Research in political communication recognizes the role of
interpersonal communication in the campaign process, with research
looking at interpersonal influence, opinion leadership, and anticipated
conversations with other individuals (Powell & Cowart, 2003). Voters
discuss political topics with others (Huckfeldt & Sprague, 1995),
using those discussions to evaluate information while forming their own
opinions (Mutz, 1998). Their conversations include exchanges of
information, political arguments, and issue-specific news (Wyatt, Kim
& Katz, 2000). Other research has found a link between television
viewing and interpersonal attraction of public figures (Antecol, 1998).
Media audiences often respond to celebrities and television characters
on a quasi-interpersonal level similar to that of an interpersonal
friendship (Powell & Anderson, 1984). Surlin (1974), for example,
argued that media audiences consider television characters to be
surrogate friends. Similarly, television characters can be perceived as
if they were real people under a social facilitation paradigm (Gardner
& Knowles, 2008), particularly when the individual has unfulfilled
social needs (Ashe & McCutcheon, 2001; Giles & Maltby, 2004;
Wang, Fink, & Cai, 2008). Rubin and McHugh (1987) reported that such
behavior was related to social attraction, while Turner (1993)
identified homophily as a factor in the strength of para-social
relationships. Homophily, is what Barker (2008) described as an affinity
for "people like me" (p. 21). However, past research has
focused primarily on media celebrities (Rubin, Perse, & Powell,
1985; Turner, 1993) or athletes (Brown & Basil, 1995; Brown, Basil,
& Bocarnea, 2003; Brown, Duane, & Fraser, 1997). Little research
has looked at the effect in terms of political candidates. This is an
oversight, considering the implications of both interpersonal
communication and Internet communication in political campaigns.
With the Internet playing an increasingly important role in both
mass and interpersonal communication, the need for research in the area
has also grown (Morris & Ogan, 1996; Newhagen & Rafaeli, 1996).
However, the relationship between the Internet and political
communication has been limited in that many political campaigns often
relied on traditional media when communicating with voters. That changed
with the 2008 presidential election, which became the first presidential
election in which both candidates were positioned to use social networks
as a campaign tool (Hendricks & Denton, 2010).
Previous research has identified four primary motivations for
seeking online political information: guidance, information-seeking,
entertainment, and social utility (Kaye & Johnson, 2002). Of primary
interest to this study is social, or interpersonal, utility, i.e.,
"using the Internet to reinforce decisions and arm individuals with
information to use in discussions with others" (p. 62). The concept
involves the need to keep up with current political events for the
purpose of discussing them with friends and co-workers (Swanson, 1976).
Social utility motivations have been reported more often for individuals
with a high interest in political campaigns and issues (Kaye &
Johnson, 2002). One by-product of social utility is that it can increase
information seeking and knowledge of political events (Kitchens, Powell,
& Williams, 2003). Diddi and LaRose argue that this behavior is
particularly common among college students, since college students read
more news online than their non-college peers and turn to the Internet
more often than any other source. Students who seek out news and
political information on the Internet are more politically active than
other students, more likely to be politically involved, and reported
higher informational motivation--including motivation for social utility
(Park, Kee, & Valenzuela, 2009). These results indicate that online
social networking sites could influence interpersonal images in the
political arena. Such sites have experienced explosive growth in recent
years, particularly among young people. Websites such as MySpace and
Facebook account for the largest usage, with possibly as many as 90% of
all college students checking Facebook daily (Rosenbush, 2006). Most
college students use Facebook to keep up with friends, but it is also
effective at establishing new relationships (Clark, Boyer, & Lee,
2007). Further, students respond to new online acquaintances in terms of
homophily. Similarly another study identified the interpersonal
communication aspect of these sites as a way to attach social identity
and a sense of social inclusion (Urista, Qingwen, & Day, 2008). That
effect is important to politics, since social inclusion is a means for
increasing civic and voter participation among young people (Edwards,
As noted earlier when discussing para-social behavior, previous
research (Gardner & Knowles, 2008; Rubin & McHugh, 1987; Surlin,
1974; Turner, 1993; Wang, Fink, & Cai, 2008) indicates some
individuals relate to political candidates on a quasi-interpersonal
basis, at least to the extent that social networking via the Internet
can simulate that type of communication. The quasi-interpersonal format
would imply that some individuals--particularly those who participate in
social networking sites related to a candidate--could evaluate those
candidates in terms of variables that traditionally fall within the
realm of interpersonal communication.
The effects of credibility, attraction, and homophily have received
more communication research interest than any other interpersonal
perception variables (Wensch, McCroskey, & Richmond, 2008).
Credibility refers to how believable we perceive a source to be. It is a
multidimensional construct with three components of competence,
trustworthiness, and caring/goodwill (McCroskey, & Teven, 1999).
Consistent with the image model developed by Nimmo and Savage (1976),
credibility was tested to provide an indication of cognitive impressions
of the candidates. Interpersonal attraction was initially developed as a
measure and studied in terms of three dimensions--physical attraction,
social attraction, and task attraction (McCroskey & McCain, 1974).
Rubin and McHugh (1987) subsequently identified a relationship between
social attraction and para-social behavior. The present study extended
the work of Rubin and McHugh by testing a possible relationship between
social attraction and para-social behavior related to political
The principle of homophily holds that the more similar two
communicators are, the more likely they are to interact with one
another, the more likely their communication will be successful, and the
more similar they become (McCroskey, Richmond, & Daly, (1975). The
more similar we perceive a source to be, the more likely we are to be
attracted to her/his beliefs, the more influence we allow her/him to
have over us. The three dimensions of homophily, as identified by
McCroskey, Richmond and Daly, are: demographic, background, and
attitude. Like interpersonal attraction, previous research had
identified a relationship between homophily and para-social behavior
(Turner, 1993). Thus, homophily was tested in the study to see if the
previous research related to para-social behavior would extend to
para-social behavior of political candidates.
The present study tested the variables, in terms of how they were
impacted by social media, in the 2008 presidential campaign that
featured Democrat Barack Obama and Republican John McCain. The use of
social media in the Obama campaign has been well documented (e.g.,
Hendricks & Denton, 2010). The use of social media by the McCain
campaign is sometimes overlooked, but McCain was actually a pioneer in
using the Internet campaign, first using it in his 2000 Republican
primary campaigns (Powell & Cowart, 2003). Young voters were
particularly receptive to the Obama campaign (Simba, 2009), and
Obama's campaign heavily used online media to communicate with
these voters (Hendricks & Denton, 2010). Further, the Obama campaign
was an active user of online communication outlets. By the end of the
2008 presidential election, the Obama campaign had more than 13 million
e-mail addresses in its computer, sent more than a billion e-mails and
had more than one million members in its texting program (Vargas, 2008).
As Kenski, Hardy and Jamieson (2010) concluded, "Better than any
campaign before it, the Democratic one located channels of direct
communication with its supporters" (p. 307).
Thus the dependent variables included in this study were
credibility, interpersonal attraction, and homophily--three constructs
pertinent to voter perceptions of political candidates. As a result,
this study tested the following hypotheses:
H1: Individuals who participated in a social networking site for a
political candidate will have higher ratings on (a) interpersonal
attraction, (b) homophily, and (c) credibility for that candidate than
supporters who did not participate in social networking sites.
H2: Information seeking will be positively associated with
perceptions of (a) interpersonal attraction, (b) homophily, and (c)
H3: Participation in an online social network for a candidate will
positively influence the relationship between information seeking and
perceptions of (a) interpersonal attraction, (b) homophily, and (c)
The participants were 265 students in introductory communication
classes at a state university. Participation on the part of the students
was voluntary. The use of this convenience sample is appropriate in this
particular study, given that young people (particularly college
students) are active users of the Internet and they also played an
active role in the 2008 election (Ruggeri, 2009). The participants were
provided with a written summary of the purpose of the study, and
assurance of the anonymous nature of their responses. Collection of the
data occurred in March 2009. Of the 265 respondents in the study, 65 had
participated in a Democratic/Obama social networking site, 45 in a
Republican/McCain social networking site, and 155 did not participate in
a political social networking site.
Participants were given a questionnaire with a series of questions
about the candidates in the 2008 presidential election. The
questionnaire included items to test candidate preference, and the
dependent variables of homophily, credibility, and interpersonal
attraction ratings for both candidates (Democrat Barack Obama and
Republican John McCain). Other questions gathered data on information
seeking, interpersonal utility, and participation in online social
Homophily was measured with the two-dimensional homophily scale
(attitude, background) scale developed by McCroskey, Richmond, and Daly
(1975). The alpha reliability for these scales is .80.
Credibility was measured with the three-dimensional source
credibility scale (character, competence, caring) developed by McCroskey
and Teven (1999). The alpha reliability for the three measures ranges
from .80 to .94.
Interpersonal attraction was measured with the three-dimensional
(social, task and physical attraction) interpersonal attraction scale
developed by McCroskey and McCain (1974). The alpha reliability for the
three measures ranges from the low .80s to the high 80s.
Interpersonal utility was measured with items taken from the
uses-gratification scales in mass communication research (Kaye &
Johnson, 2002). The three-item interpersonal utility measure had an
alpha reliability of .70.
Information seeking was measured with the five questions previously
employed by Kitchens, Powell, and Williams (2003). It previously
demonstrated reliability in excess of .80.
Participation in online social networking groups was determined by
asking the participants if they had participated in such a group during
the election and on behalf of which candidate.
Potential scores for each homophily factor ranged from 4 (low) to
28 (high) with a midpoint of 16. The overall mean for attitudinal
homophily in this study was 14.49 (SD = 6.22). The overall mean for
background homophily was 12.11 (SD = 5.55).
Scores on the interpersonal (social) attraction dimensions ranged
from 6 (low) to 30 (high), with a midpoint of 18. The overall mean for
social attraction was 16.17 (SD = 5.39).
Scores on the three credibility dimensions ranged from 7 (low) to
49 (high), with a midpoint of 28. The mean for competence was 29.61 (SD
= 8.14). The mean for character was 27.71 (SD = 8.04), and the mean for
caring was 27.00 (SD = 7.78).
Hypothesis One predicted that individuals who participated in a
social networking site for a political candidate would have higher
ratings on (a) interpersonal attraction, (b) homophily, and (c)
credibility for that candidate than supporters who did not participate
in social networking. Hypothesis One was tested with a series of
two-tailed, paired t-tests that compared the ratings for the candidates
from supporters who participated and did not participate in an online
social networking group for their candidate.
Hypothesis 1a was supported. There were significant differences
between social networking and non-social networking supporters in terms
of interpersonal (social) attraction ratings. The results were
consistent for both Obama (M = 16.49, SD = 6.16 for networking, M =
14.75, SD = 5.01 for non-networking, t = 4.71, p < 001, df = 216,
effect size = .019) and for McCain (M = 20.73, SD = 5.21 for networking,
M = 17.32, SD = 4.46 for non-networking, t =4.29, p <.001, df = 216,
size = .023).
Hypothesis 1b was supported for both candidates on attitudinal
homophily and background homophily. Obama's online supporters gave
him higher ratings on attitudinal homophily (M = 21.96, SD = 4.92) than
did those who did not participate in an online group (M = 15.12; SD =
5.81, t = 6.84, df = 216, p < 001, effect size = .090). McCain's
online supporters also gave him higher ratings on attitudinal homophily
(M = 19.95, SD = 5.42, versus M = 13.35, SD = 6.03; df = 216, t = 7.53,
p <.001, effect size = .090).
Similarly, Obama's ratings for background homophily were
higher for those who participated in social networking sites for the
candidate (M = 16.44, SD = 5.26) than for those who did not join such
social networks (M = 11.97, SD = 4.71, McCain's ratings on
background homophily were also higher for social network participants (M
= 19.95, SD = 5.42) than for non-social network users (M = 13.35, SD =
5.28) t = 4.20, p <.001.
Hypothesis 1c was supported for ratings of Barack Obama, but not
for John McCain. Credibility ratings for Barack Obama were higher among
online supporters than among those who did not belong to a social
networking group. This result was consistent for character ratings (M =
37.25, SD = 5.03, versus M = 30.18, SD = 8.05, t = 1.97, df = 119,
effect = .070, p <.05), competence ratings (M = 37.34, SD = 4.29,
versus M = 36.01, SD = 6.33, t = 1.96, df = 119, effect = .047, p <
.05), and caring ratings (M = 36.19, SD = 5.14, versus M = 32.78, SD =
7.11, t = 2.90, df = 119, effect size = .105, p <.001).
Hypothesis Two predicted that information seeking would be
positively associated with perceptions of (a) interpersonal attraction,
(b) homophily, and (c) credibility. Hypothesis Two was tested with
Pearson's correlation coefficient to detect the predicted
association between information seeking and the dependent
variables--credibility, interpersonal attraction, and homophily
Hypothesis Two was supported for ratings of Barack Obama, but not
for ratings of John McCain. Information seeking was negatively
associated with ratings of Obama on attitudinal homophily (r = -.29, p
< .01), background homophily (r = -.29, p < .01), character (r =
-.23, p < .05), competence (r = -.22, p < .05) and caring (r =
-.24, p < .05). Information seeking was positively associated with
ratings of Obama on social attraction (r = 29, p < .01), task
attraction (r = .24, p < .05) and physical attraction (r = .22, p
Hypothesis Three predicted that participation in an online social
network for a candidate would influence the relationship between
information seeking and perceptions of (a) interpersonal attraction, (b)
homophily, and (c) credibility. Hypothesis Three was also tested with
Pearson's correlation coefficient, with separate analyses for those
who participated in online social networking groups for a candidate and
those who did not participate in such activity.
Hypothesis 3a was not supported. There were no significant
relationships between information seeking and ratings of interpersonal
attraction between participants (r = .20 for Obama, r = .35 for McCain)
and non-participants (r = .15 for Obama, r = .01 for McCain) in social
networking sites for a candidate. The n for McCain supporters was 50,
thus the .35 coefficient barely missed significance at the .05 level.
Hypothesis 3b was supported. For Obama supporters, there was a
significant relationship for those who did not participate in online
social networks in terms of both attitudinal homophily (r = -.26, p <
background homophily (r = -.25, p < .05). For McCain supporters
who did not participate in an online group, there was a positive
relationship on background homophily (r = .20, p < .05).
Hypothesis 3c was not supported. There were no significant
differences in the relationship between information seeking and ratings
of credibility between participants (r = .09 for Obama, r = .05 for
McCain) and non-participants (r = .13 for Obama, r = .05 for McCain) in
social networking sites for a candidate.
Voters made interpersonal judgments about candidates, and those
judgments were partially related to participation in an online social
network for that candidate. In addition, how voters used information
also influenced interpersonal judgments of a candidate. Specifically,
there were positive correlations between interpersonal attraction
ratings of Barack Obama and information seeking and interpersonal
utility. For John McCain, there was a relationship between interpersonal
utility needs and ratings of both social and task attraction.
Homophily was a key variable in this study. Seeking information
related to Barack Obama was negatively associated with homophily and
credibility ratings. The more respondents viewed themselves as
dissimilar to Obama, the more likely they were to seek information about
him. This relationship was significant for both attitudinal and
background homophily) (r = -.29 for each, p < .05), but accounted for
less than 9% of the variance in each case. In terms of credibility,
information seeking related to Obama was negatively associated with
respondents' ratings of his character (r = -.23, p < .05),
competence (r = -.22, p < .05) and caring (r = -.24, p < .05).
This indicates that information seeking on political candidates may be
triggered more by negative campaigning than by interest in positive
information about the candidate.
However, seeking information about Obama was positively related to
the respondents' interpersonal attraction ratings of the candidate.
That effect was consistent for task attraction, social attraction and
physical attraction. Information seeking for McCain was significantly
but minimally associated positively with his character ratings.
Information seeking was related to participation in a social
networking group, but not in a positive manner. Mean information scores
were higher for those who were not members of a political social
networking group (M = 14.13, SD = 3.92) than for those who were part of
an Obama social networking group (M = 12.06, SD = 2.98). Mean scores for
those in a McCain social networking group (M = 13.59, SD = 3.33) were
slightly below those of non-network respondents. However, there was no
significant difference on means of interpersonal utility in terms of
social networking participation. This result is counter to previous
research on para-social behavior that found that audiences developed
positive relationships, or, "friendships" with television
characters that increased their susceptibility to attitude change
(Surlin, 1974). Still, the current finding indicates that at least some
of the concepts related to para-social behavior cannot be directly
applied to Internet communication.
The results of this study indicate that information seeking was
positively related to the interpersonal attraction the voters felt
toward Obama. This is consistent with what would be expected from a
highly charismatic campaigner. However, information seeking was not
linked to homophily, but to the dissimilarities of the seeker. These
were unexpected findings that add to the depth of our understanding of
political theory because they are counter to previous work by Surlin
(1974) and Kitchens, Powell, and Williams (2003). More research is
needed to identify the impact of dissimilarity on information seeking.
There are limitations to the study. The survey pool was young,
instead of seasoned voters, and future research should expand the sample
beyond college-age participants. However, it should be noted that these
are the voters of the present and the future, and social networking is a
major part of their lives. Still, the concepts tested in this study need
to be replicated with a broader range of participants.
However, this study indicates that there is an interpersonal
connection between a candidate and his or her audience. Currently that
connection is greater outside of a social network site, but that could
change as the Internet becomes a more frequent means of communicating
with voters. Future research should continue to examine this
relationship. Further, since young people use social media more
frequently than the elderly (Urista, Qingwen, & Day, 2008), as the
current pool of young voters mature, research should follow them to
identify if these findings remain consistent or to verify if social
networking has an increasing impact on voting behavior.
In general, these results indicate that respondents do form images
of political candidates that can be measured with interpersonal image
variables. Further, this study indicates that Internet usage,
particularly in terms of social networking, is related to those
interpersonal images. However, the specific nature of those
relationships appears to vary, depending up the voter perception of
individual candidates. Thus additional research is needed to refine our
understanding of how the Internet impacts candidate images.
Antecol, M. (1998). Understanding McLuhan: Television and the
creation of the global village. ETC: A Review of General Semantics, 54,
Ashe, D. D., & McCutcheon, L. E. (2001). Shyness, loneliness,
and attitude toward celebrities. Current Research in Social Psychology,
6(9), 124-133. [on-line]. Available:
Barker, V. E. (2008-05-22). Older adolescents' motivations for
use of social networking sites: The influence of group identity and
collective self-esteem. Paper presented at the annual meeting of the
International Communication Association. Montreal, Quebec.
Brown, W. J., & Basil, M. D. (1995). Media celebrities and
public health: Response to "Magic" Johnson's HIV
disclosure and its impact on AIDS risk and high risk behaviors. Health
Communication, 7, 345-370.
Brown, W. J., Basil, M. D., & Bocarnea, C. (2003). The
influence of famous athletes on health beliefs and practices: Mark
McGwire, child abuse prevention, and androstenedione. Journal of Health
Communication, 8, 41-57.
Brown, W. J., Duane, J. J., & Fraser, B. P. (1997). Media
coverage and public opinion of the O.J. Simpson trial: Implications for
the criminal justice system. Communication Law and Policy, 2, 261-287.
Clark, N., Boyer, L., & Lee, S., (2007) A place of their own:
An exploratory study of college students ' uses of Facebook. Paper
presented at the annual meeting of the International Communication
Association. San Francisco, CA.
Diddi, A., & LaRose, R. (2006). Getting hooked on news: uses
and gratifications and the formation of news habits among college
students in an internet environment. Journal of Broadcasting &
Electronic Media, 50, 193-210.
Edwards, K. (2008). Social inclusion and youth participation. Youth
Studies Australia, 27, 11-17.
Gardner, W. L., & Knowles, M. L. (2008). Love makes you real:
Favorite television characters are perceived as 'real' in a
social facilitation paradigm. Social Cognition, 26, 156-168.
Giles, D. C., & Maltby, J. (2004). The role of media figures in
adolescent development: Relations between autonomy, attachment, and
interest in celebrities. Personality and Individual Differences, 36,
Hendricks, J. A., & Denton, R. E. (eds.) (2010).
Communicator-in-chief: How Barack Obama used new media technology to win
the White House. Lanham, MD: Lexington Books.
Huckfeldt, R. R., & Sprague, J. (1995). Citizens, politics
and social communication: Information and influence in an election
campaign. New York: Cambridge University Press.
Kaye, B. K., & Johnson, T. (2002). Online and in the know: Uses
and Gratifications of the web for political information. Journal of
Broadcasting & Electronic Media, 46, 54-71.
Kenski, K., Hardy, B. W., & Jamieson, K. H. (2010). The Obama
victory: How media, money and message shaped the 2008 election. New
York: Oxford University Press.
Kitchens, J. T., Powell, L., & Williams, G. (2003). Information
please? Information seeking, mass media, and the undecided voter.
Communication Research Reports, 20, 73-81.
McCroskey, J. C., & McCain, T. A. (1974). The measurement of
interpersonal attraction. Speech Monographs, 41, 261-266.
McCroskey, J. C., Richmond, V. P., & Daly, J. A. (1975). The
development of a measure of perceived homophily in interpersonal
communication. Human Communication Research, 1, 323-332.
McCroskey, J. C., & Teven, J. J. (1999).Goodwill: A
reexamination of the construct and its measurement. Communication
Monographs, 66, 90-103.
Morris, M., & Ogan, C. (1996). The Internet as a mass medium.
Journal of Communication, 46, 39-50.
Mutz, D. C. (1998) Interpersonal influence: How perceptions of mass
collectives affect political attitudes. Cambridge: Cambridge University
Newhagan, J., & Rafaeli, S. (1996). Why communication
researchers should study the Internet: A dialog. Journal of
Communication, 46, 4-13.
Nimmo, D., & Savage, R. L. (1975). Candidates and their images.
Pacific Palisades, CA: Goodyear.
Park, N., Kee, K., & Valenzuela, S. (2009). Being immersed in
social networking environment: Facebook groups, uses and gratifications,
and social outcomes. CyberPsychology & Behavior, July 20,
Powell, L., & Anderson, R. G. (1984). The impact of TV comedy:
A test of the friendship theory. Journal of Communication Studies, 3,
Powell, L., & Cowart, J. (2003). Political campaign
communication: Inside and out. Boston: Allyn & Bacon.
Rosenbush, S. (2006). Socializing for collars. Business Week
Online, April 10. Rubin, A. M., & McHugh, M. P. (1987). Development
of parasocial interaction relationships. Journal of Broadcasting &
Electronic Media, 31, 279-292.
Rubin, A. M., Perse, E., & Powell, R. A. (1985). Loneliness,
parasocial interaction and local television viewing. Human Communication
Research, 12, 155-180.
Ruggeri, A. (2009, November). The youth movement. U.S. News &
World Report, 146, 30-33.
Simba, M. (2009). The Obama campaign 2008: A historical overview.
Western Journal of Black Studies, 33, 186-191.
Surlin, S. H. (1974). Bigotry on the air and in life: The Archie
Bunker case. Pubic Telecommunication Review, 2, 34-41.
Swanson, D. L. (1976). Information utility: An alternative
perspective in political communication. Central States Speech Journal,
Turner, J. R. (1993). Interpersonal and psychological predictors of
parasocial interaction with different television performers.
Communication Quarterly, 41, 443-453.
Urista, M., Qingwen, D., & Day, K. (2008). Explaining why young
adults use MySpace and Facebook through uses and gratifications theory.
Paper presented at the National Communication Association Annual
Conference. San Diego, CA.
Vargas, A. J. (2008). Obama raised half a billion online.
WashingtonPost.com http://voices.washingtonpost.com/44/2008/11/obama-raised-half-a-billion- on.html
Wang, Q., Fink, E. L., & Cai, D. A. (2008). Loneliness, gender
and parasocial interaction: A uses and gratifications approach.
Communication Quarterly, 56, 87-109.
Wensch, J. S., McCroskey, J. C., & Richmond, V. P. (2008).
Human communication in everyday life: Explanations and applications.
Pearson, Boston: MA.
Wyatt, O., Kim, J., & Katz, E. (2000). How feeling free to talk
affects ordinary political conversation, purposeful argumentation, and
civic participation. Journalism and Mass Communication Quarterly, 77,
Virginia P. Richmond
University of Alabama at Birmingham
Glenda C. Williams
University of Alabama
Author info: Correspondence should be sent to: Dr. Larry Powell,
Department of Communications, University of Alabama at Birmingham, 1401
University Ave., Birmingham, AL 35294. e-mail firstname.lastname@example.org.