Information, Communication & Society

ISSN: 1369-118X (Print) 1468-4462 (Online) Journal homepage: http://www.tandfonline.com/loi/rics20

The great equalizer? Patterns of social media use

and youth political engagement in three advanced
democracies

Michael Xenos, Ariadne Vromen & Brian D. Loader

To cite this article: Michael Xenos, Ariadne Vromen & Brian D. Loader (2014) The
great equalizer? Patterns of social media use and youth political engagement in three
advanced democracies, Information, Communication & Society, 17:2, 151-167, DOI:
10.1080/1369118X.2013.871318

To link to this article: http://dx.doi.org/10.1080/1369118X.2013.871318

Published online: 03 Jan 2014.

Submit your article to this journal

Article views: 4382

View related articles

View Crossmark data

Citing articles: 26 View citing articles

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=rics20

Download by: [University of Liverpool] Date: 30 September 2016, At: 02:35

Information, Communication & Society, 2014
Vol. 17, No. 2, 151–167, http://dx.doi.org/10.1080/1369118X.2013.871318

The great equalizer? Patterns of social media use and youth political
engagement in three advanced democracies
Michael Xenosa*, Ariadne Vromenb and Brian D Loaderc
a
Communication Arts, University of Wisconsin-Madison, 821 University Ave., Madison 53706, WI, USA;
b
Department of Government, University of Sydney, Sydney 2006, Australia; cDepartment of Sociology,
University of York, York, UK
(Received 19 November 2013; accepted 28 November 2013)

Recent developments suggest a strong relationship between social media use and political
engagement and raise questions about the potential for social media to help stem or even
reverse patterns of political inequality that have troubled scholars for years. In this paper,
we articulate a model of social media and political engagement among young people, and
test it using data from representative samples of young people in Australia, the USA,
and the UK. Our results suggest a strong, positive relationship between social media use
and political engagement among young people across all three countries, and provide
additional insights regarding the role played by social media use in the processes by which
young people become politically engaged. Notably, our results also provide reasons to be
optimistic concerning the overall influence of this popular new form of digital media on
longstanding patterns of political inequality.
Keywords: social media; political engagement; political participation; political socialization;
citizenship norms; political inequality

Introduction
From the events of the Arab Spring to the occupation of Zuccotti Park, stories of mass protests
saturated with the aura of young people’s savvy use of social media platforms have produced
headlines with growing regularity. It has also become conventional wisdom to attribute US Pre-
sident Barack Obama’s initial and later re-election victories to his campaign’s deft deployment of
social media to mobilize the youth vote. In the USA and other advanced democracies, social
media activity is disproportionately concentrated among young people (Australian Communi-
cations and Media Authority, 2013; Brenner, 2013; Woollaston, 2013). As a result, there has
been an explosive growth in studies examining relationships between social media use and pol-
itical engagement, sometimes with a specific focus on young people (Bode, 2012; Conroy,
Feezell, & Guerrero, 2012; Gil de Zúñiga, Jung, & Valenzuela, 2012; Vitak et al., 2011).
These developments contribute to a growing popular understanding of social media as a potent
tool for moving young people to political engagement.
Such developments are particularly significant for political communication research because
they raise questions about the potential for social media to help stem or reverse patterns of
political inequality that have troubled scholars for years (Brady, Verba, & Schlozman, 1995;

*Corresponding author. Email: [email protected]

© 2013 Taylor & Francis

152 M. Xenos et al.

Carlisle & Patton, 2013; Morris & Morris, 2013; Schlozman, Verba, & Brady, 2010). Underlying
these studies is a belief that by lowering the costs of many different forms of engagement and
providing new ways to discover and get involved with issues, social media may have great poten-
tial for not just mobilizing, but also broadening political participation. The optimistic tone of these
studies echoes earlier waves of enthusiasm surrounding the potential impact of general internet
use on political engagement and political equality. Indeed, after summarizing earlier scholarship,
the authors of a recent study explain that they ‘see social media as having the same influence but
taking it one step further to provide the individual greater flexibility to actively engage in the
public sphere’ (Carlisle & Patton, 2013, p. 3). Just as first-generation studies of internet use
and participation quickly seized on questions of whether that new technology would alter funda-
mental dynamics of political voice, contemporary political communication scholars have begun to
ask similar questions regarding social media.
Prospects for significant progress on these questions, however, are limited by a number of
factors. As we will explain in our review of the existing literature, these include a number of
empirical, conceptual, and theoretical issues. In this paper we seek to overcome limitations in
this literature by articulating a model of social media and political engagement among young
people that draws on contemporary scholarship concerning the processes by which contemporary
young people actively construct their roles as citizens. We test hypotheses derived from this
model using data from nationally representative surveys of young people (aged 16–29) in three
advanced democracies: Australia, the USA, and the UK. Our results suggest a strong, positive
relationship between social media use and political engagement across all three countries, and
suggest that social media may be helping to soften traditional patterns of political inequality.

Social media and political engagement

Social media and political engagement are both concepts that are subject to a variety of interpret-
ations. We consider ‘social media’ to include a variety of internet-based tools that users engage
with by maintaining an individual profile and interacting with others based on a network of con-
nections. In an effort to better synthesize general patterns of relationships between social media
use and political participation, we also adopt an expansive conception of ‘political engagement.’
The rapidly growing literature on social media use and political engagement has so far pro-
duced mixed results. On the optimistic side, a number of studies have suggested a positive
relationship between social media use and various indicators of political engagement (Bode,
2012; Valenzuela, Park, & Kee, 2009; Zhang, Seltzer, & Bichard, 2013). Other studies report
positive relationships between social media use and political engagement, but delimit those
relationships to specific kinds of social media use or a circumscribed set of engagement outcomes
(Gil de Zúñiga et al., 2012; Vitak et al., 2011). These studies suggest that the spread of social
media among young people and the broader public has had salutary effects on political engage-
ment, but at the same time suggest that the relationship between social media use and engagement
may be limited to individuals who would likely be relatively engaged without social media. Still
other studies suggest a very weak relationship between social media use and political engage-
ment, or none at all (Baumgargner & Morris, 2009; Dimitrova & Bystrom, 2013). Baumgartner
and Morris (2009), for instance, found small and inconsistent effects of social networking site use
on a range of engagement outcomes in their study of social networking site use surrounding the
2008 Iowa caucuses, ultimately concluding that ‘the hyperbole surrounding new Web develop-
ments … as they relate to citizenship may be just that – hype’ (p. 38).
Despite this mixed pattern of results, a few scholars have recently started to explore whether
the mobilizing potential of social media is powerful enough to directly affect classic patterns of
stratification by socioeconomic status and other factors in political engagement behaviors. Here
 Information, Communication & Society 153

too, however, results have not been consistent. On the one hand, the venerable scholars of stra-
tification and participation. Schlozman et al. (2010) ultimately found little evidence of ‘counter-
stratificational effects’ with respect to social networking site use and socioeconomic status in their
analysis of nationally representative Pew Internet and American Life Project survey data. They
did, however, find that social networking site use was associated with countertrends to traditional
disparities in participation based on age (Schlozman et al., 2010). In a similar study, however,
Morris and Morris (2013) found patterns implicating internet use in the closing of traditional
socioeconomic status gaps in participation, using a measure of internet use in which social
media figured quite prominently. Thus in the emerging literature on the implications of social
media use for patterns of stratification in political engagement, we find a much smaller, but simi-
larly mixed set of results.
A number of factors may explain the elusiveness of clear patterns of findings within the emer-
ging literature on social media and political engagement. First, as is understandable in early
exploratory research, many existing studies have been relatively limited in empirical scope.
Many rely on samples of college students, collected with varying degrees of sophistication
(Baumgartner & Morris, 2009; Bode, 2012; Valenzuela et al., 2009; Vitak et al., 2011). Others
have used nationally representative samples of adults of varying sizes (Gil de Zúñiga et al.,
2012; Zhang et al., 2013). Most are centered on a particular focusing event, such as the Iowa cau-
cuses (Baumgartner & Morris, 2009; Bode, 2012; Dimitrova & Bystrom, 2013; Zhang et al.,
2013). A consistent pattern within the sets findings reported in this literature based on these vari-
ations could point the way toward an orderly synthesis of the mixed picture of results reported so
far. Unfortunately, however, these methodological variations cut across the range of results found
within this growing literature. A similar set of observations can be made regarding variations in
conceptualization of the primary independent and dependent variables of social media use and
political engagement. As noted earlier, both of these concepts enjoy a multifaceted existence
within the broader literature. Once again, there appear to be no clear patterns based on these
kinds of differences among existing findings on social media and political engagement.
Indeed, while some studies suggest a direct relationship between the sheer time spent with
social media and political engagement (Bode, 2012; Zhang et al., 2013), others suggest that
such a relationship should only be expected with politically oriented activities in social media
(Gil de Zúñiga et al., 2012; Vitak et al., 2011). A few studies suggest a slight relationship
between the kinds of political engagement examined and the strength of the relationship with
social media use, such that mobilizing effects appear stronger for civic as opposed to political
engagement, and so-called ‘lightweight’ or online forms of involvement versus their more com-
plicated offline cousins (Baumgartner & Morris, 2009; Valenzuela et al., 2009; Vitak et al., 2011).
For each of these examples, however, there are studies in which such patterns are not readily
apparent (Bode, 2012; Zhang et al., 2013). Thus as with the methodological differences, it
appears that variations in the conceptualization of independent and dependent variables mainly
contribute to a general shortage of clear and definitive patterns with respect to the relationship
between social media use and political engagement.
In addition to these issues within the current literature on social media and political engage-
ment, we would further identify an important substantive limitation. Specifically, given the well-
known skew of social media use toward younger cohorts, and the veritable tradition of attending
to issues related to youth in research on digital media and politics, it is surprising that the connec-
tion between social media use and youth has rarely been explicitly theorized in research focused
on social media use and political engagement. To be sure, it is often mentioned that social media
use is concentrated among young people; but typically this is discussed in passing or, more often,
simply as a justification for using readily available samples of college students. This overlooks an
important opportunity to directly incorporate valuable insights about a core element of the
154 M. Xenos et al.

principal user base of social media into efforts aimed at understanding its implications for patterns
of political engagement. For example, a large amount of research on ‘digital natives’ and related
concepts documents the sense in which contemporary young people have a unique relationship
with digital media, particularly its newer features that are the heart of social media (Tapscott,
2008). Additionally, scholarship on youth and politics has shown that young people also have
a distinct set of political interests and interactions with politics (Levine, 2007). Moreover, scholar-
ship on political socialization and related concepts reminds us that many of the most avid users of
social media have recently undergone, or are in the process of undergoing, important experiences
that will powerfully shape their political engagement over time (Amna, Ekstrom, Kerr, & Stattin,
2009; Lee, Shah, & McLeod, 2012; McLeod & Shah, 2009; Sapiro, 2004; Shah, McLeod, & Lee,
2009; Torney-Purta, 2000). As we will discuss in more detail shortly, attention to these youth-
focused factors offers a useful set of concepts for understanding how social media use may be
initiating new patterns of political engagement in the digital age.

Theoretical model, hypotheses, and research questions

In constructing a theoretical model of social media use and political engagement among young
people, we draw from previous research on general internet use and political engagement. In con-
crete terms, we expect a mix of direct and differential relationships, with the latter conditioned by
predispositions and individual characteristics independently associated with political engagement
activities (Xenos & Moy, 2007). Along these lines, we expect that social media should be directly
associated with political engagement, but we also assume that the mobilizing power of social
media will not affect all young people in the same way. To elaborate on these expectations, we
first consider the mobilizing properties of social media, and then explore how key characteristics
and predispositions of social media users may interact with social media use to produce specific
patterns of differential relationships between social media use and political engagement.
Perhaps the most basic argument for the mobilizing potential of social media use follows the
sentiment expressed by Carlisle and Patton (2013) and cited in the introduction. By this logic,
social media possess the same quintessential mobilizing features of the internet as a whole, but
take them all ‘one step further’ by overcoming key limitations of the Web 1.0 world. Moreover,
as quintessential examples of Web 2.0 functionalities, social media have the added qualities of
making it much easier for individuals to engage in online versions of a number of traditional and
nontraditional acts of political involvement, such as participating in political discussions, persuad-
ing others how to vote, as well as engaging in a variety of forms of online activism. In addition, we
expect that the unique properties of social media could also directly stimulate political engagement
through other pathways. Perhaps the most discussed in existing research literature flows through
social media’s fostering of social capital and informal discussions about politics, which have
long been associated with higher levels of political engagement (Bode, 2012; Chong, Farquharson,
Choy, Lukman, & Mokhtar, 2011; Conroy et al., 2012; Ellison, Steinfeld, & Lampe, 2007; Gil de
Zuniga et al., 2012). Another possible pathway stems from the unique sense in which social media
depart from most other web experiences. Through habitual reviews of their ‘news feeds’, users are
often presented with information that they may not have been originally seeking, creating rare
opportunities for incidental exposure. This kind of dynamic has already been documented for like-
lihood of exposure to political content that runs counter to users’ existing political ideologies
(Messing & Westwood, 2012), and others have applied a similar logic to incidental exposure
across lines of general interest in politics (Morris & Morris, 2013). Individuals who may be rela-
tively uninterested or unengaged in politics need only ‘know’ a handful of interested and
engaged others in order to be regularly presented with incidental cues and information about pol-
itical issues that could result in greater engagement, especially during times of high political activity
 Information, Communication & Society 155

such as a national election. Based on these considerations, we offer our first hypothesis. H1: Social
media use will be positively related to political engagement.
Such considerations, however, focus mainly on attributes of social media that might be
equally experienced by any and all users. Young people are the heaviest users of social media,
and are also in the process of forming norms and habits of citizen engagement that are typically
stable across the life course (Amna et al., 2009; Sapiro, 2004). These factors are likely indepen-
dently related to political engagement, but are also of additional theoretical interest based on their
possible interactions with social media use in patterns of youth engagement. We now turn to two
such factors: political socialization and newly emerging norms of citizenship believed to be con-
centrated among young people in advanced post-industrial democracies.
Research on political socialization suggests that young people who come from homes in
which politics are regularly discussed among family members are significantly more likely to
engage in a variety of political behaviors (Andolina & Jenkins, 2003; Lee et al., 2012). In
addition, research has also highlighted the positive influence that school-based civic learning
experiences can have on later patterns of political engagement (Torney-Purta & Amadeo,
2003; Galston, 2004; Kahne & Westheimer, 2006). Socialization is also likely to have effects
on the relationship between social media and political engagement. Specifically, we posit that
young people who might be stimulated to political involvement through incidental exposure or
other political interactions facilitated by social media may be significantly more apt to express
that involvement through various political activities, if they have already developed civic compe-
tencies through family and school experiences. Based on these considerations we thus offer two
related hypotheses about social media use, political socialization and political engagement. H2a:
Political socialization experiences will be positively related to political engagement. And, H2b:
Political socialization experiences will significantly enhance the relationship between social
media use and political engagement.
An additional factor that is vitally important for understanding the relationship between social
media use and political engagement is the emergence of distinctly new norms of citizenship. Citing
a variety of broader social and economic currents commonly experienced within advanced democ-
racies, some have argued that contemporary youth are beginning to relate to politics and public life
in ways that are distinctly different from the dutiful conception of political involvement dominant
among previous generations (Bennett, 2012; Dalton, 2008). Work in this vein has suggested that
young people may be abandoning traditional modes of so-called ‘dutiful’ citizen participation
(voting, party membership, reading the newspaper), in favor of a more personalized politics of
self-actualization and expressive engagement with greater emphasis on non-traditional modes of
engagement such as digital networking, volunteering, and consumer activism (Loader, 2007;
Bennett, Wells, & Freelon, 2011). Though this ‘norm set for the next era’ (Bennett, 2012, p. 30)
involves both a turn away from more traditional forms of political engagement such as voting as
well as a turn toward newer forms of political action such as political consumerism, we believe
it is reasonable to treat ‘actualizing citizenship’ norms (Bennett, 2008) as a general predictor of pol-
itical engagement for contemporary youth, all else equal.
As with political socialization, we further expect actualizing citizenship norms to be impli-
cated in a set of contingent relationships involving social media use and political engagement
among young people. Because such norms reflect a distinct expression of contemporary interest
in politics among young people, we expect social media use to serve as a moderator for the
relationship between norms and political engagement. Here, we emphasize the affordances and
capabilities of social media for the particular kinds of political engagement scholars such as
Bennett (2008, 2012) suggest are a natural outgrowth of the adoption of actualizing citizenship
norms. We thus offer the following hypotheses concerning actualizing citizenship norms,
social media use, and political engagement. H3a: Actualizing norms of citizenship will be
156 M. Xenos et al.

Figure 1. Model of social media and political engagement (control variables omitted).

positively related to political engagement. And, H3b: Social media use will significantly enhance
the relationship between actualizing norms of citizenship and political engagement. Our general
theoretical model encompassing each of these predictions is summarized in Figure 1.
As noted earlier, our project of constructing a model explaining youth political engagement as
a function of social media use and other factors is in many ways but a means to an end. Specifi-
cally, we are interested in the extent to which relationships between social media use and youth
engagement may have implications for patterns of unequal political voice, primarily along the
lines of socioeconomic status, but also along other dimensions such as age and race or ethnicity.
There are at least two ways in which social media use could affect such patterns of inequality.
First, as discussed earlier, social media use could result in direct ‘counterstratificational’ effects
by raising political engagement among the previously uninvolved to levels that are much
closer to those exhibited by the more involved. Another possibility is that the direct influence
of widespread social media use on political engagement could serve to broaden the overall
pool of young people engaged in politics, contributing to a less rapid, but no less significant soft-
ening of political inequality patterns over time, through generational replacement. Given the rela-
tive dearth of clear patterns in this literature, and an absence of strong theoretical expectations on
this question, rather than formulating a clear hypothesis regarding patterns of political inequality,
we simply offer the following research question to guide our analysis. RQ1: To what extent does
social media use among young people affect patterns of political inequality?
As noted earlier, another impetus for our study is that existing research in this area includes
few large-scale empirical investigations. It is thus important to explore the extent to which
relationships between social media use and political engagement are stable or generalizeable
across a broad array of advanced democracies. Though many factors in our model are believed
to be consistent across advanced democracies, political socialization experiences and the insti-
tutional arrangements within which political engagement is practiced vary distinctly from
country to country (Hahn, 1998; Torney-Purta, 2000; Torney-Purta & Amadeo, 2003). Our
focus on these issues is expressed in our second research question: RQ2: To what extent do pat-
terns of social media use and youth political engagement vary based on differences in national and
political contexts among advanced democracies?

Data
To test our model and shed light on our research questions we analyze survey data collected in
three advanced democracies, Australia, the USA, and the UK. These countries represent an
 Information, Communication & Society 157

excellent set of cases for study in that they are all subject to the kinds of social and economic
trends identified earlier as particularly salient to understanding contemporary patterns of youth
engagement. They also share a common language and many cultural references, while offering
a reasonable amount of variation in civic education contexts, political institutions, and other
factors. As noted earlier, we chose to limit our survey populations to individuals aged 16–29,
which includes the range of ages typically associated with contemporary individuals’ experience
of ‘emerging adulthood’ (Arnett, 2000), as well as those age groups routinely identified as the
heaviest users of social media. We contracted with a marketing firm to administer our survey
to nationally representative samples of young people in each country. The surveys were identical,
with the exception of minor modifications of some items to reflect differences in language usage
or, in some cases, relevant options (e.g. educational levels). Participants were systematically
recruited from online panels to create samples that mirrored census data in each country on
key dimensions such as gender and age. Across all three countries, 3685 young people completed
our surveys, between late March and early May of 2013, including 1216 in Australia, 1228 in the
UK, and 1241 in the USA.
We constructed two distinct measures of our dependent variable, political engagement. The
first, individual engagement, was based on questions involving 12 individual acts of civic or pol-
itical engagement, modeled on items used by Zukin, Keeter, Andolina, Jenkins, & Carpini (2006).
Given the age range of our sample and compulsory voting in Australia, we did not ask about
voting. The otherwise inclusive list featured conventional political activities such as trying to
influence how others might vote in an election, as well as more civic-oriented acts like raising
money for charitable causes, and nontraditional political activities such as buying (or not
buying) goods or services based on political or ethical reasons. Participants were asked
whether they engaged in these activities in the past year, either online, offline, or in some com-
bination of online or offline forms (M = 4.10, SD = 3.88). Our second political engagement vari-
able, collective political engagement, was designed to capture activities that specifically involved
working with others in organizations. Again, we sought to capture a wide array of different kinds
of engagement, so we included items related to ‘political groups or causes’, ‘nonpolitical or chari-
table groups’, and groups associated with political candidates or parties. In all, we asked partici-
pants whether they joined, worked, or volunteered with five different kinds of groups, allowing
them to count activities that may or may not have involved the internet to varying degrees (M =
1.00, SD = 1.34).
To capture social media use, we created an index of social media use based on the frequency
with which participants used nine popular social media platforms. Respondents specified usage
on an eight-point scale ranging from ‘Never’ (0) to ‘Multiple times per day’ (7). We combined
responses to these items into a single scale of social media use (M = 1.83, SD = 1.12). Our
index of frequency of use across the 10 platform options enjoys reasonable validity (Chronbach’s
α = .76).
To represent the other major factors in our theoretical model, political socialization and actua-
lizing citizenship norms, we created four distinct independent variables. Three of these were
related to socialization. Political talk in the home was measured using responses to a series of
survey items that asked participants to indicate the frequency with which their early family experi-
ences included talking about news and political affairs using a five-point frequency scale ranging
from ‘Never’ (0) to ‘All the time’ (4). Across the full sample this variable has a mean of 1.64 and a
standard deviation of 0.89. We also created two measures of civic education experiences. Given
the significant differences in the ways that civic education is handled in each of the study
countries, our civic education items focused on general experiences involving teachers and pol-
itical material. The first of these was intended to tap ‘traditional’ or ‘general’ civic education
experiences and was based on three items that asked students to reflect on their secondary or
158 M. Xenos et al.

high-school experiences and express agreement or disagreement (on a standard five-point scale)
with statements like ‘When people had different opinions on political or social issues, teachers
encouraged us to discuss things’ (Chronbach’s α = .80, M = 3.41, SD = 0.91). We also created a
variable that captured digital literacy as part of civics education. This variable was based on
three items similar to those just described, but focused on the internet (e.g. ‘Teachers provided
instruction on how to assess the trustworthiness of information found on the internet’ and ‘Stu-
dents were required to use the internet to learn about politics or political issues’). This variable
allows us to distinguish between ordinary civic education experiences and those that might
reasonably be expected to be particularly relevant to social media and political engagement
(Chronbach’s α = .76, M = 2.99, SD = 0.97).
Finally, to capture actualizing norms of citizenship we used a series of items that asked par-
ticipants to indicate the extent to which various kinds of political activities were personally impor-
tant to them using a four-point scale ranging from ‘Not important at all’ (1) to ‘Extremely
important’ (4). Items used here were constructed based on the explanations of actualizing citizen-
ship norms elaborated in earlier theoretical work by Bennett (2008), which emphasize a personal
as opposed to dutiful orientation toward politics, as well as a relatively broad conception of poli-
tics and a ‘networked’ view of social relations. They included the following statements: ‘Volun-
teering your time or donating money to community organizations’, ‘Taking moral, ethical, or
political considerations into account when buying products or services’, and ‘Communicating
with others about social issues of personal concern’ (M = 2.45, SD = 0.76).
We also include a number of additional variables in our analysis, including a range of demo-
graphic variables and other control variables that were selected based on their demonstrated
relationships with political engagement in previous studies. In terms of demographics, we
include measures of age (M = 23, SD = 3.88), gender (female = 1, 52% across the three
samples), race (non-white = 1, 24% across the three samples), and a measure of parents’ edu-
cation. Parents’ education was selected as our best proxy for socioeconomic status, given that
our age range includes minor children, individuals at various stages of the normal course of edu-
cational attainment, and the difficulty of obtaining reliable estimates of household income from
the youngest members of our target populations. Since educational levels are different across
the three countries, we converted the raw measures of parental education to a zero-to-one scale
and then mean-centered these within each country, creating a measure of parents’ education
across the comparative data set that ranges from –0.47 to 0.48, and has a standard deviation of
0.20. Additional control variables included a measure of attention to political news, which was
based on standard four-point scales of attention paid to local, national, political, and international
news (Chronbach’s α = .80, M = 1.79, SD = 0.67), a two-item measure of internal political efficacy
(r = .92, M = 3.10, SD = 1.00), and a binary measure of whether participants identified with, or
considered themselves members of, an identifiable political party (61.4% across the full data set).

Methods
Our analytic approach involved specifying a series of hierarchical ordinary least squares
regression models, in which we entered blocks of variables based on their assumed causal
order. Blocks were entered in the following order: demographics, media attention, efficacy, and
partisanship, followed by a block including our socialization and norms variables (to test H2a
and H3a), a block for social media use (to test H1), and finally a block including interactions
between social media use and demographic variables, as well as interactions between social
media use and socialization experiences (to test H2b), and between social media use and actualiz-
ing citizenship norms (to test H3b). Whereas we report upon-entry coefficient estimates for the
first four blocks, we report before-entry coefficient estimates in the final block to limit the
 Information, Communication & Society 159

effects of multicollinearity on our evaluation of the interaction effects. The total adjusted R 2
reported in each column, however, is derived from a fully specified version of each model. To
simplify our analysis, we estimated models separately for each country. Cases were deleted
list-wise in the event of missing data. Finally, where relevant, we used the Clarify package in
STATA to create plots of significant interactions, reflecting expected values on the dependent vari-
ables (and 95% confidence intervals) for various combinations of relevant independent variables
(typically one standard deviation above or below each variable’s appropriate country mean).

Findings
The results from our regression analyses across each of the three countries in which we collected
data, and across our two indicators of political engagement, are reported in Tables 1 and 2.
Overall, the models perform quite well, producing F values well within acceptable significance
levels (in all cases p < .000), explaining roughly one-third of the variation in individual political

Table 1. Individual political engagement as explained by social media use, political socialization, and
actualizing norms of citizenship.
AUS USA UK
Demographics
Age .04 .00 .09**
Gender (female) −.05 −.13*** −.10**
Race (non-white) .17*** .04 .12
Parents’ education .20*** .13*** .15***
Incremental adjusted R 2 .08 .04 .05
Media use, efficacy, and partisanship
Attention to political news .08* .25*** −.04
Internal political efficacy .31*** .24*** .30***
Party identification/membership .12*** .04 .09
Incremental adjusted R 2 .23 .21 .18
Political socialization and norms
Political talk in home .08* .07*** .08*
Civic education (traditional) −.02 −.05 −.04
Civic education (digital) .13*** .05 .09*
Actualizing norms of citizenship .20*** .25*** .25***
Incremental adjusted R 2 .29 .27 .23
Social media use
Time spent with social media platforms .28*** .30*** .33***
Incremental adjusted R 2 .36 .34 .32
Interactions
Social media × Age .27 .17 −.03
Social media × Parents’ education .00 .04 −.03
Social media × Race (non-white) .09 .00 −.03
Social media × Political talk in home .05 .07 −.06
Social media × Civic education (traditional) .04 .15 −.02
Social media × Civic education (digital) .11 .16 .17
Social media × Actualizing norms .01 −.02 .01
Total adjusted R 2 .36 .34 .33
N 924 972 897
Note: Cell entries are standardized regression coefficients.
*p < .05.
**p < .01.
***p < .000.
160 M. Xenos et al.

Table 2. Group political engagement as explained by social media use, political socialization, and
actualizing norms of citizenship.

AUS USA UK
Demographics
Age −.04 −.10** −.05
Gender (female) −.01 −.10** −.05
Race (non-white) .09** .06# .13***
Parents’ education .16*** .13*** .18***
Incremental adjusted R 2 .04 .04 .06
Media use, efficacy, and partisanship
Attention to political news .14*** .25*** .09*
Internal political efficacy .17*** .10** .15***
Party identification/membership .07* .04 .11**
Incremental adjusted R 2 .11 .14 .12
Political socialization and norms
Political talk in home .07# .08* .06#
Civic education (traditional) −.06 −.01 −.03
Civic education (digital) .11** .07# .04
Actualizing norms of citizenship .19*** .22*** .22***
Incremental adjusted R 2 .15 .19 .16
Social media use
Time spent with social media platforms .21*** .21*** .20***
Incremental adjusted R 2 .19 .23 .20
Interactions
Social media × Age .27 .11 −.10
Social media × Parents’ education .16** .05 .03
Social media × Race (non-white) −.01 .14* .02
Social media × Political talk in home .22* .18* .24*
Social media × Civic education (traditional) .41** .22 .16
Social media × Civic education (digital) .50*** .27* .28*
Social media × Actualizing norms .16 .17 .50***
Total adjusted R 2 .21 .23 .21
N 924 972 897
Note: Cell entries are standardized regression coefficients.
#
p < .10.
*p < .05.
**p < .01.
***p < .000.

engagement, and explaining just over one-fifth of the variation in collective political engagement,
across the three countries.
The clearest pattern in our data is unambiguous support for H1 across all three countries and
for both indicators of political engagement. The results reported in Table 1 indicate that social
media use is significantly related to individual political engagement in Australia (β = .28,
p < .000), the USA (β = .30, p < .000), and the UK (β = .33, p < .000). Similarly, the results sum-
marized in Table 2 reflect the same pattern (Australia: β = .21, p < .000, USA: β = .21, p < .000,
UK: β = .20, p < .000). To provide a more real-world sense of these results, on the basis of the
unstandardized regression coefficients (not reported), movement of one standard deviation on
the social media use scale is generally associated with one additional act of individual partici-
pation, and nearly a third of an act of collective participation, all else equal. These estimates
hold equally across the advanced democracies included in our analysis. Considering that the
mean number of individual acts of participation is four, and the mean number of collective
acts is one, we interpret this as substantial demonstration of a direct relationship between
social media use and political engagement.
 Information, Communication & Society 161

Recall that H2a and H2b concerned the relationship between civic education experiences and
interactions between these experiences and social media use. Results here were somewhat mixed,
but ultimately fairly consistent with our expectations. Specifically, we found political talk in the
home to be significantly associated with individual political engagement across the three country
cases (Australia: β = .08, p < .05, USA: β = .07, p < .000, UK: β = .08, p < .05), and the corre-
sponding results for collective political engagement were marginally or statistically significant
(Australia: β = .07, p < .10, USA: β = .08, p < .05, UK: β = .06, p < .10). Traditional civic edu-
cation experiences were not significantly associated with either form of political engagement.
However, ‘digital’ civic education experiences, which mix digital media literacy with civic or pol-
itical discussion topics, were found to be significantly and positively related to individual political
engagement in Australia (β = .13, p < .000) and the UK (β = .09, p < .05), significantly and posi-
tively related to collective political engagement in Australia (β = .11, p < .01) and marginally and
positively related to collective political engagement in the USA (β = .07, p < .10). Taken as a
whole, we interpret these results as moderate support for H2a.
With respect to the conditioning of the relationship between social media use and political
engagement by political socialization (H2b), we found little support with respect to individual pol-
itical engagement, but considerable support when we turn to collective acts of political engage-
ment. Specifically, we find robust positive and significant results for the interactive term ‘Social
media × Political talk in home’ (Australia: β = .22, p < .05, USA: β = .18, p < .05, UK: β = .24,
p < .05). We also find a similar result for the traditional civics education and social media use
interaction in the Australian case (β = .41, p < .01). The strongest results here, however,
concern the interaction between ‘digital’ civic education experiences and social media use,
which again are positive and statistically significant across all three countries (Australia:
β = .50, p < .000, USA: β = .27, p < .05, UK: β = .28, p < .05). Figure 2 illustrates this interaction
within the Australian data.

Figure 2. Collective participation as a function of civics education and social media use (Australia).
162 M. Xenos et al.

Our remaining hypotheses, H3a and H3b, pursue a similar set of predictions with respect to
‘actualizing’ norms of citizenship. Overall, our results lend considerable support to the notion
that such norms are of critical importance to understanding contemporary youth engagement.
The clearest pattern of findings here is robust support for H3a. As seen in Table 1, the results
here are positive, significant, and strikingly consistent across the country cases for individual pol-
itical engagement (Australia: β = .20, p < .000, USA: β = .25, p < .000, UK: β = .25, p < .000).
A similar pattern of findings for collective political engagement is clearly visible in Table 2
(Australia: β = .19, p < .000, USA: β = .22, p < .000, UK: β = .22, p < .000). Results associated
with the interaction term between actualizing norms and social media use, however, lent little
support for H3b, except with respect to collective political engagement among British youth
(β = .50, p < .000). We plot this interaction in Figure 3.
In addition to providing a comprehensive series of tests of our hypotheses, our results also
shed considerable light on our two research questions. With respect to our primary research ques-
tion, we identified two plausible outcomes. The interaction terms between social media use and
demographic characteristics in each of the regression models provide direct leverage over the first,
that of counterstratificational effects. These interaction terms enable us to explore the possibility
that social media use may be directly associated with political inequality, either increasing or
decreasing stratification. As a careful examination of the coefficients for the interaction terms
reveals, in 16 out of 18 trials relevant to this question, we find no significant results. Aside
from the two significant results, the overall pattern presents no clear implication regarding a
direct relationship between social media use and political inequality.
The second plausible scenario related to RQ1 was that reasonably strong and direct relation-
ships between social media use and indicators of political engagement could signal the possibility
of a generational softening of stratification in political engagement. Taken as a whole, our results
are much more suggestive of this state of affairs. To be sure, consistent with the findings of
Schlozman et al.’s analysis (2010), our models reveal a persistent, positive, and significant

Figure 3. Collective political engagement as a function of actualizing norms and social media use (UK).
 Information, Communication & Society 163

relationship between socioeconomic status (SES) and political engagement whether the depen-
dent variable is individual (Australia: β = .20, p < .000, USA: β = .13, p < .000, UK: β = .15,
p < .000) or collective (Australia: β = .16, p < .000, USA: β = .13, p < .000, UK: β = .18,
p < .000) political activity. At the same time, however, a number of additional factors suggest
that social media use may be indirectly related to softening patterns of political inequality.
First, as noted earlier, we find a strong, significant, and robust positive relationship between
social media use and political engagement. Perhaps more telling, however, is the pattern of
explained variance across the various blocks in the six models. Though SES remains a significant
predictor of political engagement, for instance, it is worth noting that the entire block of demo-
graphic variables explains no more than roughly 30% of the total explained variance in any of
the models, and generally explains between 15% and 20% of the total explained variance
across most of the models. Additionally, in four of the six models, social media use alone explains
as much as or more total variance than the combined block of demographic variables. In two
models, those for individual political engagement in the USA and the UK, the social media
use variable explains nearly twice as much variance in the dependent variable as compared to
demographic factors. Stated plainly, our results suggest that if one were seeking an efficient
single indicator of political engagement among young people in the countries studied here,
social media use would appear to be as good as, or better than, SES.
In terms of our second research question concerning the extent to which our findings are con-
sistent across the three advanced democracies in which we collected our survey data, the fore-
going presentation of results has already revealed many of the relevant patterns. In addition to
indicating widespread support for H1, findings were also consistent across the three countries
in terms of support for H3a, concerning actualizing norms of citizenship, and the persistent
relationship between SES and political engagement just discussed. Though a number of variables
performed differently across the models, two patterns of country-based variations are worth
noting here. The first is civic education, particularly in ‘digital’ form, which overall appears to
have the most explanatory power in Australia. The second is actualizing norms, which as men-
tioned earlier appears to only exhibit the kind of contingent relationship with social media use
with respect to collective political engagement in the British case.

Discussion
In this paper we have advanced a theoretical model of social media use and political engagement, in
an effort to understand the possible implications of increasingly widespread use of internet services
with social media functionality for longstanding patterns of political inequality. Paying particular
attention to the fact that social media are largely dominated by younger users, we have focused on
theoretical concepts that are particularly relevant to contemporary youth in advanced democracies,
and we have deliberately narrowed our empirical focus to users between the ages of 16 and 29. In
search of broadly applicable findings and a parsimonious model, we have used particularly inclus-
ive measures of our primary variables of interest. Our results lend strong support to our hypothesis
that social media are positively related to political engagement, and suggest a number of patterns
consistent with a flattening out of social asymmetries in political engagement over time, via a
process of generational replacement. Though our principal findings are relatively consistent
across the advanced democracies in which we conducted our research, we also identified a
number of areas in which country-specific variations on these findings may deserve further inquiry.
Before discussing the implications of these findings for future research in the area of digital
media use and political engagement, it is important to first acknowledge and discuss some limit-
ations of the present study. First, as a cross-sectional design, the present study is inherently unable
to provide any definitive conclusions regarding causal relationships. Even without clear leverage
164 M. Xenos et al.

over causal questions, however, the present investigation still offers the most comprehensive
study of social media use and political engagement among contemporary youth to date.
Whereas previous work has focused on convenience or otherwise limited samples of college stu-
dents, or perhaps representative samples of a single country, the analyses presented here provide
an unprecedented level of reach in terms of examining patterns of social media use, political
engagement, and political equality across contemporary advanced democracies. Even with con-
siderable support and advances in research infrastructure, conducting research in this area still
involves a fundamental dilemma between an emphasis on internal versus external validity.
Given the current state of research in this area, we made a calculated choice to emphasize the
latter.
Additional limitations stem from issues related to the specific variables and measures included
in our study. For example, while being the best available proxy, our measure of socioeconomic
status is somewhat rough. However, we take our finding of a significant relationship between
this indicator and political engagement as an indication that it nonetheless provides a useful
measure of SES that behaves similarly to other measures used in the literature. Finally, it is undoubt-
edly a further limitation that our principal independent and dependent variables are more broadly
cast than most others used in the literature. Prior studies in this area have demonstrated the value of
attending to distinctions between various social media platforms (Pasek, more, & Romer, 2009),
types of social media use (Gil de Zúñiga et al., 2012), or forms of what we consider political engage-
ment (Dalton, 2008; Zukin et al., 2006). As useful as the insights that flow from such distinctions
may be, however, we believe there is also value in establishing a set of basic relationships between
general versions of the variables explored in our analysis. Moreover, in the case of opting for a
general (as opposed to a more politically focused) approach to social media use, such a tack pre-
serves an important separation between independent and dependent variables. In a world in
which most acts of political engagement can and are carried out within social media, a focus on
general use behaviors as independent variables may be the only way to avoid measuring the
same activity on both sides of the regression equation (Morris & Morris, 2013).
The aforementioned limitations of the present study notwithstanding, we believe the findings
reported here offer a number of significant implications and questions for future research. Most
important, the results of our inquiry offer a clear signal regarding the fundamental relationship
between social media use and political engagement that is not provided by the existing body
of research on these questions. On the basis of these findings, future studies can more confidently
explore causal relationships between key variables of interest. Additionally, it is hoped that our
results also contribute to research in this area by further demonstrating the importance of attend-
ing to variables particularly relevant to young people, such as political socialization and newly
emerging norms of citizenship, in all areas of research on digital media and citizenship. Differ-
ences in their experiences with socialization, and their norms and attitudes toward politics and
public affairs, are clearly relevant to understanding how young citizens interact with social
media and the political world. Future research in this area should explore these contingent
relationships in more detail.
A number of additional questions for future research are suggested by the country-specific
findings we report on the roles played by ‘digital’ civic education experiences and actualizing
norms of citizenship. Clearly, the results reported here suggest that there may be something par-
ticularly interesting happening with civics education in Australia. Indeed, readers familiar with
comparative studies of civics education should find our results particularly interesting given
that unlike the other two countries in our study, Australia lacks explicit and consistent civics cur-
ricula across all of its schools. At the same time, scholars interested in actualizing norms and
forms of political engagement and organizing may take particular interest in British politics
based on our findings.
 Information, Communication & Society 165

As a final reflection, we believe the present study provides an important and unprecedented
empirical validation of many claims regarding social media use and youth political engagement,
while pointing the way toward critical areas for future scholarship. We hope that future studies
will be able to build upon these findings to not only identify hyperbole and hype where appropriate,
but move us ever closer to a more comprehensive understanding of how new information technol-
ogies affect our youngest political actors, and thus the shape of political engagement overall.

Notes on contributors
Michael Xenos is an associate professor in the Department of Communication Arts at the University of
Wisconsin-Madison. [email: [email protected]]
Ariadne Vromen is an associate professor in the Department of Government and International Relations at
the University of Sydney. [email: [email protected]]
Brian Loader is an associate director of the Science and Technology Studies Unit (SATSU), and senior
lecturer in the Department of Sociology at the University of York. [email: [email protected]]

References
Amna, E., Ekstrom, M., Kerr, M., & Stattin, H. (2009). Political socialization and human agency. The
Development of civic engagement from adolescence to adulthood. Statsvetenskaplig Tidskrift, 111(1),
27–40.
Andolina, M., & Jenkins, K. (2003). Habits from home, lessons from school: Influences on youth civic
engagement. PS: Political Science & Politics, 36(2), 275–280. Retrieved from http://journals.
cambridge.org/production/action/cjoGetFulltext?fulltextid=147396
Arnett, J. J. (2000). Emerging adulthood: A theory of development from the late teens through the twenties.
American Psychologist, 55(5), 469–480. doi:10.1037//0003-066X.55.5.469
Australian Communications and Media Authority. (2013). Like, post, share: Young Australians’ experience
of social media. Retrieved from http://www.acma.gov.au/∼/media/mediacomms/Report/pdf/Like post
share Young Australians experience of social media Quantitative research report.pdf
Baumgartner, J. C., & Morris, J. S. (2009). MyFaceTube politics: Social networking web sites and political
engagement of young adults. Social Science Computer Review, 28(1), 24–44. doi:10.1177/
0894439309334325
Bennett, W. L. (2008). Changing citizenship in the digital age. In W. L. Bennett (Ed.), Civic life online:
Learning how digital media can engage youth (pp. 1–24). Cambridge, MA: MIT Press.
Bennett, W. L., Wells, C., & Freelon, D. (2011). Communicating civic engagement: Contrasting models
of citizenship in the youth web sphere. Journal of Communication, 61(5), 835–856. doi:10.1111/
j.1460-2466.2011.01588.x
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(1), 20–39.
doi:10.1177/0002716212451428
Bode, L. (2012). Facebooking it to the polls: A study in online social networking and political behavior.
Journal of Information Technology & Politics, 9(4), 352–369. doi:10.1080/19331681.2012.709045
Brady, H., Verba, S., & Schlozman, K. (1995). Beyond SES: A resource model of political participation.
American Political Science Review, 89(2), 271–294. Retrieved from http://www.jstor.org/stable/10.
2307/2082425
Brenner, J. (2013). Pew Internet: Social networking (full detail). Pew Internet and American Life Project.
Retrieved from http://pewinternet.org/Commentary/2012/March/Pew-Internet-Social-Networking-full-
detail.aspx
Carlisle, J. E., & Patton, R. C. (2013). Is social media changing how we understand political engagement?
An Analysis of Facebook and the 2008 Presidential Election. Political Research Quarterly. doi:10.1177/
1065912913482758
Chong, S., Farquharson, K., Choy, E. A., Lukman, Z. M., & Mokhtar, M. K. (2011). Enhancing youth civic
engagement and generalized trust through bonding social capital among friends. Pertanika Journal of
Social Sciences and Humanities, 19, 57–64.
166 M. Xenos et al.

Conroy, M., Feezell, J. T., & Guerrero, M. (2012). Facebook and political engagement: A study of online
political group membership and offline political engagement. Computers in Human Behavior, 28(5),
1535–1546. doi:10.1016/j.chb.2012.03.012
Dalton, R. J. (2008). The good citizen: How a younger generation is reshaping American politics (Rev. ed.,
p. 200). CQ Press. Retrieved from http://www.amazon.com/The-Good-Citizen-Generation-Reshaping/
dp/1604265566
Dimitrova, D. V., & Bystrom, D. (2013). The effects of social media on political participation and candidate
image evaluations in the 2012 Iowa Caucuses. American Behavioral Scientist. doi:10.1177/
0002764213489011
Ellison, N. B., Steinfield, C., & Lampe, C. (2007). The benefits of Facebook “friends”: Social capital and
college students’ use of online social network sites. Journal of Computer-Mediated Communication,
12(4), 1143–1168. doi:10.1111/j.1083-6101.2007.00367.x
Galston, W. (2004). Civic education and political participation. PS: Political Science & Politics, 37(2),
263–266.
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(3), 319–336. doi:10.1111/j.1083-6101.2012.01574.x
Hahn, C. (1998). Becoming political: Comparative perspectives on citizenship education. Albany, NY:
SUNY Press.
Kahne, J., & Westheimer, J. (2006). The limits of political efficacy: Educating citizens for a democratic
society. PS: Political Science & Politics, 39(2), 289–296.
Lee, N.-J., Shah, D. V., & McLeod, J. M. (2012). Processes of political socialization: A communication mediation
approach to youth civic engagement. Communication Research. doi:10.1177/0093650212436712
Levine, P. (2007). The future of democracy: Developing the next generation of American citizens (civil
society: historical and contemporary perspectives) (p. 308). Tufts. Retrieved from http://www.
amazon.com/The-Future-Democracy-Contemporary-Perspectives/dp/1584656484
Loader, B. (2007). Introduction: Young citizens in the digital age: Disaffected or displaced? In B. Loader
(Ed.), Young citizens in the digital age: Political engagement, young people and new media (pp. 1–
18). New York: Routledge.
McLeod, J. M., & Shah, D. V. (2009). Communication and political socialization: Challenges and opportu-
nities for research. Political Communication, 26(1), 1–10. doi:10.1080/10584600802686105
Messing, S., & Westwood, S. J. (2012). Selective exposure in the age of social media: Endorsements trump
partisan source affiliation when selecting news online. Communication Research. doi:10.1177/
0093650212466406
Morris, D. S., & Morris, J. S. (2013). Digital inequality and participation in the political process: Real or
imagined? Social Science Computer Review. doi:10.1177/0894439313489259
Pasek, J., more, E., & Romer, D. (2009). Realizing the social internet? Online social networking meets
offline civic engagement. Journal of Information Technology & Politics, 6(3–4), 197–215. doi:10.
1080/19331680902996403
Sapiro, V. (2004). Not your parents’ political socialization: Introduction for a new generation. Annual Review
of Political Science, 7(1), 1–23. doi:10.1146/annurev.polisci.7.012003.104840
Schlozman, K. L., Verba, S., & Brady, H. E. (2010). Weapon of the strong? Participatory inequality and the
Internet. Perspectives on Politics, 8(02), 487–509. Retrieved from http://journals.cambridge.org/
abstract_S1537592710001210
Shah, D. V., McLeod, J. M., & Lee, N. (2009). Communication competence as a foundation for civic com-
petence: Processes of socialization into citizenship. Political Communication, 26(1), 102–117. doi:10.
1080/10584600802710384
Tapscott, D. (2008). Grown up digital: How the net generation is changing your world (p. 384). McGraw-
Hill. Retrieved from http://www.amazon.com/Grown-Up-Digital-Generation-Changing/dp/0071508635
Torney-Purta, J. (2000). Comparative perspectives on political socialization and civic education.
Comparative Education Review, 44(1), 88–95. Retrieved from http://www.jstor.org/stable/10.2307/
1189228
Torney-Purta, J., & Amadeo, J. (2003). A cross-national analysis of political and civic involvement among
adolescents. PS: Political Science & Politics, 36(2), 269–274. Retrieved from http://journals.cambridge.
org/production/action/cjoGetFulltext?fulltextid=147394
Valenzuela, S., Park, N., & Kee, K. F. (2009). Is there social capital in a social network site?: Facebook use
and college students’ life satisfaction, trust, and participation. Journal of Computer-Mediated
Communication, 14(4), 875–901. doi:10.1111/j.1083-6101.2009.01474.x
 Information, Communication & Society 167

Vitak, J., Zube, P., Smock, A., Carr, C. T., Ellison, N., & Lampe, C. (2011). It’s complicated: Facebook users’
political participation in the 2008 election. Cyberpsychology, Behavior and Social Networking, 14(3),
107–114. doi:10.1089/cyber.2009.0226
Woollaston, V. (2013). Britons are the second most prolific Facebook and Twitter users in EUROPE with a
fifth now aged over 65. Daily Mail Online. Retrieved from http://www.dailymail.co.uk/sciencetech/
article-2340893/Britons-second-prolific-Facebook-Twitter-users-EUROPE-fifth-aged-65.html
Xenos, M., & Moy, P. (2007). Direct and differential effects of the internet on political and civic engagement.
Journal of Communication, 57(4), 704–718. doi:10.1111/j.1460-2466.2007.00364.x
Zhang, W., Seltzer, T., & Bichard, S. L. (2013). Two sides of the coin: Assessing the influence of social
network site use during the 2012 U.S. Presidential campaign. Social Science Computer Review.
doi:10.1177/0894439313489962
Zukin, C., Keeter, S., Andolina, M., Jenkins, K., & Carpini, M. X. D. (2006). A new engagement?: Political
participation, civic life, and the changing American citizen (p. 272). Oxford University Press. Retrieved
from http://www.amazon.com/New-Engagement-Political-Participation-Changing/dp/0195183177