.028
2.213
1.028
.012
.285
1.012 .0.42
3.824 1.043
.041
2.571
1.042
SNS: General
use
.138
32.122***
1.148
.136
28.952***
1.146 -.025
-.021
.539
.979
1.151
large town vs.
village
Democratic
satisfaction
.910 -.038
2.176
.800
.976
National trust
.041
2.878#
1.042
.141
25.141***
European trust
.061
6.066*
1.063
.111
15.877***
1.118
Trust in SNS
.087
15.390***
1.090
.047
3.597
1.048
SNS: General
use × Trust in
SNS
.034
Constant
Total R2
(Nagelkerke)
N
.004
.004
.996
.233
2.640#
1.034
10.060
1.262 .662
70.509
1.938
.094
.098
.266
.273
14,598
10,628
14,598
10,628
Note. Data: Eurobarometer 76.3. Logistic regressions (metric variables are z-transformed); Exp(b) = odds ratio
values above 1 indicate an increase, odds ratio values below 1 indicate a decrease in the dependent variable.
a
Variable (15 categories) is only used to control for country effects (see note 5).
# p < .10, * p < .05, ** p < .01, *** p < .001
37
37 38 Beaufort & Seethaler: Citizen Participation, Social Media Use, and the Shift in Political Trust
Figure 1
Civic Engagement as a Function of General Use of SNS and Trust in SNS
Civic engagement
0.6
Low
trust in
SNS
High
trust in
SNS
0.5
0.4
Low general use of SNS
High general use of SNS
Note. The figure is based on an Excel worksheet, provided by Jeremy Dawson, that plots two-way interaction
effects for a logistic regression analysis (www.jeremydawson.co.uk/slopes.htm).
38