Facebook and political engagement: A study of online political group membership
and offline political engagement
Meredith Conroy a
, Jessica T. Feezell b
, Mario Guerrero c,â‡‘
aOccidental College, Department of Politics, 1600 Campus Road, Los Angeles, CA 90041, USA
bUniversity of California, Santa Barbara, Department of Political Science, Mailcode 9420, Santa Barbara, CA 93106-9420, USA
c California State Polytechnic University, Pomona, Department of Political Science, 3801 West Temple Avenue, Building 94, Room 303, Pomona, CA 91768, USA
Available online 24 April 2012
In what ways do online groups help to foster political engagement among citizens? We employ a multimethod design incorporating content analysis of online political group pages and original survey research
of university undergraduates (n = 455) to assess the relationship between online political group membership and political engagementâ€”measured through political knowledge and political participation surrounding the 2008 election. We find that participation in online political groups is strongly correlated
with offline political participation, as a potential function of engaging members online. However, we fail
to confirm that there is a corresponding positive relationship between participation in online political
groups and political knowledge, likely due to low quality online group discussion.
2012 Elsevier Ltd. All rights reserved.
New media is a growing force in the study of civic engagement.
There are many levels of analysis within the discussion of new
media effects ranging from the global economy to personal use
of the Internet. Our research exists on the level of the democratic
divide (Norris, 2001), where researchers study individual-level
usage of the Internet and analyze its effect in terms of civic engagement. We join an active discussion of whether political Internet
use will be helpful, harmful, or irrelevant in its effects on civic society and political engagement.
There is some controversy concerning the effects of the Internet
on political engagement. While the impact of general Internet use
on political efficacy and trust is still contested,1 many are optimistic
about the ability of political Internet use to increase offline and conventional forms of political participation (Cho et al., 2009; Mossberger, Tolbert, & McNeal, 2008; Rojas & Puig-i-Abril, 2009; Shah, Cho,
Eveland, & Kwak, 2005; Xenos & Moy, 2007), knowledge (Xenos &
Moy, 2007) and civic engagement through social capital (Jennings
& Zeitner, 2003; Norris, 2001; Shah, Kwak, & Holbert, 2001; Valenzuela, Park, & Kee, 2009).
Understanding the influence of political Internet use, and especially new venues and capacities for social interaction, on offline
conventional forms of political participation and political knowledge is especially pertinent to understanding younger citizens,
who are more active online than previous generations. In 2007,
Pew reported that 93% of teens use the Internet. Additionally, as
Internet use goes up, participation on social networking sites
(SNS) increases as well: â€˜â€˜more [teens] than ever are treating [the
Internet] as a venue for social interactionâ€”a place where they
can share creations, tell stories, and interact with othersâ€™â€™ (Lenhart,
Madden, Macgill, & Smith, 2007). To better understand whether
heightened Internet use has a positive or negative impact on political engagement of youth, it is important for our analysis to incorporate measures of different types of social interactions online. As
time goes on, we are developing more robust measures for online
activities and effects through increased research efforts related to
the effects of new media. This paper is an early attempt to accurately capture measurements of these online social interactions.
The proliferation of online venues for all purposes, from social
interaction to consumerism, suggests that Internet use alone is
too blunt a measure. Recently, researchers have begun to examine
specific forms of â€˜â€˜political useâ€™â€™ of the Internet and SNS, an approach we find to be more indicative of the mechanisms through
which new media impacts political engagement. This project contributes to this line of more specified research by further exploring
how online political group membership affects offline conventional
forms of political participation and political knowledge among
youth. Political groups are defined as any social connection shared
by individuals, which can enable political discussion and interaction. Political groups have long existed offline through formal
group organizations and even informal interaction amongst
friends. However, new media is providing opportunities for
0747-5632/$ – see front matter 2012 Elsevier Ltd. All rights reserved.
â‡‘ Corresponding author. Tel.: +1 805 893 3432; fax: +1 805 893 3309.
E-mail addresses: [email protected] (M. Conroy), [email protected]
(J.T. Feezell), [email protected] (M. Guerrero). 1 For an extended discussion of how the Internet erodes engagement and
demobilizes citizens, see Ansolabehere and Iyengar, 1995; Nie, 2001; Pasek, More,
& Romer, 2009; Putnam, 2000. Also, more recent literature on Internet use and
slacktivism, see Vitak et al., 2011.
Computers in Human Behavior 28 (2012) 1535â€“1546
Contents lists available at SciVerse ScienceDirect
Computers in Human Behavior
journal homepage: www.elsevier.com/locate/comphumbeh
citizens in political groups to engage politically in ways that we
have not yet seen.
Focusing on the social networking website Facebook, we use a
multi-method design to learn more about the content of online
political groups and potential influence they have on political
engagement. Political engagement is defined here as offline conventional forms of political participation and political knowledge.
We begin with analysis of original survey data (n = 455) to measure
membership in online political groups and levels of offline conventional forms of political activity and political knowledge. We find
that increased online political group membership is correlated
with increased levels of offline conventional forms of political participation but not necessarily increased levels of political knowledge. To elaborate on these findings, we conduct a content
analysis of political group pages and group wall commentary
(walls are a shared social space where group members post messages), where we find information quality to be quite low and relatively opinionated rather than information rich. Through survey
design, we confidently establish correlation between online political groups and political engagement, while the content analysis
corroborates this relationship. We conclude with a discussion of
our findings and suggest direction for future research in this area.
2. Online political activity effects
Certain uses of the Internet and new media yield civically
redeeming effects in users. Mossberger et al. (2008) find that chat
rooms, political email correspondence, and online news exposure
predict higher voting rates. Shah, Kwak et al. (2001) demonstrate
that information exchange over the Internet fosters civic engagement, trust, and life contentment in younger generations, while social recreation on the Internet is negatively correlated with trust
and life contentment. Both of these studies highlight the more
deliberative uses of Internet, and more specifically, political discussion. McLeod, Scheufele and Moy (1999), Hardy and Scheufele
(2005), Shah et al. (2007), and Cho et al. (2009) find interpersonal
processes, such as discussion, are central to learning and action,
perhaps licensing the positive effects on civic engagement and
SNS often propagate deliberative activity through their use of
discussion walls, online chat, information sharing, and networking.
One function of SNS that has received little attention so far is the
ability to easily create and join groups. Social scientists have celebrated the advantages of group membership and associations for
decades and some have prescribed participation in groups as an
â€˜â€˜all-purposive elixir for the ills of societyâ€™â€™ (Dekker & Uslaner,
2001). Existing research demonstrates that group membership
encourages trust (Brehm & Rahn, 1997; Jennings & Stoker, 2004),
democratic values, and the development of important political
skills (Fowler, 1991; McFarland & Thomas, 2006). Furthermore,
membership in a group provides necessary motivation and incentive to be politically informed (Coleman, 1988; Fishkin, 1991). Indeed, described as a â€˜â€˜nation of joinersâ€™â€™ in the 18th century by
foreign visitor Alexis de Tocqueville (1990 , p. 118), political
engagement in the US has historically been spurred by group
In one of the more crucial calls for attention to groups, Putnam
(2000) details an alarming trend amid group membership and civic
engagement in the United States; as membership in civic groups
decreases so too does civic engagement. Putnam believes the stock
of social capital underpinning civic engagement is built up though
participation in voluntary organizations, largely offline. Yet the
Internet is changing the ways in which we communicate, organize,
and socialize (Bimber, Flanagin, & Stohl, 2005; Ellison, Steinfield, &
Lampe, 2007; Gil de Zuniga & Valenzuela, 2011; Hampton &
Wellman, 2001; Klein, 1999; Rich, 1999; Shah et al., 2005).
Technological development has spurred what is known as â€˜â€˜networked individualismâ€™â€™ where individuals are more likely to share
information and work in collaborative networked groups (Wellman, 2001). More specifically, the Internet revolution has brought
about the inception of online groups that appear to resemble offline
groups in function, if not in form. Even as some disagree that offline
groups have decreased in prominence, most agree that the Internet
has brought significant changes in how offline groups function.
The perceived decline in offline groups paired with growth
among online groups raises an important question for civic
engagement and new media: In what ways does online political
group participation benefit offline political participation? In this
paper, we anticipate advancing scholarship on the effects of online
political group membership specifically in terms of political engagement. Heeding advice from Berger (2009), we avoid measuring effects on civic engagement broadly. We focus more directly on
political engagement in the form of offline conventional forms of
political participation during the 2008 election and political
knowledge, generally, measured using a standard set of civics
questions (Delli Carpini & Keeter, 1997). We argue that online
political group membership is likely to encourage offline political
participation, but is unlikely to contribute to substantial increases
in political knowledge among joiners.
2.1. Group membership as a mechanism for political engagement
Group membership is thought to encourage political engagement though a number of mechanisms. First, group membership
can provide an opportunity for members to discuss politics. Discussion is thought to be integral to feelings of efficacy among citizens,
leading to higher rates of political activity (Andersen & Hansen,
2007; Cho et al., 2009; Delli Carpini, Cook, & Jacobs, 2004; Delli
Carpini & Keeter, 1997; Fishkin, 1991; Gastil & Dillard, 1999;
Robinson & Levy, 1986). Discussion in a group setting can also promote learning by necessitating the expression of views (Taber &
Lodge, 2006) and forcing more thoughtful consideration of
viewpoints (Eveland, 2004; Huckfeldt, 2007). Benhabib (1994, p.
30â€“31) notes that, â€˜â€˜when presenting their point of view and position to others, individuals must support them by articulating good
reasons in a public context to their co-deliberators. This process
of articulating good reasons in public forces the individual to think
of what would count as a good reason for all others involved.â€™â€™
Eveland (2004) finds that anticipation of discussion that is counter
to oneâ€™s own viewpoint motivates individuals to become more informed and elaborate on their own opinions. Reasoning, in this
general sense, promotes learning (Cho et al., 2009).
However, deliberation effects are precarious. Studies have
found the diversity of discussion to be imperative to knowledge
gains, whereas homogenous discussion or one-sided arguments
are detrimental to knowledge gains. This is especially evident in
the framing literature, which finds that the availability of counter-arguments limits framing effects (Druckman & Chong, 2007;
Druckman & Nelson, 2003; Sniderman & Theriault, 2003). Diverse
discussion is important in helping people to develop skills that
encourage deeper understanding, yet message exposure is
only as varied as a personâ€™s network (Gastil, Deess, & Weisler,
2002; Nisbet & Scheufele, 2004; Scheufele, 2002).
When most people discuss politics, â€˜â€˜their conversations usually
take place within primary groups of family and close friends â€“ that
is, among like-minded people who largely resemble each other socially and politicallyâ€™â€™ (Price & Capella, 2002, p. 304; see also Wyatt,
Katz, & Kim, 2000). Mutz and Martin (2001, p. 99) find cause for
concern as they show a trend toward ever-homogenizing discussant networks, however, they go onto note that our media environments, such as the news we read and watch, are more diverse than
our social environments and that when compared to personal
1536 M. Conroy et al. / Computers in Human Behavior 28 (2012) 1535â€“1546
interactions, people have less ability and desire to exercise selective exposure on media content (see also Brundidge & Rice,
2009). Additionally, some have taken the hopeful view that new,
albeit intangible, venues unrestrained by geography will enable
diversified discussion groups and a more engaged citizenry or that
online messaging will lead to more civic participation.
The Internet exists as a potential emporium of diverse information where people can communicate freely, without the restriction
of time and space. However, selection capabilities and little regulation of material available online might instead produce groups
with members who share similar values and ideas; in other words,
the Internet may lead to a heightening of selective exposure. By the
sheer choice lent to Internet users, those surfing the web can
choose to only join groups and discuss politics with like-minded
others (Bimber, 2008). Thus, although deliberation is a characteristic of groups both online and offline, deliberative criteriaâ€”that the
participant have an open mind, and that there be diversity among
messagesâ€”are not a necessarily organic characteristics of group
participation (Barabas, 2004).
Indeed, the Internet provides minority groups and subcultures
with views counter to the mainstream a place to meet and gather.
As Dahlgren (2006, p. 15) notes:
One can see an expansion in terms of available communicative
spaces for politics, as well as ideological breadth, compared to
the mass media. Structurally, this pluralization not only extends
but also disperses the relatively clustered public sphere of the
Yet by providing minority groups a digital space to meet, the
Internet becomes a place of external ideological balance, where
individual groupâ€™s space lack internal balance (Norris, 2001). Sparks
(2001) argues that this could lead to cynicism and disengagement.
Indeed, when individuals with strong political positions do meet to
discuss politics, it often dissolves into argumentation (Huckfeldt &
Mendez, 2007), defeating the purpose of deliberation.
Using survey design, Norris (2004) found that Americans most
active in online groups report that membership in online groups
both widened their social networks to include others with different
backgrounds and beliefs, as well as deepened their existing social
ties, though she notes that the type of group matters. Thus, the potential for groups to deepen existing cleavages exists.
Additionally, discussion that takes place online is asynchronous; although a number of social networking sites have a forum
or application for chatting and discussion, most dialogue takes
place on message boards and over the course of a few days or
months. For instance, a group member can post a comment on a
message board and either never returns to see if others have responded, or returns several days later to continue the conversation.
Either way, this type of discussion adds dimension to our traditional understanding of deliberation, acting as a potential detriment to the learning process.
Another valuable component of group membership that may or
may not be present online is accountability. Olson (1965) famously
argues that face time is an important component in the enforcement of member participation. By ensuring individuals will physically run into each other, small groups will enhance membership
participation. Correspondingly, larger groups are less likely to
incentivize member participation and individual accountability.
The Internet, alternatively, does not necessarily bring people into
physical contact or require as much commitment from its members and this has potentially harmful associated effects (Putnam,
If we apply this logic to group membership online, it suggests
that online groups may be unlikely to hold members accountable.
Members of online groups usually do not meet face to face, and
members of online groups can also choose to be anonymous, or
have multiple or false identities (Kolko & Reid, 1998; Nie & Erbring,
2000). For example, Fung (2002) found that journalists sometimes
pretend to be ordinary citizens online in Hong Kong. This can have
negative effects on oneâ€™s online community. As Kolko and Reid
(1998) suggest, individuals who are anonymous or have multiple
identities online are less likely to maintain their online identities,
and as a result, may be unlikely to contribute in a meaningful
way to an online community. These findings also suggest that elements of political engagement may be hindered as a result. Yet
Bargh, McKenna, and Fitzsimmons (2002) report that respondents
from their experiment felt more comfortable discussing politics
online than in person, and were more their â€˜â€˜true selvesâ€™â€™ (see also
Stromer-Galley, 2003). In conclusion, several attributes of offline
groups, such as face-to-face communication, physical group membership, and fluid personal discussion are not staples of online
groups, lending to the overall need of an increased study of this online function.
As the scholarship now stands, there are many competing as
well as consistent expectations of the influence of online group
activity on political engagement. Early work quickly called into
question the exchange of face-to-face interaction for online correspondence, but later work has begun to identify areas in which the
two forms are similar and even complementary (Krueger, 2002;
Norris & Jones, 1998; Shah, Kwak et al., 2001; Shah, McLeod, &
Yoon, 2001; Tidwell & Walthers, 2002; Wellman, Haase, Witte, &
Hampton, 2001; Williams, 2007). Recent scholarship has assessed
the impact of SNS network size and strength (Ellison, Steinfield,
& Lampe, 2011; Gil de Zuniga & Valenzuela, 2011; Rojas, 2008;
Son & Lin, 2008) on political engagement, but not necessarily smaller groups or cliques that exist within social networking sites. We
argue that as scholarship continues to move toward understanding
the effects of online activity on offline activity, greater attention
must be paid to the forums of exchange, the type of information
being exchanged, and the quality of opinions being expressed online. We move this line of research forward by opening up social
networking sites and looking at the specific ways in which groups
within these sites foster activities that may relate to offline conventional forms of political participation and knowledge. We focus
specifically on the social networking site, Facebook, which as of
late 2011, is the most visited website on the Internet. We chose
Facebook because in addition to its popularity as the most visited
site on the Internet, Facebook is the most popular social networking site. We do not expect that Facebook has a disproportionate
political use, but we assert that as it is the most popular social networking site, it potentially has substantial effects on the political
The main analysis examines whether membership in political
Facebook groups is independently related to both participation
and knowledge after controlling for other personal level indicators. These personal level indicators and participation will not
be the focus of the analysis as previous literature has documented
these relationships. Since the dependent variableâ€”political
engagementâ€”is likely to influence membership in online political
groups, the model may violate the recursivity assumption in OLS
regression. Thus, two-stage least squares (2SLS) models are used
to estimate this relationship, as OLS models cannot rigorously
control for any potential endogeneity between the dependent
and predictor variables. Before examining these results, we explain Facebook and the choice of using groups as our medium
of analysis. Next, we move onto describing the methodology
and limitations of the sample. In considering these limitations,
we assess the findings of the statistical analysis and use a content
analysis in an attempt to corroborate those findings. Through detailed analysis of the information exchanged among online
groups, online group membership rates, and measures of political
engagement and knowledge of group members on Facebook, we
M. Conroy et al. / Computers in Human Behavior 28 (2012) 1535â€“1546 1537
are able to empirically explore the theoretical expectations of online group participation.
Facebook is an online medium that lets users interact with each
other by sharing information about themselves via personal profiles. Users share their information by â€˜â€˜friendingâ€™â€™ others and
allowing them access to their profile. As of Winter 2011, Facebook
is currently considered the largest online social network with over
800 million active users, surpassing other online social networks
such as MySpace, Friendster, and Bebo. Originally created by Harvard University students in February 2004, Facebook was modeled
after paper pages that Harvard circulated profiling staff, faculty,
and students. Facebook originally began as a service only offered
to universities, but continually expanded its availability until Facebook allowed global registration in September 2006. Since then,
Facebook has grown rapidly, becoming especially popular among
younger generations and college students.
Although the premise of Facebook rests with sharing information via an online profile that contains basic information about
the user, there have been important additions to the site that have
fundamentally changed how users interact with others on Facebook. Facebook introduced the â€˜â€˜groupsâ€™â€™ application in September
2004 as one of its basic features. Groups allows users to share common interests with each other by providing a common space where
users can meet others interested in a specific topic, disseminate
information about that topic, and have public discussions relevant
to that topic. Groups also have the ability to connect individuals
who are not friends, yet share a common political interest. Individuals who would not be normally connected on Facebook are networked together in these groups based on their shared interests.
Groups are unique in the sense that they have a powerful networking ability. Groups also allow members to directly converse with
each other over one-on-one private messages, which signifies the
powerful potential of groups to facilitate political communication.
The group application was one of the earliest and still remains one
of the most pivotal features contributing to the interactive nature
of Facebook. Facebook has also made the wall (where users can
post messages on other peopleâ€™s profiles), notes (where users can
share their views with blog-like posts), share (where users can post
links to external websites on their profile), and fan pages (where
users can show support for a public figure) features, enabling users
to continually interact with each other in a variety of different
3. Research methodology
This is a study of a specific application of SNS, namely online
political group participation on Facebook. Of the work to date that
has focused on SNS (Ellison et al., 2007; Lewis, Kaufman, & Christakis, 2008; Zywica & Danowski, 2008) we were only able to find one
study that has looked at group participation through SNS and finds
a positive correlation between online political group participation
and offline political participation (Valenzuela et al., 2009). Our
study advances the field by looking at the relationship between online political group membership on Facebook and offline conventional forms of political engagement. We use a multi-method
design, which employs a survey and two-stage least squares model
to test our hypotheses, and also explore if a direct association exists. In the interest of adding depth to the findings, we conduct a
content analysis to investigate the manner in which online political
group membership discussion facilitates political engagement.
Very few large, national surveys include measures for the specific types of SNS usage that we would like to explore in our study.
Consequently, we designed a survey that allows us to open up general SNS usage and learn more about the specific ways in which
young people use these SNS applications as well as the political
and civic ramifications of their usage. We sample undergraduate
college students who are heavy users of Facebook, thus making
them an important sample population to examine for potential effects. Based on this survey we can begin to gain a better understanding of whether the political group application on Facebook
is civically virtuous.
Offline conventional forms of political participation and political knowledge are critical components of general political engagement; therefore, we examine these as our primary dependent
variables. Of the available Facebook applications, we specifically
expect that online political group participation will be positively
correlated with offline participation because online political
groups promote activity that transfers readily to the offline world.
H1. Increased levels of online political group membership are correlated with increased levels of offline political participation.
We also expect to find a direct significant association between
the relationship of online political group membership and offline
conventional forms of political participation and political knowledge. An individual who joins an online political group does not
necessarily have to be interested in politics. Rather, shared interests and social concerns may push an individual to interact and
learn about politics through online political groups. Online political
groups allow members to express their opinion through posts and
to engage on many levels through the group discussion and information sharing. For example, individual group members may learn
that an organization is sponsoring a local meeting, the location and
time of a political rally, a link to an online petition, or members
may generally feel obligated to take action after participating in
online political groups. These activities provide a psychological
connection with political activity online that we believe will stimulate political empowerment offline. An individual may be
empowered through interaction or information they found online,
that they might not otherwise have been exposed to. Although not
a specific hypothesis, we attempt to shine light on this association
by using a two-stage least squares regression model, which uses
instrumental variables to control for endogeneity. Separately, because of the nature of online political groups, we expect a different
outcome regarding the relationship between online political group
membership and political knowledge.
H2. Increased levels of online political group membership will have no
effect on levels of political knowledge.
In online political groups, we expect that the information
sources that are provided for members to access and share would
serve to increase levels of political knowledge only if these sources
were fully exploited and informed. Where, in face-to-face interactions people are physically held accountable for their statements
and conversation, we believe that the anonymity or lack of personal interaction online may lead to lower quality information
sharing and provision. Because it is very easy to post comments online, and because people can do so without much social risk, we expect to find a paucity of high quality information being shared on
group walls. Therefore we expect to accept the hypothesis of H2
and find no correlation between political group membership and
increased levels of political knowledge.
In summary, we expect to find that increased levels of online
political group membership via Facebook will be correlated with
increasing levels of political participation among members, offline,
yet have no effect on political knowledge. To analyze this, we conduct a two-part study. In the first part of the study, we use survey
design to assess the relationships between online political group
1538 M. Conroy et al. / Computers in Human Behavior 28 (2012) 1535â€“1546
membership via Facebook and political engagement. The second
part of the study focuses on a content analysis of political groups
to provide rich detail about how Facebook users interact in political groups and the quality of information and deliberation users
are exposed to.
4. Survey design
For the first portion of the study, we administer a survey to college undergraduates at a public university in California (n = 455).
This study is based on a convenience sample, which invariably
raises questions about the external validity of the findings (Sears,
1986). The survey allows us to gather cross-sectional data about
Facebook usage among a limited yet relevant population including
new measures for distinctly political versus non-political usage.
The survey takes roughly 15 min to complete and all surveys were
administered over the course of one week. We survey students in
three large political science classes; two lower division courses
and one upper division course during the first week of the spring
quarter of 2009. Nearly 70% of the students in the sample are, or
intend to become, political science majors. Political science majors
are admittedly more interested in politics. For the purposes of the
study, this biases the results because political science majors are
more likely to have knowledge about political issues and/or participate more in politics offline. Thus, in addition to the sample being
college-aged, the population of the study is biased towards being
politically active.2 The inclusion of political science majors in the
sample does offer some benefits, as young people are known for
their general lack of political participation.3 Twenty-two percent of
the sample was made up of freshman; 28% were sophomores; 33%
were juniors; 15% were seniors; the remaining 2% failed to specify
their year in school. However, the sample is not homogenous on
key demographic variables4 and is easily comparable to the wider
university population and similar state public universities. We do
not claim the findings on participation and knowledge to be generalizable to the wider population. It is reasonable to assume that this
sample might be hyper-users of Facebook, in addition to being younger, more politically interested and active than other cohorts. We do
believe, however, that the effects found in this study may not differ
drastically across similar groups, specifically younger generations of
politically active college students who use social networking sites in
large numbers. In considering that Facebook and online social networking is growing considerably over time, this sample is intrinsically interesting. The findings, as applied to hyper-users of
Facebook who are politically interested, provide interesting implications and new research questions for Internet use and civic engagement amongst the broader population.
We suspect that there may be some simultaneity between people who choose to join online political groups and people who engage politically. While endogeneity is a potential problem that we
anticipate, we run an ordinary least squares (OLS) regression as a
starting point for determining a significant direct association between online political group membership and offline political
participation and political knowledge. To establish this direct association, we address the problem of endogeneity by using a twostage least squares regression model (2SLS), which instruments
length of Facebook membership and frequency of log-on for political group membership on Facebook, to control for endogeneity.
Furthermore, we do recognize some limitation in the methodology
of this design as causality between online political groups and offline political participation is better established through a panel
Our independent variable is a self-identified measure of how
many political groups the respondent is a member of as a proportion of their total amount of group membership (PG), ranging on a
5-point ordinal scale from none to all. The two primary dependent
variables that are of most interest are offline political participation
(PP) and political knowledge (PK). To measure offline political participation, we create an aggregate scale composed of ten modes of
political participation coded on a four-point scale indicating participation frequency. The political participation scale ranges from 0 to
40 or low participation to high participation (M = 19.29, SD = 4.18,
a = .733).5 The measure for political knowledge is also an aggregate
scale composed of dummy variables for correct answers to 11 political knowledge questions. The political knowledge scale ranges from
0 to 11 or low knowledge to high knowledge (M = 9.47, SD = 1.64,
a = .621).6 The descriptive statistics for the predictor variable and
the two scales are listed in Table 1.
The model for political participation and political knowledge is:
Ã°PPÃž;Ã°PKÃž Â¼ a Ã¾ b1PG Ã¾ b2S Ã¾ b3A Ã¾ b4Y Ã¾ b5I Ã¾ b6ID
Ã¾ b7R Ã¾ b8PI Ã¾ b9ON Ã¾ b10P Ã¾ e Ã°1:1Ãž
Descriptive statistics for the dependent and predictor variables.
Variables Categories N Percentage
Political group membership None 108 25.12
A few 204 47.44
Some 89 20.70
Most 27 6.28
All 2 0.47
Participation index 0 5 1.43
1â€“2 37 10.57
3â€“4 158 45.14
5â€“6 106 30.29
7â€“8 37 10.57
9â€“10 7 1.90
Knowledge index 0 1 0.24
1â€“2 7 1.69
3â€“4 55 13.25
5â€“6 197 47.47
7 155 37.35
2 We controlled for major in an earlier iteration of the models presented in this
study. Major does not significantly affect the outcome. To address concerns of an
overly interested and knowledgeable population, we include controls for political
interest, which does affect the two dependent variables.
3 74.5% have persuaded one or more person to vote; 14% have donated to a
campaign; 3% have worked for a campaign; 17% have volunteered for a campaign;
34% have attended a political rally; 20% have put a political bumper sticker on their
car, or window; 11% have participated in a boycott; 56% have signed a petition. 34.1%
of the sample answered all political knowledge questions correctly; 29.0% missed one
question on the knowledge scale; 14.3% missed two questions on the knowledge
scale; 7.7% missed three questions on the knowledge scale; 14.9% missed four or more
questions on the knowledge scale.
4 The sample was compared on ethnicity and gender, which are the only relevant
demographic variables that are made publicly available by the university.
5 The participation scale includes measures of whether the subject voted in 2008,
plans to vote in the 2010 election, tried to persuade someone to vote, donated money
to a political candidate or campaign, worked as a paid employee for a candidate or
campaign, worked as a volunteer for a candidate or campaign, attended a political
rally, stuck a campaign sticker on window or car, participated in a boycott, and signed
6 The knowledge scale includes measures for whether the subject provided the
correct response to the following 11 questions provided by Delli Carpini and Keeter
(1997): which party holds the majority in the House of Representatives, vote required
to override a presidential veto, which party is more conservative, whose responsibility is it to determine if a law is constitutional, how many terms can the President
serve, how many members are on the Supreme Court, what political office is held by
Nancy Pelosi (write-in), can you vote online in a presidential election, do you need to
pass a literacy test to vote in CA, which 2008 presidential candidate most favored
universal health care, which 2008 presidential candidate most favored troop
reduction in Iraq.
M. Conroy et al. / Computers in Human Behavior 28 (2012) 1535â€“1546 1539
In the above equation, a number of measures are used to control
for socioeconomic, demographic, and political factors that are
thought to have an influence on participation and knowledge
(Rosenstone & Hansen, 1993; Verba, Schlozman, & Brady, 1995).
To control for the possible impact of sex on political participation
we include a dummy variable for sex (S), coded 0 for male and 1
for female. We include a scale for age (A),7 which is a self-reported
measure asking respondents how old they are, as well as an ordinal
measure for year in school (Y), coded low to high.
We ask subjects to report their family income (I) because this is
likely a better indicator of their socioeconomic status than the income of a student. We measure family income on an ordinal scale
ranging from â€˜â€˜under $50,000â€™â€™ to â€˜â€˜over $250,000â€™â€™ in $50,000 increments, with 6 representing over $250,000. We measure party identification (PI) using a 5-point scale moving from â€˜â€˜strong
Democrat,â€™â€™ â€˜â€˜weak Democrat,â€™â€™ â€˜â€˜Independent,â€™â€™ â€˜â€˜strong Republican,â€™â€™
and â€˜â€˜strong Republican.â€™â€™ Highly differentiated racial diversity
proved to be an insignificant factor correlated to participation
and knowledge, so we use a basic dummy variable (R) here where
â€˜â€˜white/non-Hispanicâ€™â€™ is coded as 1 and all else coded as 0. We
measure political interest (PI) using a 7-point scale indicating the
respondentâ€™s self-identified overall interest in politics.8 Lastly, recent work suggests that online news gatherers (ON) who use Internet
websites for political information, like Google and Yahoo News, are
more likely to vote (Mossberger et al., 2008), and we suspect that
privacy (P) online may also correspond with political reclusiveness,9
so we include a dummy variable for online news readers and a measure for online privacy that ranges on a 4-point scale from few
restrictions to many restrictions that a respondent places on their
Facebook profile. The descriptive statistics of the sample based on
these variables are listed in Table 2.
We first test the hypotheses that increased political group
membership on Facebook is correlated with increased offline political participation and political knowledge by using a multivariate
OLS regression. However, we also use a two-stage least squares
regression to control for any simultaneity between our primary
variables, to try and establish a direct association between the predictor and dependent variables. In the two-stage least squares
model on both participation and knowledge, we instrument political group membership (PG) using:
PG Â¼ a Ã¾ b1LM Ã¾ b2FL Ã¾ e Ã°1:2Ãž
We use two variables as instruments for political group membership that correlate with increased group membership but are
not politically motivated: how long respondents have been Facebook members (LM), calculated on a 5-point scale from â€˜â€˜less than
6 monthsâ€™â€™ to â€˜â€˜more than 3 yearsâ€™â€™ and how frequently respondents
log onto Facebook (FL) calculated on a 6-point scale from â€˜â€˜never or
almost neverâ€™â€™ to â€˜â€˜I always stay logged on.â€™â€™10 Thus, these variables
satisfy the requirement that instruments not be correlated with the
predictor variable for two-stage least squares regression models.
These two instruments are used to estimate our variable â€˜Predicted
probability of group membershipâ€™ for the 2SLS models in Tables 1
and 2. The second stage of the two-stage least squares model is similar to Eq. (1.1). As proper instruments in 2SLS, the instruments are
independently correlated with the independent variable but not
the dependent variable.11
4.2. Content analysis
In the second portion of the study, through in-depth content
analysis of political groups, we gain a better understanding of the
type of information and discourse to be found among these online
political groups. Survey design can only tell us so much about the
nature of online political group membership. The richness of information provided to us by looking directly at online political group
discussion gets to the central aspect of online social interaction. In
the survey design, we argue that political information levels may
remain unaffected by increased political group membership. In
assessing our hypothesis, the content analysis should underscore
why this seemingly contradictory effect occurs. Although some
limitations do exist in the content analysis, analyzing the content
and quality of information posted to these online political groups
can help us assess the merits of political knowledge acquisition
through this medium.
We analyzed the content of 39 randomly selected political Facebook group pages, accounting for numerous dimensions of information content and quality available through these online
political groups. Political groups are identified on Facebook as a
distinct category, called â€˜â€˜Politics.â€™â€™ The groups that the respondents
in the survey identified as political are presumably included in this
sample. We coded the information for these political group pages
(Cohenâ€™s Kappa = .71), gathering general group content and information including: number of news posts, links posted by the group
administrators, shared videos, advertised events, and group wall
discussion. Groups usually allow for members to post comments
on the â€˜â€˜wallâ€™â€™ to be viewed by members and non-members alike.
These comments on the group wall can potentially be seen as a
proxy for discussion that might occur in face-to-face interaction
in a traditional offline group. As we addressed earlier, wall posts
are a form of asynchronous communication, and thus not traditional discussion. Privacy standards on Facebook protect users
and their demographic information from being readily identified
through these discussions, thus the discussion that occurs may
be disjointed as a result. In addition, the discourse that occurs online may differ from the rhetorical and discursive practices that occur online (Bentivenga, 2006). These limitations prevent the results
of the content analysis from being generalizable. But given that
wall posts function as the primary mode of discussion for online
political groups on Facebook, the wall posts warrant a systematic
look. Discussion through these wall posts typifies the behaviors
that are common in most online social networks. Although the
7 Age is the first of three variables that pose potential problems in reference to the
assumptions of a regression. Age has a skewed distribution toward younger
individuals within the sample. This reflects the limitation of a sample of college
students. In addition, our variables measuring political interest and online news
gathers are also skewed, reflecting a sample that is more politically interested than
expected. While our study recognizes the limitations of the sample, we did conduct
the OLS and 2SLS by transforming these three variables. When we conduct the
analysis with these transformations, our results do not significantly change as the
magnitude, direction and significance of the relationships remain the same. We do
not report the regression analyses with the transformations as interpreting regression
results with transformations poses significant problems in comparing non-transformed to transformed variables.
8 The question on political interest is measured by using one question asking
respondents to report their own level of self-interest. The question used asked
respondents to indicate, on a seven-point scale, â€˜â€˜How interested are you in politics?â€™â€™
The scale was labeled at three points, â€˜â€˜Not interested,â€™â€™ â€˜â€˜Moderately interested,â€™â€™ and
â€˜â€˜Very interested.â€™â€™ Respondents are asked to interpret the 7-point scaled based on
their placement between these three labels within the seven points.
9 For an extended discussion on Facebook and privacy, see Lewis et al., 2008.
Privacy on Facebook can be predicted by the influence of cultural, social, and personal
10 Users who â€˜â€˜always stay logged onâ€™â€™ are presumed to be active users throughout
the day. They stay logged onto Facebook in order to intermittently check the service
whenever they are online.
11 Robust instruments are notoriously difficult to identify and employ. Further
testing through the Craggâ€“Donald Wald weak identification test reveals that these
estimators are categorically weak. However, we repeated the regression analysis with
the instrumental variables using a limited information maximum likelihood regression with a conditional likelihood ratio test that is robust to weak estimators and
conclude very similar findings to those found in the 2SLS regression (the results are
reported in Appendix C). â€˜Instrumented group membershipâ€™ is the equivalent of Eq.
1540 M. Conroy et al. / Computers in Human Behavior 28 (2012) 1535â€“1546
content analysis is limited, it can speak directly to the nature of online political discussion on networks. The results of the content
analysis have the ability to illuminate and generate future questions in the field of online social networks. We randomly selected
20 comments from each wall (n = 780) to be coded according to
opinion strength and overall information quality.12
In generating the codes for the analysis, we rely on a relatively
simple yet analytic strategy to assess the sophistication of the discussion. We gauge opinion strength based on the adjectives and
adverbs used in the comments, and whether or not the commenter
expresses a personal plea or statement. Opinions were coded as
â€˜â€˜Not Opinionated,â€™â€™ â€˜â€˜Low,â€™â€™ and â€˜â€˜High.â€™â€™ For example, the comment
â€˜â€˜Wait, Jeanine, your telling me Obama rode in on the good things
Bush Did??? Are you insane?â€™â€™ was left on the wall for the group
â€˜â€˜Is Obama Qualified? No, But He Did Stay at a Holiday Inn Express
Once.â€™â€™ This comment is categorized as highly opinionated based on
the strength of the language used. The strength of opinion from
these individuals with highly opinionated wall posts is typically
derived from existing knowledge of the political system. Those
posts that are not highly opinionated, and tend to be one-word,
two-word statements, or sentence fragments are devoid of any
information indicating outside knowledge of the topic under
The evaluation of wall post information quality was based on
the accuracy of the comment, and whether or not the comment
was supported by evidence, with an explanation or accurate statistics. Those comments which met this criteria were considered
â€˜â€˜Excellent;â€™â€™ comments that were accurate and supported with evidence, but did not elaborate on the evidence were considered
â€˜â€˜Good;â€™â€™ comments that were not supported with evidence were
considered â€˜â€˜Average;â€™â€™ inaccurate comments are considered
â€˜â€˜Poor.â€™â€™ For example, the comment â€˜â€˜According to former bank regulator Bill Black, more than 12 million Americans are at risk of
going into foreclosure because the mortgages they hold are worth
far more than the value of the homesâ€™â€™ left on the â€˜â€˜Occupy Bostonâ€™â€™
group would be considered to be an excellently informed post. The
post makes a comment based on statistics that would not be readily available on the group. In addition, we coded posts for the actual informational content they provided. Had the post provided
an informational perspective that was not yet introduced on the
group, it would be considered to be â€˜â€˜very informativeâ€™â€™. In the
example from the â€˜â€˜Occupy Bostonâ€™â€™ group, that post would be categorized as â€˜â€˜very informativeâ€™â€™ as the statistic had yet to be
brought to light on the group wall prior to the posting. When information is introduced on the groups in this manner, it has the ability to facilitate and encourage learning amongst most discussants
in the group.
The coding procedure followed standards for content analysis,
including the development of a codebook, coder training, coder
practice, and establishing intercoder reliability. In the next section,
we present the results from the content analysis and discuss how
this method contributes to a deeper understanding of general online discussion.
Table 3 presents the findings from the OLS and 2SLS models
with political participation as the dependent variable (H1).13 The
political participation scale is coded so that higher scores are associated with higher probabilities of participating offline. This table indicates that our primary independent variable, political group
membership through Facebook, has a significant effect on offline
political participation after controlling for other influential factors.
In fact, in the OLS model, a respondent indicating â€˜â€˜allâ€™â€™ versus â€˜â€˜noneâ€™â€™
of their Facebook groups as political is likely to score higher on our
participation scale by seven points, or the equivalent to performing
two political activities with the highest frequency. In controlling
for endogeneity in the two-stage least squares model, we call the
dependent variable â€˜predicted probability of group membershipâ€™.
The two-stage least squares model uses Eq. (1.2) as instruments in
order to avoid the problem of endogeneity. We also confirm the relationship between political group membership on Facebook and offline political participation.
Table 4 examines political knowledge as the dependent variable
(H2), where we observe that online political group membership
through Facebook is significantly but weakly correlated to political
knowledge in the OLS model. The difference between a respondent
with no political group memberships and one with all political
group memberships is the equivalent of answering slightly more
than one additional political knowledge question correctly. However, when using the two-stage least squares model, the predicted
probability of group membership fails to achieve statistical significance. The instrument used in Table 4 is identical to that of Table 3.
â€˜Predicted probability of group membershipâ€™ is instrumented for
using Eq. (1.2). Unlike the results for Table 3, the two-stage least
Descriptive statistics for the sample.
Variables Categories N Percentage
Sex (S) Male 212 47.11
Female 238 52.89
Age (A) 18â€“19 183 40.76
20â€“21 211 46.99
22â€“23 41 9.13
24 and above 14 3.12
Family income (I) Less than $50,000 72 18.51
$50,001â€“100,000 108 27.76
$100,001â€“150,000 82 21.08
$150,001 and above 127 32.65
Party identification (ID) Strong Republican 26 6.36
Weak Republican 50 12.22
Independent 62 15.16
Weak Democrat 153 37.41
Strong Democrat 118 28.85
Year in school (Y) Freshman 102 22.71
Sophomore 129 28.73
Junior 152 33.85
Senior 66 14.71
Political interest (PI) Not interested 40 8.88
Moderately interested 38 8.44
Very interested 372 82.68
White (R) White 244 57.41
Non-white 181 42.59
Online news user (ON) 0 h/week 144 39.89
1â€“3 h/week 45 12.47
4â€“6 h/week 57 15.79
7 h/week and above 115 31.85
Privacy restrictions (P) None 340 79.07
Some level of privacy 90 20.93
Length of FB membership (LM) Less than 6 months 16 3.72
6 monthsâ€“1 year 52 12.09
1â€“2 years 116 26.98
2â€“3 years 159 36.98
3 years or above 87 20.23
Frequency log-in (FL) Never 14 3.50
1â€“2 times/week 51 12.72
Several times/day 305 76.05
Always logged on 31 7.73
12 We used a random number generator to produce 20 numbers ranging from 1 to n.
We then scrolled through the wall pages to find the corresponding wall post and
pasted it onto a sheet for our research assistants to code.
13 The OLS models for both political participation and political knowledge are
reported in Appendixes A and B, respectively.
M. Conroy et al. / Computers in Human Behavior 28 (2012) 1535â€“1546 1541
squares model for political knowledge in Table 4 suggests that the
relationship observed in the OLS model is endogenous.
As such, it seems that we can make the case that membership in
online political groups via the Facebook platform encourages offline
political participation, even when the simultaneity problem is taken into account. At the very least, we can be confident that online
political groups encourage offline political participation and therefore we confirm H1. When we turn to political knowledge, however,
H2 is supported in the 2SLS model which controls for simultaneity
showing that political group membership does not confidently bolster levels of political knowledge. Therefore, while political engagement encompasses both political participation and political
knowledge, our study cannot confirm the fact that Facebook creates
fully politically engaged participants, rather it seems that it encourages political participation but not a corresponding effect on political knowledge. To understand more about why this might be the
case, we turn to the content analysis of the group pages.
Table 5 shows that the content of Facebook political pages provides potential sources of information, particularly in self-guided
OLS and Two-stage least squares estimates of political group membership on Facebook (PG) and offline political participation (PP).
Variables b (se) p > |z| b (se) p > |z|
Political group membership on FB (PG) 1.785 (.315) .000
Predicted probability of group membershipa 2.742 (1.334) .041
Female (S) .103 (.475) .830 .269 (.537) .617
Age (A) .231 (.299) .442 .297 (.320) .354
Family income (I) .280 (.150) .063 .240 (.163) .143
Party identification (ID) .441 (.188) .020 .351 (.227) .124
Year in school (Y) .096 (.393) .807 .057 (.406) .888
Political interest (PI) .923 (.206) .000 .667 (.404) .100
White (R) .636 (.522) .225 .495 (.568) .385
Online news user (ON) .038 (.024) .116 .030 (.027) .274
Privacy restrictions (P) .538 (.298) .073 .482 (.315) .128
R2 .411 .382
Adjusted R2 .380 .349
F 13.12 9.86
N 199 199
Note: Data derived from survey of 455 college undergraduates. Unstandardized regression coefficients with standard error in parentheses. All tests are two-tailed tests. To
control for possible interdependence between group membership and political participation offline, we estimated a two-stage least squares model.
a Predicted probabilities from first-stage OLS regression where the dependent variable is political group membership, and independent variables are female, age, family
income, party identification, year in school political interest, White, Asian, Black, Hispanic, Online news user, and privacy restrictions. Years on Facebook and frequency of
Facebook log-in are the instrumental variables.
OLS and two-stage least squares estimates of political group membership on Facebook (PG) and political knowledge (PK).
Variables b (se) p > |z| b (se) p > |z|
Political group membership on FB (PG) .295 (.140) .036
Predicted probability of group membershipa 1.045 (783) .184
Female (S) .590 (.225) .009 .695 (.265) .009
Age (A) .286 (.138) .040 .392 (.185) .035
Family income (I) .099 (.070) .158 .114 (.077) .139
Party identification (ID) .008 (.087) .931 .063 (.118) .596
Year in school (Y) .340 (.178) .058 .375 (.195) .056
Political interest (PI) .336 (.098) .001 .127 (.239) .596
White (R) .498 (.242) .041 .352 (.301) .244
Online news user (ON) .007 (.011) .568 .001 (.014) .963
Privacy restrictions (P) .015 (.138) .916 .112 (.179) .532
R2 .243 .123
Adjusted R2 .201 .074
F 5.79 4.79
N 191 191
Note: Data derived from survey of 455 college undergraduates. Unstandardized regression coefficients with standard error in parentheses. All tests are two-tailed tests. To
control for possible interdependence between group membership and political knowledge, we estimated a two-stage least squares model.
a Predicted probabilities from first-stage OLS regression where the dependent variable is political group membership, and independent variables are female, age, family
income, party identification, year in school political interest, White, Asian, Black, Hispanic, Online news user, and privacy restrictions. Years on Facebook and frequency of
Facebook log-in are the instrumental variables.
Political Facebook group page content.
Percentage of groups
No (%) Yes (%)
Provide additional contact information 39 62
Provide web links in the â€˜â€˜linksâ€™â€™ section 18 82
Provide video links 41 59
Event information posted 80 20
News posts provided 26 74
Photos posted 13 87
Discussion topics posted on â€˜â€˜discussion boardâ€™â€™ 13 87
Note: Data from content analysis of 39 randomly selected political Facebook groups.
1542 M. Conroy et al. / Computers in Human Behavior 28 (2012) 1535â€“1546
formats, which members can optionally take advantage of. For
example, 62% of the pages we coded provided additional contact
information for the group outside of Facebook and 82% posted
additional website links in the designated â€˜â€˜linksâ€™â€™ space. In addition, a large number of the groups posted news links, photos, and
discussion topics for the visitors to engage in online. Administratively, these political pages seem to provide a wealth of information where members have the opportunity to seek and learn
more about their group. While it seems that political groups have
the ability to encourage the acquisition of political knowledge, performing a content analysis of the discussion among group members within these groups can highlight whether or not members
take advantage of these opportunities.
Fig. 1 presents our findings across three critical dimensions
among the 780 wall posts randomly selected for analysis. We presume that members who are politically knowledgeable should be
having online discussions that are well informed, characterized by
strong opinions, and of high-information quality. Even if all members are not necessarily politically knowledgeable, if the discussion
between members is reasonably well-informed, then the discussion
should encourage the attainment of political knowledge.
Overall the informational content and quality of discussion on
the walls was very low. Forty-one percent of the wall posts were
â€˜â€˜not veryâ€™â€™ informative posts, which are described as posts that
did not share any new information. Only 16% were classified as
sharing â€˜â€˜veryâ€™â€™ thoughtful information within the post; these posts
are coded as those that offered a new perspective or information to
the group. The strength of the opinion offered in the post was
coded not opinionated or neutral, low opinion, or high opinion.
Low opinion posts are those that have a perspective on the issue
at hand, as contrasted to high opinion posts which have a strong
perspective by inciting people or advocating for action. Overwhelmingly, 523 of the 780 (67%) posts offered low or high opinion
strengths suggesting that the general discussion in Facebook
groups is opinionated. The wall posts were also coded for their
overall information quality, ranging from â€˜â€˜Poorâ€™â€™ where the information was inaccurate, incoherent, or did not support thought
with evidence, to â€˜â€˜Excellentâ€™â€™ where the post supports their
thought with evidence and/or thoughtful explanation. The overall
quality of the posts we coded was poor and only 4% were thought
to offer excellent quality discussion.
The finding that the overall quality of wall posts on these
groups is low may be on account of the nature of most Facebook
groups themselves. Online political groups have the potential to
foster discussion, but are also mediums for other types of activity.
For example, members create political groups without the intention of being an information-rich space. Indeed, groups such as â€˜â€˜I
have more foreign policy experience than Sarah Palinâ€™â€™ are formed
to be a source of comedic material, as opposed to a place to talk
seriously about Sarah Palinâ€™s foreign policy experience. In our content analysis we do not organize groups by their intention or â€˜â€˜seriousnessâ€™â€™ of discussion. In this manner, differences in overall
quality may be linked to the groupâ€™s original purpose. Nonetheless,
regardless of each groupâ€™s intention, these online political groups
create awareness around political issues. We do recognize the limitation around the existing content analysis and suggest future
work and description is necessary to fully understand the discussion taking place in online political groups on Facebook.
Our content analysis indicates that political Facebook group
users, in general, often do not share much new information and
the information they do share tends to be somewhat inaccurate,
incoherent, or not very well supported with evidence. As a forum
for people to easily engage and share their opinions, online political groups are beneficial; however, as a forum to learn new political information online political groups are ineffective due in part to
low quality wall discussion. The survey design suggests that while
online group membership in political groups encouraged political
participation, it had no effect on political knowledge. The content
analysis suggests that while political group members may feel
more empowered or efficacious through the opportunity presented
by online participation in political groups, the low quality of group
interaction does not encourage members to learn new political
Fig. 1. Content analysis of political group wall posts.
M. Conroy et al. / Computers in Human Behavior 28 (2012) 1535â€“1546 1543
This research and its findings are significant on three important levels. First, we illustrate the need to start looking deeper
into SNS usage, and political Internet usage more generally. Social
network sites are not a use in and of themselves, as much as they
are a platform for various applications that have important
implications for studying how people interact today. This study
highlights one SNS application from one specific SNS at one point
in time. We investigate a specific service provided by Facebook
through which groups can encourage political participation in
ways very similar to offline groups. While we show that Facebook
groups foster political participation, we believe this is just a
preliminary step, to understanding the potential of SNS to impact
the political process. Indeed, Facebook and other SNS have
created new ways to bridge the gap between users through
groundbreaking interactive technologies. These groundbreaking
technologies provide ample opportunity for other scholars to
investigate Facebook, SNS sites, and their effect on political
Second, we show that online political groups produce similar effects to traditional offline groups, specifically in their ability to foster political engagement. The 2008 election solidified the
importance of the Internet broadly, and SNS specifically, as critical
elements of politics and campaigning today. We find that Facebook
allows for the creation of online political groups that provide many
of the benefits that we have known face-to-face groups to provide
for decades, such as information, motivation for political action,
and a forum for discussion and communicative exchanges. In this
sense, Facebook is fostering political engagement. However, this
finding points to other interesting questions regarding the nature
of online political groups. While we know online political groups
have the potential to encourage offline political engagement, future research should focus on how and whether online political
groups influence feelings of efficacy, and how membership in
groups changes or solidifies political opinions and attitudes. The
full extent of the effect of online political groups on political behavior needs to be thoroughly examined as the effects of online political groups can have potentially important ramifications for the
The last point we make emphasizes the dual nature of the
findings. The fundamentals of democracy assume a knowledgeable public, one that is capable of representing its own self-interest effectively. A healthy democracy, then, should see tandem
movement between political knowledge and political participation. Here we find that while online political group membership
is correlated with offline political participation, we do not see
an equally significant correlation with levels of political knowledge. Although a panel study would be a more ideal method for
shining more light on the potential for information gains, we executed a content analysis of group wall posts as an attempt to
understand these findings. Our results offer a suggestion for
why this might be the caseâ€™ the information content and quality
of most wall posts were found to be very poor, generally lacking
support for their claims, incoherent, or simply opinionated. In
other words, political group members are exposed to little new
or well-articulated information about the political causes around
which these groups form. The information is more likely to be
reinforcing and therefore mobilizing, but not enlightening and
Through content analysis of online political group pages
coupled with a survey of high-level Facebook users, we offer a step
forward in understanding the political nature and effects of online
social networking sites and online political groups. We find that
online political groups that are facilitated through SNS platforms
such as Facebook perform many similar functions to their offline
counterparts. Online political group membership is positively related to offline political participation, but appear to fall short on
our measures of political knowledge. We suggest this is the case
because while the groups offer many applications that members
can use to feel engaged and politically empowered, the group wall
discussion falls short of quality deliberation and offers little substantive information sharing.
There are clear limitations to this project however, mainly
the sample frame and external validity of our findings. While
our sample closely approximates the campus in terms of demographics, a better sample would be a nationally representative
panel. Future research should investigate these measures, these
findings, and other relevant questions with better sampling techniques in a panel study. We anticipate that as research in this
field continues to grow in demand and interest that this will become easier to do. Furthermore, an interesting line of future endeavor should look at specific forms of political participation as
they are facilitated through new media. As part of this, we
should continue to expand our understanding of what it means
to be a political participant in the era of new media as these
definitions, and survey measures, should continue to change
The authors would like to thank Bruce Bimber for his thoughtful
feedback throughout our research, as well as M. Kent Jennings and
Eric R.A.N. Smith for their comments on previous versions of this
First-stage regression on political group membership for political participation (PP).
Variables b (se) p > |z|
Female (S) .21 (.11) .05
Age (A) .04 (.07) .55
Family income (I) .07 (.03) .05
Party identification (ID) .09 (.04) .02
Year in school (Y) .12 (.09) .21
Political interest (PI) .26 (.04) .00
White (R) .16 (.12) .18
Online news user (ON) .01 (.01) .13
Privacy restrictions (P) .06 (.07) .34
Length of FB membership (LM) .15 (.06) .01
Frequency log-in (FL) .08 (.05) .09
Adjusted R2 .263
Note: Instrumental variables in bold. Data derived from survey of 455 college
undergraduates. Unstandardized regression coefficients with standard error in
First-stage regression on political group membership for political knowledge (PK).
1544 M. Conroy et al. / Computers in Human Behavior 28 (2012) 1535â€“1546
Variables b (se) p > |z|
Female (S) .17 (.12) .16
Age (A) .13 (.07) .07
Family income (I) .04 (.04) .24
Party identification (ID) .10 (.05) .03
Year in school (Y) .01 (.10) .93
Political interest (PI) .27 (.05) .00
White (R) .21 (.13) .10
Online news user (ON) .01 (.01) .24
Privacy restrictions (P) .14 (.07) .06
Length of FB membership (LM) .11 (.06) .06
Frequency log-in (FL) .08 (.05) .11
Adjusted R2 .235
Note: Instrumental variables in bold. Data derived from survey of 455 college
undergraduates. Unstandardizedregression coefficients with standard error in
Limited information maximum likelihood regression.
Variables b (se) p > |z| b (se) p > |z|
Female (S) .28 (.52) .60 .71 (.27) .01
Age (A) .30 (.31) .34 .41 (.19) .03
.24 (.16) .14 .12 (.08) .13
.35 (.22) .12 .07 (.12) .55
Year in school
.06 (.40) .89 .38 (.19) .05
.65 (.40) .11 .09 (.25) .70
White (R) .49 (.56) .38 .33 (.30) .28
.03 (.03) .27 .00 (.01) .99
.48 (.31) .12 .13 (.18) .49
2.80 (1.33) 0.04* 1.16 (.83) 0.17**
R2 .378 0.085
Adjusted R2 .345 0.034
F 4.00 4
N 199 191
Note: Data derived from survey of 455 college undergraduates. Unstandardized
regression coefficients with standard error in parentheses.
* Conditional likelihood ratio p = .0661. ** Conditional likelihood ratio p = .1773.
Andersen, V. N., & Hansen, K. M. (2007). How deliberation makes better citizens:
The Danish deliberative poll on the Euro. European Journal of Political Research,
Ansolabehere, Stephen, & Shanto Iyengar (1995). Going negative: How political
advertisements shrink and polarize the electorate. New York, NY: The Free Press.
Bargh, J. A., McKenna, K. Y. A., & Fitzsimmons, G. M. (2002). Can you see the real me?
Activation and expression of the â€˜true selfâ€™ on the Internet. Journal of Social
Issues, 59, 33â€“48.
Barabas, J. (2004). How deliberation affects policy opinions. American Political
Science Review, 98(4), 687â€“701.
Benhabib, S. (1994). Deliberative rationality and models of democratic legitimacy.
Constellations, 1(1), 26â€“52.
Bentivenga, S. (2006). Rethinking politics in the world of ICTs. European Journal of
Communication, 21(3), 331â€“343.
Berger, B. (2009). Political theory, political science and the end of civic engagement.
Perspectives on Politics, 7(2), 335â€“350.
Bimber, B. (2008). The Internet and political fragmentation. In P. F. Nardulli (Ed.),
Domestic perspectives on contemporary democracy. Chicago, IL: University of
Bimber, B., Flanagin, A., & Stohl, C. (2005). Reconceptualizing collective action in the
contemporary media environment. Communication Theory, 15(4), 365â€“388.
Brehm, J., & Rahn, W. (1997). Individual-level evidence for the causes and
consequences of social capital. American Journal of Political Science, 41(3),
Brundidge, J., & Rice, R. E. (2009). Do the information rich get richer and the likeminded more similar? In A. Chadwick & P. N. Howard (Eds.), Routledge handbook
of Internet politics. New York, NY: Routledge Press.
Coleman, J. S. (1988). Social capital in the creation of human capital. American
Journal of Sociology, 94(s1), S95.
Cho, J., Shah, D. V., McLeod, J. M., McLeod, D. M., Scholl, R. M., & Gotlieb, R. M. (2009).
Campaigns, reflection, and deliberation: Advancing an O-S-R-O-R model of
communication effects. Communication Theory, 19(1), 66â€“88.
Dahlgren, P. (2006). The Internet, public spheres, and political communication:
Dispersion and deliberation. Political Communication, 22(2), 147â€“162.
de Tocqueville, A. (1990). Democracy in America. New York, NY: Vintage Press.
Dekker, P., & Uslaner, E. M. (2001). Social capital and participation in everyday life.
New York, NY: Routledge Press.
Delli Carpini, M. X., Cook, F. L., & Jacobs, L. R. (2004). Public deliberation, discursive
participation, and citizen engagement: A review of the empirical literature.
Annual Review of Political Science, 7, 315â€“344.
Delli Carpini, M. X., & Keeter, S. (1997). What Americans know about politics and why
it matters. New Haven, NH: Yale University Press.
Druckman, J., & Chong, D. (2007). Framing theory. Annual Review of Political Science,
Druckman, J., & Nelson, K. R. (2003). Framing and deliberation: How citizensâ€™
conversations limit elite influence. American Journal of Political Science, 47,
Ellison, N., Steinfield, C., & Lampe, C. (2011). Connections strategies: Social capital
implications of Facebook-enabled communication practices. New Media &
Society, 13(6), 873â€“892.
Ellison, N., Steinfield, C., & Lampe, C. (2007). The benefits of Facebook â€˜â€˜friends:â€™â€™
Social capital and college students use of online social network sites. Journal of
Computer-Mediated Communication, 12(4), 1143â€“1168.
Eveland, W. P. (2004). The effects of political discussion in producing informed
citizens: The roles of information, motivation and elaboration. Political
Communication, 21(2), 177â€“193.
Fishkin, J. S. (1991). Democracy and deliberation: New directions for democratic
reform. New Haven, NH: Yale University Press.
Fung, A. (2002). One city, two systems: Democracy in an electronic chat room in
Hong Kong. Javnost/The Public, 9(2), 77â€“94.
Gastil, J., Deess, E. P., & Weisler, P. (2002). Civic awakening in the jury room: A test
of the connection between jury deliberation and political participation. Journal
of Politics, 64(2), 585â€“595.
Gastil, J., & Dillard, J. P. (1999). Increasing political sophistication through public
Deliberation. Political Communication, 16(1), 3â€“23.
Gil de Zuniga, H., & Valenzuela, S. (2011). The mediating path to a stronger
citizenship: Online and offline networks, weak ties, and civic engagement.
Communication Research, 38(3), 397â€“421.
Hardy, B. W., & Scheufele, D. A. (2005). Examining differential gains from Internet
use: Comparing the moderating role of talk and online interactions. Journal of
Communication, 55(1), 71â€“84.
Hampton, K., & Wellman, B. (2001). Long distance community in the network
society: Contact and support beyond Netville. American Behavioral Scientist,
Huckfeldt, R. (2007). Unanimity, discord, and the communication of public opinion.
American Journal of Political Science, 51(4), 978â€“995.
Huckfeldt, R., & Mendez, J. M. (2007). Moths, flames and political engagement:
managing disagreement within communication networks. Journal of Politics,
Jennings, M. K., & Stoker, L. (2004). Social trust and civic engagement across time
and generations. Acta Politica, 39(4), 342â€“379.
Jennings, M. K., & Zeitner, V. (2003). Internet use and civic engagement: A
longitudinal analysis. Public Opinion Quarterly, 67(3), 311â€“334.
Klein, H. (1999). Tocqueville in Cyberspace. Using the Internet for citizen
associations. The Information Society, 15(4), 213â€“220.
Kolko, B., & Reid, E. (1998). Dissolution and fragmentation: Problems in on-line
communities. In S. Jones (Ed.), Cybersociety 2.0: Revisiting computer-mediated
communication and community. Thousand Oaks, CA: Sage Publications.
M. Conroy et al. / Computers in Human Behavior 28 (2012) 1535â€“1546 1545
Krueger, B. S. (2002). Assessing the potential of Internet political participation in the
United States: A resource approach. American Politics Research, 30(5), 476â€“498.
Lenhart, A., Madden, M., Macgill, A. & Smith, A. (2007). Teens and social media: The
use of social media gains a greater foothold in teen life as they embrace the
conversational nature of interactive online media. <www.pewinternet.org/
Reports/2007/Teens-and-Social-Media.aspx> Retrieved 14.07.11.
Lewis, K., Kaufman, J., & Christakis, N. (2008). The taste for privacy: An analysis of
college student privacy settings in an online social world. Journal of ComputerMediated Communication, 14(1), 79â€“100.
McFarland, D. A., & Thomas, R. J. (2006). Bowling young: How youth voluntary
associations influence adult political participation. American Sociological Review,
McLeod, J. M., Scheufele, D. A., & Moy, P. (1999). Community, communication and
participation: The role of mass media and interpersonal discussion in local
political participation. Political Communication, 16, 315â€“336.
Mossberger, K., Tolbert, C. J., & McNeal, R. S. (2008). Digital citizenship: The Internet,
society, and participation. Cambridge, MA: MIT Press.
Mutz, D. C., & Martin, P. S. (2001). Facilitating communication across lines of
political difference. The role of mass media. The American Political Science
Review, 95(1), 97â€“114.
Nie, N. H. (2001). Sociability, interpersonal relations, and the internet: reconciling
conflicting findings. American Behavioral Scientist, 45(3), 420â€“435.
Nie, N., & Erbring, L. (2000). Internet and society: A preliminary report. <www.
Nisbet, M. C., & Scheufele, D. A. (2004). Political talk as a catalyst for online
citizenship. Journalism and Mass Communication Quarterly, 81, 877â€“896.
Norris, P. (2004). The bridging and bonding role of online communities. Harvard
International Journal of Press/Politics, 7, 3â€“13.
Norris, P. (2001). Digital Divide: Civic engagement, information poverty, and the
Internet. Cambridge, UK: Cambridge University Press.
Norris, P., & Jones, D. (1998). Editorial: Virtual democracy. Harvard International
Journal of Press/Politics, 3(1), 1â€“4.
Olson, M. (1965). The logic of collective action: Public goods and the theory of groups,
second printing with new preface and appendix. Cambridge, MA: Harvard
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), 197â€“215.
Price, V., & Capella, J. (2002). Online deliberation and its influence. The electronic
dialogue project in campaign 2000. IT & Society, 1(1), 303â€“329.
Putnam, R. D. (2000). Bowling alone: The collapse and revival of American community.
New York, NY: Simon & Schuster.
Rich, P. (1999). American voluntarism, social capital, and political culture. The
Annals of the American Academy of Political and Social Science, 565(1), 15â€“34.
Robinson, J. P., & Levy, M. R. (1986). Interpersonal communication and news
comprehension. Public Opinion Quarterly, 50(2), 160â€“175.
Rojas, H. (2008). Strategy versus understanding: How orientations toward political
conversation influence political engagement. Communication Research, 8,
Rojas, H., & Puig-i-Abril, E. (2009). Mobilizers mobilized: Information, expression,
mobilization and participation in the digital age. Journal of Computer Mediated
Communication, 14(4), 902â€“927.
Rosenstone, S. J., & Hansen, J. M. (1993). Mobilization, participation, and democracy in
America. New York, NY: Macmillan Publishing Company.
Scheufele, D. A. (2002). Examining differential gains from mass media and their
implications for participatory behavior. Communication Research, 29(1), 46â€“65.
Sears, D. (1986). College sophomores in the laboratory: Influences of a narrow data
base on Social Psychologys view of human nature. Journal of Personality and
Social Psychology, 51(3), 515â€“530.
Shah, D. V., Kwak, N., & Holbert, R. (2001). â€™â€™Connectingâ€™â€™ and â€˜â€˜disconnectingâ€™â€™ with
civic life: Patterns of Internet use and the production of social capital.
Communication Abstracts, 24(6), 743â€“793.
Shah, D. V., McLeod, J. M., & Yoon, S. (2001). Communication, context, and
community: An exploration of print, broadcast, and Internet influences.
Communication Research, 28(4), 464â€“506.
Shah, D., Cho, J., Eveland, W., & Kwak, N. (2005). Information and expression in a
digital age. Communication Research, 32(5), 531â€“565.
Shah, D. V., Cho, J., Nah, S., Gotlieb, M. R., Hwang, H., Lee, N., et al. (2007). Campaign
ads, online messaging, and participation: extending the communication
mediation model. Journal of Communication, 57, 676â€“703.
Son, J., & Lin, N. (2008). Social capital and civic action: A network-based approach.
Social Science Research, 37, 330â€“349.
Sparks, C. (2001). The Internet and the global public sphere. In W. L. Bennett & R. M.
Entman (Eds.), Mediated politics: Communication in the future of democracy. New
York: Cambridge University Press.
Stromer-Galley, J. (2003). Diversity of political conversation on the internet: Usersâ€™
perspectives. Journal of Computer-Mediated Communication, 8 <http://
Taber, C. S., & Lodge, M. (2006). Motivated skepticism in the evaluation of political
beliefs. American Journal of Political Science, 50(3), 755â€“769.
Tidwell, L., & Walthers, J. B. (2002). Computer-mediated communication effects on
disclosure, impressions, and interpersonal evaluations: Getting to know one another a bit at a time. Human Communication Research, 28, 317â€“348.
Valenzuela, S., Park, N., & Kee, K. F. (2009). Is there social capital in a social network
site?: Facebook use and college studentsâ€™ life satisfaction, trust, and
participation. Journal of Computer-Mediated Communication, 14(4), 875â€“901.
Verba, S., Schlozman, K. L., & Brady, H. (1995). Voice and equality: Civic voluntarism in
American Politics. Cambridge, MA: Harvard University Press.
Vitak, J., Zube, P., Smock, A., Carr Ellison, N., & Lampe, C. (2011). Itâ€™s complicated:
Facebook usersâ€™ political participation in the 2008 election. Cyberpsychology,
Behavior & Social Networking, 14(3), 107â€“114.
Wellman, B. (2001). Little boxes, glocalization, and networked individualism. In M.
Tanabe, P. van den Besslaar, & T. Ishida (Eds.), Digital cities. Berlin: Springer
Wellman, B., Haase, A. Q., Witte, J., & Hampton, K. (2001). Does the Internet increase,
decrease, or supplement social capital? Social networks, participation, and
community commitment. American Behavioral Scientist, 45(3), 436â€“455.
Williams, D. (2007). The impact of time online social capital and
cyberbalkanization. CyberPsychology and Behavior, 10(3), 398â€“406.
Wyatt, R. O., Katz, E., & Kim, J. (2000). Bridging the spheres: Political and personal
conversation in public and private spaces. The Journal of Communication, 50(1),
Xenos, M., & Moy, P. (2007). Direct and differential effects of the Internet on political
and civic engagement. Journal of Communication, 57(4), 704â€“718.
Zywica, J., & Danowski, J. (2008). The faces of Facebookers: Investigating social
enhancement and social compensation hypotheses: Predicting Facebook and
pffline popularity from sociability and self-esteem, and mapping the meanings
of popularity with semantic networks. Journal of Computer-Mediated
Communication, 14(1), 1â€“34.
1546 M. Conroy et al. / Computers in Human Behavior 28 (2012) 1535â€“1546
Get Professional Assignment Help Cheaply
Are you busy and do not have time to handle your assignment? Are you scared that your paper will not make the grade? Do you have responsibilities that may hinder you from turning in your assignment on time? Are you tired and can barely handle your assignment? Are your grades inconsistent?
Whichever your reason is, it is valid! You can get professional academic help from our service at affordable rates. We have a team of professional academic writers who can handle all your assignments.
Why Choose Our Academic Writing Service?
- Plagiarism free papers
- Timely delivery
- Any deadline
- Skilled, Experienced Native English Writers
- Subject-relevant academic writer
- Adherence to paper instructions
- Ability to tackle bulk assignments
- Reasonable prices
- 24/7 Customer Support
- Get superb grades consistently
Online Academic Help With Different Subjects
Students barely have time to read. We got you! Have your literature essay or book review written without having the hassle of reading the book. You can get your literature paper custom-written for you by our literature specialists.
Do you struggle with finance? No need to torture yourself if finance is not your cup of tea. You can order your finance paper from our academic writing service and get 100% original work from competent finance experts.
While psychology may be an interesting subject, you may lack sufficient time to handle your assignments. Don’t despair; by using our academic writing service, you can be assured of perfect grades. Moreover, your grades will be consistent.
Engineering is quite a demanding subject. Students face a lot of pressure and barely have enough time to do what they love to do. Our academic writing service got you covered! Our engineering specialists follow the paper instructions and ensure timely delivery of the paper.
In the nursing course, you may have difficulties with literature reviews, annotated bibliographies, critical essays, and other assignments. Our nursing assignment writers will offer you professional nursing paper help at low prices.
Truth be told, sociology papers can be quite exhausting. Our academic writing service relieves you of fatigue, pressure, and stress. You can relax and have peace of mind as our academic writers handle your sociology assignment.
We take pride in having some of the best business writers in the industry. Our business writers have a lot of experience in the field. They are reliable, and you can be assured of a high-grade paper. They are able to handle business papers of any subject, length, deadline, and difficulty!
We boast of having some of the most experienced statistics experts in the industry. Our statistics experts have diverse skills, expertise, and knowledge to handle any kind of assignment. They have access to all kinds of software to get your assignment done.
Writing a law essay may prove to be an insurmountable obstacle, especially when you need to know the peculiarities of the legislative framework. Take advantage of our top-notch law specialists and get superb grades and 100% satisfaction.
What discipline/subjects do you deal in?
We have highlighted some of the most popular subjects we handle above. Those are just a tip of the iceberg. We deal in all academic disciplines since our writers are as diverse. They have been drawn from across all disciplines, and orders are assigned to those writers believed to be the best in the field. In a nutshell, there is no task we cannot handle; all you need to do is place your order with us. As long as your instructions are clear, just trust we shall deliver irrespective of the discipline.
Are your writers competent enough to handle my paper?
Our essay writers are graduates with bachelor's, masters, Ph.D., and doctorate degrees in various subjects. The minimum requirement to be an essay writer with our essay writing service is to have a college degree. All our academic writers have a minimum of two years of academic writing. We have a stringent recruitment process to ensure that we get only the most competent essay writers in the industry. We also ensure that the writers are handsomely compensated for their value. The majority of our writers are native English speakers. As such, the fluency of language and grammar is impeccable.
What if I don’t like the paper?
There is a very low likelihood that you won’t like the paper.
- When assigning your order, we match the paper’s discipline with the writer’s field/specialization. Since all our writers are graduates, we match the paper’s subject with the field the writer studied. For instance, if it’s a nursing paper, only a nursing graduate and writer will handle it. Furthermore, all our writers have academic writing experience and top-notch research skills.
- We have a quality assurance that reviews the paper before it gets to you. As such, we ensure that you get a paper that meets the required standard and will most definitely make the grade.
In the event that you don’t like your paper:
- The writer will revise the paper up to your pleasing. You have unlimited revisions. You simply need to highlight what specifically you don’t like about the paper, and the writer will make the amendments. The paper will be revised until you are satisfied. Revisions are free of charge
- We will have a different writer write the paper from scratch.
- Last resort, if the above does not work, we will refund your money.
Will the professor find out I didn’t write the paper myself?
Not at all. All papers are written from scratch. There is no way your tutor or instructor will realize that you did not write the paper yourself. In fact, we recommend using our assignment help services for consistent results.
What if the paper is plagiarized?
We check all papers for plagiarism before we submit them. We use powerful plagiarism checking software such as SafeAssign, LopesWrite, and Turnitin. We also upload the plagiarism report so that you can review it. We understand that plagiarism is academic suicide. We would not take the risk of submitting plagiarized work and jeopardize your academic journey. Furthermore, we do not sell or use prewritten papers, and each paper is written from scratch.
When will I get my paper?
You determine when you get the paper by setting the deadline when placing the order. All papers are delivered within the deadline. We are well aware that we operate in a time-sensitive industry. As such, we have laid out strategies to ensure that the client receives the paper on time and they never miss the deadline. We understand that papers that are submitted late have some points deducted. We do not want you to miss any points due to late submission. We work on beating deadlines by huge margins in order to ensure that you have ample time to review the paper before you submit it.
Will anyone find out that I used your services?
We have a privacy and confidentiality policy that guides our work. We NEVER share any customer information with third parties. Noone will ever know that you used our assignment help services. It’s only between you and us. We are bound by our policies to protect the customer’s identity and information. All your information, such as your names, phone number, email, order information, and so on, are protected. We have robust security systems that ensure that your data is protected. Hacking our systems is close to impossible, and it has never happened.
How our Assignment Help Service Works
1. Place an order
You fill all the paper instructions in the order form. Make sure you include all the helpful materials so that our academic writers can deliver the perfect paper. It will also help to eliminate unnecessary revisions.
2. Pay for the order
Proceed to pay for the paper so that it can be assigned to one of our expert academic writers. The paper subject is matched with the writer’s area of specialization.
3. Track the progress
You communicate with the writer and know about the progress of the paper. The client can ask the writer for drafts of the paper. The client can upload extra material and include additional instructions from the lecturer. Receive a paper.
4. Download the paper
The paper is sent to your email and uploaded to your personal account. You also get a plagiarism report attached to your paper.