channel differences in political campaign...
TRANSCRIPT
CHANNEL DIFFERENCES IN POLITICAL CAMPAIGN ADVERTISEMENTS AND THEIR
EFFECTS ON INDIVIDUALS' COGNITIVE, EMOTIONAL, AND BEHAVIORAL
RESPONSES IN SOCIAL MEDIA CONTEXT
by
YEOJIN “JULIE” KIM
SHUHUA ZHOU, COMMITTEE CHAIR
WILLIAM J. GONZENBACH JENNIFER D. GREER
JAMES LEEPER CAROL B. MILLS
A DISSERTATION
Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy
in the College of Communication and Information Sciences in the Graduate School of
The University of Alabama
TUSCALOOSA, ALABAMA
2015
Copyright Yeojin “Julie” Kim 2015 ALL RIGHTS RESERVED
ii
ABSTRACT
This dissertation examined cognitive, emotional, and behavioral responses to different
channels of political campaign advertisements in the social media context. Different channels
such as social media platform (Facebook vs. Twitter), level of interactivity (high vs. low) and
modality (visual vs. audiovisual advertisements) were manipulated as predictors of campaign
learning, candidate evaluation, and political behavior. Several theories and concepts regarding
interactivity (i.e., the theory of user control, the theory of structural isomorphism, the media
richness theory, the social presence theory, cognitive load, and disorientation) and modality (i.e.,
the dual coding theory, the cue summation theory, the limited-capacity information processing
theory, and multiple resource theory) were employed to investigate what conditions of modality
and interactivity embedded within social media have the strongest effects on users’ cognitive,
emotional, and behavioral responses in political campaign advertisements. Data were collected
from 317 participants who completed the online experiment via Amazon Mechanical Turk.
Findings demonstrated that interactivity, modality, and social media platform differently
influenced campaign learning, candidate evaluation, and political behavior. Facebook generated
greater campaign learning than Twitter without consideration of other factors such as
interactivity and modality, and audiovisual advertisements generated greater campaign learning
than visual counterparts. However, no significant effects of interactivity on political outcomes
were found. Interestingly, however, under combined conditions of social media platform and
modality, a significant interaction was found in campaign learning and candidate evaluation.
iii
Further, there were significant effects of social media when political, cognitive, and emotional
involvement with social media were added as covariates, showing that Twitter generated more
political behaviors than Facebook.
Conceptually, the present study provides a better understanding of the validity of
traditional theories about medium and communication channel, especially in the social media
context, by reexamining different conceptualizations of medium, channel, mode of presentation,
and media platform and by reviewing unique characteristics of social media. This study provides
theoretical contributions to the understanding of cognitive, emotional, and behavioral reactions
to different channels in the political campaign context by adding an explanation of the effects of
interactivity and modality in relation to social media. Practically, the findings of this study may
help candidates and political campaign practitioners build appropriate strategies for political
campaigns through social media platforms to encourage their target publics to engage in political
activities including political learning, candidate evaluation, and political behaviors.
iv
DEDICATION
This dissertation is dedicated to God, my loving parents, and other people who have
supported me with prayers, encouraging words, and endless love.
v
LIST OF ABBREVIATIONS AND SYMBOLS
α Alpha: Cronbach’s index of internal consistency df Degrees of freedom: number of values free to vary after certain restrictions have been
placed on the data F Fisher’s F ratio: A ration of two variances
M Mean: the sum of a set of measurements divided by the number of measurements in the set
Mdn. Median: value that separates the highest and lowest halves of a sample population
SD Standard deviation: value that represents the degree to which measurements in a set vary
in relation to the mean.
SE Standard error: estimate of the SD based on estimated means in a statistical model
p Probability associated with the occurrence under the null hypothesis of a value as extreme as or more extreme than the observed value
η2 Eta squared: the ratio of variance explained in the dependent variable by a predictor while controlling for other predictors
t Computed value of t test
< Less than
> Greater than
= Equal to
H Research Hypothesis
RQ Research Question
vi
ACKNOWLEDGMENTS
This dissertation would not have been completed without the guidance and support of
several individuals. When I felt down or depressed about writing my dissertation while applying
for jobs, their encouragement gave me strength.
I would like to express my deepest gratitude to my chair and advisor, Dr. Shuhua Zhou.
Without his letter of acceptance, I would never have been able to study and do research at the
University of Alabama. For four years, he has supported me emotionally and intellectually. He
has spent time reading drafts of my research papers and dissertation even on weekends and has
accommodated unscheduled office visits for me. Whenever I needed assistance with
recommendation letters, doing research, and adjusting myself to campus life, he helped me as
much as he could. His support has motivated me to work hard, to do my research in a timely
manner, and to dream of becoming a professor. I would also like to thank Dr. Jennifer Greer. As
a professor, as a researcher, and as a mother, her untiring efforts have been an example to me as
a female scholar. Despite her extremely busy schedule, she always tried to make everything
possible and successful for me. I will never forget doing several research projects with her and
talking with her about my job situation as well as her letter of recommendation for my job
search. Her insightful advice and comments on my research projects, dissertation, and job search
enabled me to think about what I can do at present and what I should do for my future as a
researcher and a professor. I am grateful to Dr. Carol Mills, who inspired me to love students as
a teacher. Whenever I was troubled, she always listened to my problems carefully, understood
my heart, and gave me good, warm-hearted advice. She was also pleased to give much time in
vii
writing recommendation letters for me. I could feel and know how much she cares about me.
Thanks to her continuous emotional support, I could get over difficult situations and come out on
top again. I would like to thank Dr. William Gonzenbach for his encouragement in my research
and practical advice for my teaching skills. He provided me an opportunity to be his teaching
assistant and taught me how to teach students. When I failed in my job search and attempts at
publication, he always gave me positive words and guided me to try once more. His help was a
great encouragement to me. I am indebted to Dr. James Leeper for his time to teach me statistics
and the long discussions we had that helped me analyze the data for my dissertation. Without
him, I would not have finished my dissertation on time. He has spent time answering my
questions and giving comments even during summer break. His insightful comments and
guidance on data, methodologies, and analyses helped me proceed. I warmly appreciate his
generosity and his passion for teaching. I would like to thank my peers, professors and staff at
the University of Alabama whom I have had the chance to learn from and work with while
completing the doctoral program. I am greatly indebted to the people of faith at the Korean
Presbyterian Church of Tuscaloosa, who made unforgettable, beautiful memories with me and
who have prayed for me during the past four and a half years.
To my parents, Unsik Kim and Myoungbin Im, whom I admire the most as instructors in
my life, thank you for your enduring support, sacrifices, and prayers. To my sister, brother, uncle
and aunt, I cannot express how grateful I am for your support and care throughout these years.
Finally, thanks to God, who is always with me and will keep me wherever I go and who will
bring me back to this land.
viii
CONTENTS
ABSTRACT ................................................................................................ ii
DEDICATION ........................................................................................... iv
LIST OF ABBREVIATIONS AND SYMBOLS ........................................v
ACKNOWLEDGMENTS ......................................................................... vi
LIST OF TABLES ..................................................................................... xi
LIST OF FIGURES .................................................................................. xii
CHAPTER ONE: INTRODUCTION ..........................................................1
CHAPTER TWO: LITERATURE REVIEW ..............................................7
Communication Channel .............................................................................7
Communication Channel in Political Campaigns .....................................7
Reexamination of McLuhan’s Concept of Medium ...............................12
Interactivity .............................................................................................18
Modality ..................................................................................................26
Cognition, Emotion, and Behavior ............................................................29
Cognitive, Emotional, Behavioral Effects of Interactivity .....................29
Cognitive, Emotional, Behavioral Effects of Modality ..........................36
Effects of Political Advertising ..................................................................45
Channel Differences in Political Advertising .........................................45
Cognitive, Emotional, and Behavioral Responses to Political Advertising ..............................................................................................46
Covariates ..................................................................................................50
ix
Cognitive and Emotional Involvement with Communication Channel ...50
Political Involvement ...............................................................................54
Overall Summary, Experimental Hypotheses and Research Questions ....57
CHAPTER THREE: RESEARCH METHOD ..........................................64
Participants and Design..............................................................................64
Stimuli ........................................................................................................65
Procedure ...................................................................................................66
Dependent Variables ..................................................................................67
Campaign Learning ..................................................................................67
Candidate Evaluation ...............................................................................68
Political Behavior.....................................................................................68
Covariates ..................................................................................................69
Cognitive Involvement with Social Media Platform ...............................69
Emotional Involvement with Social Media Platform ..............................70
Political Orientation .................................................................................71
Political Participation ...............................................................................71
CHAPTER FOUR: RESULTS ..................................................................72
Study Participants ......................................................................................72
Manipulation Check ...................................................................................74
Data Analysis .............................................................................................75
CHAPTER FIVE: DISCUSSION ..............................................................93
No Significant Effects of Interactivity on Political Outcomes ..................94
Interaction Effects of Modality and Social Media on Campaign Learning .....................................................................................................96
x
Interaction Effects of Modality and Social Media on Candidate Evaluation ..................................................................................................98
Effect of Social Media on Political Behavior ..........................................100
CHAPTER SIX: CONCLUSION ............................................................102
Study Limitations .....................................................................................104
REFERENCES ........................................................................................107
APPENDIX A ..........................................................................................129
APPENDIX B ..........................................................................................132
xi
LIST OF TABLES
1. Demographics of Study Participants ....................................................72 2. Mann-Whitney U Test of High and Low Interactive Channels on
Campaign Learning ..............................................................................76 3. Comparing Candidate Evaluation and Political Behavior between High and Low Interactive Channels ....................................................77 4. Mann-Whitney U Test of Facebook and Twitter on Campaign
Learning ...............................................................................................78 5. Comparing Candidate Evaluation and Political Behavior between
Facebook and Twitter ..........................................................................79 6. Mann-Whitney Test of Visual and Audiovisual Advertisements on
Campaign Learning ..............................................................................80 7. Comparing Candidate Evaluation and Political Behavior between
Visual and Audiovisual Advertisements ..............................................81 8. Tests of Between-Subjects Main and Interaction Effects: Interactivity, Modality and Social media as Predictors of Campaign
Learning ...............................................................................................82 9. Tests of Between-Subjects Main and Interaction Effects: Interactivity, Modality and Social media as Predictors of Candidate
Evaluation ............................................................................................85 10. Tests of Between-Subjects Main and Interaction Effects: Interactivity, Modality and Social Media as Predictors of Political
Behavior ...............................................................................................88 11. Results of Hypotheses and Research Questions ..................................89
xii
LIST OF FIGURES
1. Interaction of Modality and Social Media Platform on Campaign Learning ...............................................................................................83
2. Interaction of Modality and Social Media Platform on Candidate Evaluation ............................................................................................86
1
CHAPTER ONE
INTRODUCTION
During political campaigns, the communication channel is an integral means for political
candidates or parties to share political messages and promote their political goals (Kaid, 1981).
In U.S. elections, new technologies have radically changed the way campaigns are waged
(Medvic, 2011). Growing out of a long history of print media in the nineteenth century and the
significant role of radio in political campaigns through World War II, television has become a
powerful medium in United States politics (Purvis, 2001). Since televised political advertising
was first successfully used in the 1952 Eisenhower-Stevenson election, it has been widely
utilized by candidates to promote their campaign messages and has become the biggest media
impact maker in U.S. elections (James, 2010; Kaid, 1999; Kaid & Johnston, 2001; NPR, 2014;
Poor, 2013; Tedesco, McKinnon, & Kaid, 1996). The 1960 Kennedy-Nixon debates, the first
televised presidential debates in the U.S., were pivotal in demonstrating that candidates’ images
on television dominate the electoral process (Keating & Latane, 1976).
With the development of digital technologies, candidates have taken advantage of the
wide reach of the Internet as a new, important campaign tool. Candidates use their websites to
communicate with their public, elicit campaign-relevant outcomes such as fundraising and
volunteer mobilization, and even convert traditional messages such as press releases and
television advertisements to a digital format (Kaid & Postelnicu, 2005). As such, Internet
advertising incorporates the text and visual information of print advertising as well as the
2
audiovisual information of television advertising (Dijkstra, Buijtels, & van Raaij, 2005; Snyder,
1996).
Barack Obama led the “digital revolution” in the 2008 presidential campaign (McKinney
& Banwart, 2011). When Obama was elected president, scholars claimed that “while many
presidential elections become nothing more than a mere footnote in history, the 2008 campaign
and election changed history” (McKinney & Banwart, 2011, p. 1). This notion highlights not
only that Obama is the first African-American president in the United States, but also that he
inspired young voters’ engagement by successfully using social media such as Facebook and
Twitter in his presidential campaign in the “social media election” (McKinney & Banwart, 2011;
Rutledge, 2013).
In the 2012 U.S. presidential election, the top presidential candidates, Mitt Romney and
Barack Obama, developed social media strategies to successfully engage with voters (Andrews,
2015). Political analysts claimed that social media, especially Twitter, was a big factor in
Obama’s victory (Gaudin, 2012). The 2012 election has been dubbed “the first Twitter election”
because the largest numbers of tweets in the history of social media related to a single event were
sent throughout the campaign (Capobres, 2012; Fouhy, 2012). On election night, Obama victory
tweets lit up the “Twittersphere” and became the most retweeted posts ever (Tsukayama, 2012).
Noting the major trend in the 2015 midterm elections that “candidates [are getting] really serious
about their personal brand and their connection to the public,” analysts further forecast that
social media will play a still bigger role in the 2016 U.S. presidential campaign, especially
compared with television (ABC News, 2013; Andrews, 2015; Snyder, 2014, para 6; The
Associated Press, 2013).
3
However, in spite of the increasing use of social media and its growing importance for
political campaign advertisements based on its interactive nature, whether political advertising
exposure via social media such as Facebook and Twitter affects audience response to the
political process remains unknown. Empirical findings focusing on the effects of social media as
a political campaign tool provide mixed evidence. Some studies provide evidence that social
media in political campaigns increased voter’s political cognition, emotion, and behaviors
(Bollen, Mao, & Pepe, 2010; Hesseldahl, MacMillan, & Kharif, 2008; Marchese, 2008; Owen,
2008), while others found no significant relationship between social media use and such political
outcomes (Ancu & Cozma, 2009; Gil de Zúñiga, Puig, & Rojas, 2009). These studies
investigated the relationships between social media use and political outcomes without
considering the conceptualization of the social media. Therefore, this study investigates the
conceptualization of the social media in order to look at the effects of the social media platform
per se on political outcomes.
Moreover, prior research on the effects of social media has failed to consider the unique
characteristics of social media. Although candidates have employed a variety of social media
channels such as Facebook, Twitter, and YouTube for their political campaigns (Kennedy, 2013),
candidates use each of the channels in a distinct way to reach voters. During the 2008
presidential campaign, for example, the Obama camp used Twitter to send out voting reminders
and Facebook to interact with users (Rutledge, 2013). A 2014 report from the Pew Research
Center also supports the idea that different social media channels offer a different use experience
in the political context. According to the report, although social media play a role as prominent
sources of political information, two of the most popular social media, Facebook and Twitter,
provide users with different experiences of gaining political information based on the user’s
4
different purpose of using each medium (Gottfried, 2014). Specifically, more politically
interested users tend to use Twitter more than Facebook to gain political information because
Facebook offers more opportunity for users’ involvement in various social and informational
activities than Twitter (Gottfried, 2014). Delacourt (2014) also recognized the difference
between Facebook and Twitter, stating that “Facebook is personal, Twitter is political” (para 1).
In terms of channel effects in political advertisements, most previous researchers have
investigated the direct effects of political campaign advertising on political attitudes and
behaviors by focusing on political advertisements alone rather than considering the role of
political advertisements in relation to other factors underlying the effects such as the candidate
and advertising contents. As Kaid (2002, 2003) pointed out, research on channel differences in
political advertisements has never been very successful because some candidates appear to do
better on one medium, some on another. Further, prior studies failed to compare different
channels with the same contents, and the evidence is sparse about channel differences for
cognitive, emotional, and behavioral effects of political advertisements in the social media
context (Eveland & Dunwoody, 2002; Kaid, 2002, 2003). Given that the possibility of social
media as a catalyst to mobilize potential voters’ political cognition, emotion, and behaviors has
been continuously raised, however, the effects of channel differences in social media should be
investigated while taking into account other factors affecting outcomes.
Therefore, the purpose of this study is to investigate whether and/or how different
channels of political advertisements affect the audience’s cognitive, emotional, and behavioral
responses in the social media context. This study is designed to investigate the effects of
different social media channels on users’ processing of embedded visual and audiovisual ads in
terms of cognition, emotion, and behavioral intention. To examine different characteristics of
5
social media, the current study uses the concept of interactivity. Although this concept was
originally developed to explain users’ participation from a computer-mediated communication
perspective and relationships between users from an interpersonal communication perspective,
recent studies have applied the concept to social media, where both mass and interpersonal
communication take place (Ha & James, 1998). Social media are designed for interactivity, but
the functional and emotional engagements of each social medium are distinct from one another.
Hence, this study employs the concept of interactivity to understand the different interactive
natures of different channels of social media. Further, this study focuses on Facebook and
Twitter because they are the most popular social media (i.e., having the largest numbers of
estimated monthly visitors as of 2014) as well as the most frequently used in political campaigns
(eBizMBA, 2015)1. The current study also compares the effects of different modalities
embedded within social media. This comparison will provide important insight into expanding
our understanding of human sensory modes in the social media context.
Several theories and concepts regarding interactivity (i.e., the theory of user control, the
theory of structural isomorphism, the media richness theory, the social presence theory, cognitive
load, and disorientation) and modality (i.e., the dual coding theory, the cue summation theory,
the limited-capacity information processing theory, and multiple resource theory) were used as
theoretical backgrounds of this study. An online experiment is conducted to investigate what
conditions of modality and interactivity embedded within social media have the strongest effects
on users’ cognitive, emotional, and behavioral responses in political campaign advertisements.
1 In December 2014, eBizMBA, an eBusiness guide that finds the Internet’s best resources, reported the top 15 most popular social networking sites. They ranked Facebook as the most popular social networking site, with 900,000,000 estimated monthly visitors, and Twitter as the second most popular, with 310,000,000 estimated monthly visitors.
6
Eight political advertisements are created, varying in a 2 (social media platform: Facebook,
Twitter) X 2 (interactivity: high, low) X 2 (modality: visual, audiovisual) design.
The findings of this study are significant for several reasons. First, the present study
contributes to understanding how valid traditional theories about communication channels are in
the social media environment. In particular, the study reexamined different conceptualizations of
medium, channel, mode of presentation, and media platform in the social media context and also
reviewed unique characteristics of social media. Second, this study advances social media
research in the political advertisement and persuasive contexts by furthering the understanding of
users’ responses to media channels. An answer about whether different features of
communication channels at different levels of interactivity and modality can be manipulated to
elicit individuals’ cognitive, affective, and behavioral effects is very valuable, especially in the
political campaign context. Third, the findings of this study have practical uses for designing
appropriate strategies for political campaigns through social media platforms to promote
communication between candidates and publics.
7
CHAPTER TWO
LITERATURE REVIEW
This chapter includes four sections. The first section reviews and reexamines
communication channel research including communication channels in political campaigns,
McLuhan’s concept of medium, and characteristics of social media platform including
interactivity and modality. The second section reviews relevant theoretical backgrounds related
to individual’s cognitive, emotional, and behavioral responses to the interactivity and modality of
the channel, especially in the social media context. The third section discusses previous research
regarding the effects of political advertising in different communication channels. The fourth
section presents the research hypotheses and research questions arising from the literature review.
Communication Channel
Communication Channel in Political Campaigns
Trent and Friedenberg (2008) noted that “the sophisticated use of modern technology has
brought significant alterations to political campaign communication” (p. 14). Political candidates
have used the new communication channels that result from technological advancements as tools
in their campaigns and hired specialists who understand the media topography for selecting
target voters. To illustrate, Dwight Eisenhower was the first to use televised advertising in the
1952 presidential campaign, John Kennedy was the first to use his own public opinion polling
specialist in the 1960 presidential campaign, George McGovern was the first to start mass direct
mail fundraising in the 1972 presidential campaign, Jimmy Carter used conference phone calls to
communicate with voters in Iowa and New Hampshire in the 1980 presidential campaign,
8
Ronald Reagan used satellite transmissions at fundraisers and rallies in the 1984 presidential
campaign, Ted Kennedy was the first U.S. senator to have a website in 1993, Dianne Feinstein
created the first candidate website in the 1994 U.S. Senate election in California, Bill Clinton and
Bob Dole delivered their messages to voters through their websites in the 1996 presidential
campaign, and Barack Obama and John McCain used social media such as Facebook and Twitter
to communicate with voters in the 2008 presidential campaign (Kim & Khang, 2014; Trent &
Friedenberg, 2008).
Before the emergence of the Internet technology, political campaigns in the election
context have been accompanied by massive campaign coverage by media and social interactions
through interpersonal campaigning (Cho, 2005). In light of campaign coverage by media, print
media and radio were the dominant media in political campaigns in the first half of the twentieth
century. Both print and radio advertisements provide candidates with the opportunity to present
their messages to targeted voters in the appropriate geographic area without limiting the length of
the messages (Trent & Friedenberg, 2008). With the advent of televised advertising in the 1952
presidential campaign, however, television outlets underestimated the power of print media and
radio as vehicles for political advertisements (2008). Unlike print media and radio, television is
the only medium that utilizes two senses―seeing and hearing―and thus it conveys more
campaign information to larger audiences in a shorter time (2008). However, as political
campaigns on television increasingly focused on candidate images such as personal
characteristics rather than policy issues, scholars were concerned about the decreasing discussion
of issues, while others argued that the technological attributes of television such as vivid images,
soundbites, and the dramatization of personal stories provoke emotional responses from voters,
9
learning about politics, and more positive ratings of a political candidate (Altheide & Snow,
1991; Ansolabehere, Behr, & Iyengar, 1993; Bucy & Newhagen, 1999; Davis & Owen, 1998).
While making use of media, candidates were still engaging in interpersonal campaigning.
The candidate’s interpersonal campaigning was primarily held by arranging small social
gathering such as coffees or the door-to-door canvass (Trent & Friedenberg, 2008). In the casual
or formal formats, candidates and their party got an opportunity to promote interpersonal
communication with neutral and friendly voters in the social gathering setting (2008). Such
events provided candidates with opportunities to not only pay personal attention to their
supporters but also establish a relationship with them (2008). In a similar vein, candidates
engaged in door-to-door canvassing to identify voters who are favorable or unfavorable to them
and to reinforce their positive image (2008).
Discussions about the role of old and new communication technologies in political
campaigns were already under way when the Internet appeared (Trent & Friedenberg, 2008). In
contrast to television, which has the downside of enormous production and advertising costs
compared to other media, the Internet is a cost-effective and high visible communication channel
that enables political candidates and parties to reach large numbers of voters in short periods of
time, convey information about the candidates and the campaign to voters, and elicit voters’
support (2008). The potential of the Internet was highlighted in the 2000 presidential election as
a variety of online services such as online voter registration, online fundraising, and online
volunteer sign-up were developed (2008). Through these online services, candidates and their
teams can communicate directly with the voters by sending messages to voters and getting
feedback from them (2008).
10
Further, as the Internet facilitates two-way communication between candidates and voters,
it has been recognized to have more characteristics of interpersonal campaigning than traditional
media such as the print, radio, and television (2008). Before the advent of the Internet, the
candidate’s interpersonal campaigning was primarily held by arranging small social gathering
such as coffees or the door-to-door canvass (2008). In a form of being casual or formal,
candidates and their party got an opportunity to promote interpersonal communication with
neutral and friendly voters in the social gathering (2008). Such events provided candidates with
opportunities not only to pay personal attention to their supporters but to establish a relationship
with them (2008). In a similar vein, candidates served the door-to-door canvass to identify voters
who are favorable or unfavorable to them and to reinforce their positive image (2008). The
Internet allowed candidates to simulate an interpersonal transaction with potential supporters by
hosting a small personal gathering online to recruit voters, volunteers, and donors (2008). The
Internet offers interpersonal communication not only between the candidate and voters but
between voters. Given that people tend to rely on the opinion of their friends or relatives when
they have few sources of information about the candidate or the campaign, the interpersonal
capacity of the Internet demonstrated the possibility of social capital and civic engagement
(Putnam, 1995).
Since Obama’s successful use of social media as a campaign tool in the 2008 presidential
election, scholars have predicted that social media has the great potential to bring about a radical
change in political communication strategies (Chi & Yang, 2011; Hong & Nadler, 2012; Shogan,
2010). Social media have blurred the boundary between mass communication and interpersonal
communication more than traditional media (Meijer & Thanes, 2013; Shultz, Utz, & Göritz,
2011). As a mass medium, social media play a role as a broadcasting channel like traditional
11
media (Meijer & Thanes, 2013). For example, government agencies or politicians use social
media to broadcast their messages to large audiences. As an interpersonal medium, on the other
hand, social media are better suited for two-way communication between organizations and users
as well as between users than traditional media and websites (Mayfield, 2011; Shultz, Utz, &
Göritz, 2011). In particular, scholars have paid attention to the increasingly personalized and
dialogical campaigning in social media (Enli & Skogerbø, 2013). Social media have the ability to
construct public or semipublic profiles, articulate users’ networks, and communicate with their
online and offline networks based on common interests, unlike earlier communication
technologies (Boyd & Ellison, 2007; Kushin & Kitchener, 2009; Woolley, Limperos, & Oliver,
2010; Zhang, Seltzer, & Bichard, 2013). As social media such as Facebook and Twitter provide
semi-public and semi-private settings for self-representation, the boundaries between public and
private roles online and offline have become blurred (Enli & Skogerbø, 2013; Enli & Thumin,
2012). In fact, politicians use social media to share thoughts and photos from their private lives
to attract voters, connect with the voters, get feedback from voters, discuss political information
more continuously, and engage voters (Enli & Skogerbø, 2013). Given that traditional political
campaigning has always involved personal encounters with voters such as coffees or the door-to-
door canvass, social media are recognized as appropriate to the political campaign setting
because they add a personal angle to the public spheres (2013).
The use of social media that have characteristics of mass and interpersonal media in the
political campaign context still raises a question about whether social media are effective
campaign tools that elicit cognitive (e.g., voters’ learning about politics), emotional (e.g., higher
ratings of a political candidate), and behavioral responses (e.g., voting behaviors), as indicated
by prior research on political campaigns though mass media (i.e., print, radio, and television) and
12
interpersonal communication (i.e., coffees and the door-to-door canvass). In response to this, this
study examines whether and/or how the use of social media in the political campaign context
affect the audience’s cognitive, emotional, and behavioral responses.
Reexamination of McLuhan’s Concept of Medium
Political candidates use different communication channels such as newspapers,
magazines, radio, television, websites, blogs, and social media to provide their messages and
communicate with their interested publics. Drawing on McLuhan’s notion, the media channel
used in a society is still important to deliver messages to the target audience in the digital age
(Sevin, Kimball, & Khalil, 2011).
McLuhan (1964) first popularized the concept of medium as a key ingredient in
interpreting communication meaning with his famous notion that “the medium is the message”
(Kaid, 2003; Sundar, Narayan, Obregon, & Uppal, 1998). This statement indicates that the
medium of communication facilitates the viewer’s interpretation of messages, and any
communicative technologies transmitting messages shape and control the scale and form of
human association and action. Although many scholars have criticized this perspective as being
too simple and ignoring individual differences, McLuhan’s idea is still recognized as the
foundation for the fundamental role of the medium in our cognition and social practices (Nardi &
O’Day, 1999; Williams & Edge, 1996).
According to McLuhan, different characteristics of each medium represent different
content of another medium. To illustrate, “the content of writing is speech, just as the written
work is the content of print, and print is the content of the telegraph” (McLuhan, 2011, p. 19).
Content is the important characteristic that distinguishes between a medium such as radio,
telegraph, telephone, and television and a non-medium such as electric light and power.
13
Further, McLuhan (1964) describes the evolution of media in a thermostatic manner such
as “hot” and “cold” media. His conceptualization of “hot” and “cold” media leads to the
widespread popular belief that each medium has different effects on the audience. He
distinguished a hot medium like a photograph, print media, radio, and movie from a cool one like
the telephone, speech, and television based on the state of information in the medium and users’
participation. With a hot medium, the audience extends a single sense in high definition to reach
“the state of being well filled with data” (McLuhan, 2011, p. 39). Since the hot medium provides
large amounts of information, it does not require high participation by the audience. In contrast, a
cool medium has low definition, which is the state of little information, and the medium has to
be completed by participation of the audience (2011).
McLuhan (2011) also considered effects that the medium engenders. In terms of personal
and social effects of the medium, McLuhan (2011) contended that the evolution of the medium
engendered an extension of our physical bodies including not only our private senses but also
interaction among the senses. For example, radio changed the news story into the form of voice,
while television changed radio programming into the form of an image. McLuhan (2011)
regarded print media as an extension of the sense of sight, radio as of the sense of hearing, and
television images as of the sense of touch, involving “maximal interplay of all the senses” (p.
441). Further, any medium—that is, of any extension of ourselves, or any new technology—
results in personal and social consequences because the content or use of the medium “shapes
and controls the scale and form of human association and action” (McLuhan, 1964, p. 9). For
example, the television image does not bring detailed information about objects based on its low
definition so that the viewer is involved in extremely intimate TV acting and the fascinating role
of the TV star. McLuhan (2011) additionally suggested that hot and cool media result in different
14
effects on group, social, or cultural structure. The hot medium based on high definition brings
about specialism and fragmentation in these structures, while the cool medium based on low
definition serves to restore the pattern of the structures. However, McLuhan (2011) indicated that
the effects depend on the culture where the medium is being used and each medium has different
implications in each country, culture, and society. For example, “the hot radio medium used in
cool or nonliterate cultures has a violent effect, quite unlike its effects, say in England or
America, where radio is felt as entertainment” (p. 48).
Scholars have opposing views applying McLuhan’s concept of hot and cool media to the
digital environment. The computer used to access the Internet is composed of an icon and
windows like television, but the screen is filled with large amounts of information like print
media (Havick, 2000). Some scholars believe that McLuhan’s concept of hot and cool media is
no longer applicable because the computer has the characteristics of both (Logan, 2010; van
Koten, 2009). However, most scholars regard the Internet as a hot medium in that the Internet
does not leave much to the audience’s imagination because of the flood of information it
provides (Havick, 2000; Levinson, 2003). Specifically, Havick (2000) contended that the
Internet in television-dominated cultures will engender specialization, fragmentation,
individualism, decentralization, isolation, and feelings of alienation that threaten the vitality of
government, as McLuhan pointed out that the hot medium used in cool cultures evokes
detrimental effects. Levinson (2003) also claimed that the Internet is more similar to print media
than to telephone or television because Internet users communicate with others using written
words although the personal computer has elements of three media such as books, telephone, and
television with the capability for instant interactivity.
15
The center of the controversy lies in different conceptualizations of the medium itself. No
single consistent definition of the medium exists. Indeed, the medium has been regarded as a
crucial channel, vehicle, or device in the communication process for message transmission
between a source and a receiver in classical models of communication (e.g., Berlo, 1960;
Lasswell, 1948; Shannon & Weaver, 1949). Lasswell (1948) described an act of communication
by asking questions such as “Who says what in which channel to whom with what effect?” (p.
37), defining the medium as channel such as “radio, press, film, and other channels of
communication" (p. 37). Similarly, Shannon and Weaver (1949) suggested a schematic diagram
of a general communication system which is composed of five parts: an information source, a
transmitter, the channel, the receiver, and the destination. In particular, they argued that the
medium is the similar concept of the channel, defining the channel as “merely the medium used
to transmit the signal from transmitter to receiver” including a pair of wires, a coaxial cable, a
band of radio frequencies, and a beam of light (p. 5).
Later, Berlo (1960) suggested that “channel” includes three major meanings such as the
message vehicle, the mode of encoding and decoding messages, and the vehicle-carrier. Berlo
(1960) used the metaphor of getting a package to the other side of a lake to explain the concept
of channel. As a person needs a boat to carry the package, a boat dock to connect with the boat,
and some water. Communication channels contain all three things: message-vehicles (boats),
modes of encoding and decoding messages (boat docks), and vehicle-carriers (water). Further,
Berlo (1960) articulated the message vehicles as media (e.g. radio, telephone, telegraph,
newspapers, films, magazines, direct mail, the stage, or the public platform) and the modes of
encoding and decoding messages as the mechanism of communication that allows the receiver to
decode the messages, including the five senses of seeing, hearing, touching, tasting, and smelling
16
(e.g. oral, physical, or visual communication). In line with McLuhan’s conceptualization of the
medium as a creator of different contents and an extension of our physical bodies, Berlo (1960)
claimed that the message vehicles and the modes of encoding and decoding messages are
interrelated in that one particular message vehicle can use certain kinds of modes. Although
Berlo (1960) did not pay much attention to the vehicle-carriers that the engineer, the physicist,
the biologist, and the chemist mainly concern, he regarded them as supporters of the message
vehicles. Berlo’s (1960) description of the three meanings of the channel suggests that “all
messages must be transmitted through a channel” (p. 68).
Applying the notion of the medium (or the message vehicle) to the online environment,
the desktop computer, mobile phone, and laptop have been regarded as a medium because each
carries different contents such as information and services. Jokinen and Raike (2003) defined the
medium as the physical channel (including computer input/output devices) for presenting or
saving information, while modality (or modes of encoding and decoding messages) is the
particular mechanism of perception, senses of encoding information in some medium such as
vision, audition, olfaction, touch, and taste. The Internet plays a role in as part of the vehicle-
carrier in that it supports the media to transmit messages to the audience. In communicating
online, hence, the source has to choose a medium―the desktop computer, mobile phone, or
laptop―to orally, visually, and/or physically transmit messages to the receiver through the
Internet (Jokinen & Raike, 2003).
However, scholars’ conceptualization of the medium still raises a question about how we
define other Internet services such as websites and social media. In particular, social media are
one of the channels but it is neither the message vehicles (media), the modality, nor the vehicle-
carriers. Social media such as Facebook and Twitter can be used across different media (e.g.,
17
desktop computer, laptop, smart phone, etc.) in various modes (i.e., text, graphic, audio, and/or
visual). Chong (2012) applied an appropriate emerging term, “media platform,” to represent
social media. According to Chong (2012), the concept of a media platform combines a “hard”
platform with “soft” programming in new media so that identical contents can be shifted to
different media platforms. Such interaction between the medium and content evokes a
reexamination of McLuhan’s (1964) concept of medium in the social media environment.
Although McLuhan (1964) noted that a different medium represents different content, a different
medium shifts the same contents in the social media context. In the same vein, according to
McLuhan (1964), the choice of the medium is important in determining the effectiveness of
communication. However, the effects of the social media may be attributed to the interaction
between media and contents based on its characteristics of media platform.
Based upon prior studies of the communication channels (Berlo, 1960; Chong, 2012;
Lasswell, 1948; McLuhan, 1964; Shannon & Weaver, 1949), hence, this study conceptualizes
the communication channel as an umbrella term including medium, modality, vehicle-carriers,
and media platform. Consistent with Chong’s (2012) notion of media platform, the present study
regards social media as media platforms that combine hard platforms (i.e., media: desktop
computer, laptop, smart phone, etc.) with soft programming (i.e., contents used as various
modalities such as text, graphics, audio, and/or visual) through the Internet networks. In terms of
the effects of social media, further, although some scholars argued that social media, which
focuses mainly on “interactivity, access to information and two-way communication,” are
extensions of our senses of sight, hearing, and touch with the increased power and speed of
information (Logan, 2010, p. 40), it has not been fully addressed whether and how social media
result in different users’ responses, considering the unique characteristics of social media.
18
In this sense, this study looks at the cognitive, emotional, and behavioral responses to
different channels, recognizing social media as media platforms. To examine the effects of the
media platform in social media, the study focuses on the interactivity of the medium and
modality (the mode of presentation) of the medium. As social media are designed for social
interaction between users and political campaigns increasingly take advantage of social media
with a variety of forms of communication ranging from text to pictures to audiovisuals, it is
important to examine the effects of the interactivity and the modality via social media in
contemporary political campaign advertisements. As such, literature on the characteristics of
communication channels including interactivity and modality is reviewed.
Interactivity
Scholars have argued that social media have built a new generation of Internet
technologies, the so-called “Web 2.0” (Dadashzadeh, 2010; Meijer & Thaens, 2010, 2013).
O’Reilly (2007) defined the term “Web 2.0” as:
The network as platform, spanning all connected devices; Web 2.0 applications are those that make the most of the intrinsic advantages of that platform: delivering software as a continually-updated service that gets better the more people use it, consuming and remixing data from multiple sources, including individual users, while providing their own data and services in a form that allows remixing by others, creating network effects through an “architecture of participation,” and going beyond the page metaphor of Web 1.0 to deliver rich user experiences (p. 17). Dadashzadeh (2010) argued that “Web 2.0 has transformed the Internet by allowing
previously passive web surfers to become active content creators who want to share” (p. 81).
Compared to traditional websites, in fact, social media such as Facebook and Twitter require
“less technical expertise, time, and financial resources” to design and manage, providing “free
accounts and including some dialogic features by default” such as “message posting, receiving
direct responses to a post through comments and ‘likes,’ text/video chat, e-mail, and sharing
19
multimedia files such as photos and videos” (Kim et al., 2014, pp. 593-594). Meijer and Thanes
(2013) also argued that social media are different from information websites in that information
websites provide information from one to many users, while social media provide information
from many to many. Unlike organization-operated information websites, which are based on
general information, social media rely on user-generated content which is more personalized
rather than social or organizational (Meijer & Thanes, 2013).
Many scholars explain that the characteristics of social media that facilitate user’s active
participation and reciprocal, direct, and two-way communication between users as the concept of
interactivity. As an interactive medium that provides non-linear or two-way communication,
social media attract some users to express and post their opinion and to comment on messages
(Lee & Cho, 2011; Meijer & Thanes, 2013; Rowley, 2004; Shultz, Utz, & Göritz, 2011). Lee and
Cho (2011) suggested three phases of interactivity on Twitter and Facebook; 1) facilitating two-
way communication among users, 2) giving more opportunity to talk with other users, and 3)
being effective in gathering feedback from others. Kietzmann et al. (2011) stated that social
media use “mobile and web-based technologies to create highly interactive platforms via which
individuals and communities share, co-create, discuss, and modify user-generated content” (p.
241), while Bertot and his colleagues declared that social media are designed for social
interaction (Bertot, Jaeger, & Grimes, 2012; Bertot, Jaeger, & Hansen, 2012). Meijer and Thanes
(2013) also claimed that social media allow users to engage in social interactions and to obtain
information based on the variety of interactions.
Kim, Chun, Kwak, and Nam (2014) suggested that social media expanded “the
interactive potentials of the Internet” (p. 593). Social media, as a simple personal Web service,
are “where individuals can post their personal information on profiles, ideas, and activities in a
20
variety of formats, such as text, audio, and video, and share the information within their
individual networks” (p. 593). Schultz, Utz and Göritz (2011) see social media as “more
interactive, dialogic, authentic, and credible” than traditional media or simple websites because
of their capacity of two-way communication (p. 22). Lee (2013) also pointed out that
interactivity is a key characteristic to differentiate television from social media such as Twitter.
According to Lee (2013), although both television and Twitter play a role in a candidate’s
personality-driven and soft campaign media in the political campaign context, unlike print media,
Twitter is a more interactive medium between a candidate and the audience than television
because candidate and audience can follow each other and engage in reciprocal message
exchange. On the other hand, television has a higher level of vividness than Twitter because it
presents more expressive images without text-chat modes (Lee, 2013).
Likewise, although many studies have paid attention to the interactive characteristics of
social media, “the standard for what makes one medium more interactive than another” is still
vaguely defined (Kiousis, 2002, p. 356). Although interactivity has been recognized as the
integral advantage of the Internet in computer-mediated communications, including the active
participation of users and social interactions among users, the concept of interactivity has been
defined differently from different perspectives (Ha & James, 1998; Morris & Ogan, 1996;
Pavlik, 1996; Rafaeli & Sudweeks, 1997).
Specifically, a mechanical perspective (or user-machine interaction) focuses on the time
and duration of interaction among users, defining interactivity as “the extent to which users can
participate in modifying the form and content of a mediated environment in real time” (Steuer,
1992, p. 84). However, this perspective alone does reflect the Internet environment that requires
user's interaction (Liu & Shrum, 2002). From an interpersonal communication perspective (or
21
user-user interaction), interactivity occurs in the context of the relationship between users
(similar to face-to-face communication) and is defined as “the extent to which messages in a
sequence relate to each other, and especially the extent to which later messages recount the
relatedness of earlier messages” (Rafaeli & Sudweeks, 1997, p. 3). Focusing on reciprocal and
two-way communication between the communicator and the audience, interactivity can also be
defined as “the degree to which participants in a communication process can exchange roles and
have control over their mutual discourse” (Rogers, 1995, p. 314) or “a condition of
communication in which simultaneous and continuous exchange occur, and these exchanges
carry a social, binding force” (Rafaeli & Sudweeks, 1997, p. 4). However, these definitions of
interactivity are limited because the definition from the interpersonal communication perspective
ignores the characteristics of computer-mediated communication, and the definition from the
mechanical perspective fails to consider Internet users’ individual differences in their
communication goals online (Ha & James, 1998).
McMillan (2002) noted that literature on conceptualizations and operationalizations for
interactivity can be classified into three main categories: structure, users, and process. According
to McMillan, early work on interactivity focused on the interactive structure of a communication
channel. With the advent of the Internet, however, researchers paid attention to the
characteristics of users, especially how interactive computer-mediated communication affects the
user’s cognition and emotion. Some scholars including McMillan further regard interactivity as
“a process-related construct about communication” that is associated with the concept of
cognition, motivation, emotion, and sociability (Rafaeli & Sudweeks, 1997, p. 175). From this
perspective, interactivity is not a condition but a continuum that acts as a bridge between the
psychological and the sociological, between mass and interpersonal communication, between
22
direct and mediated communication, between one-way (e.g., radio and television) and two-way
communication, and the text-based versus verbal-based communication (Rafaeli & Sudweeks,
1997).
Sundar, Kalyanaraman, and Brown (2003) conceptualized interactivity from two points
of view: functional and contingency. The functional view regards interactivity as “an interface’s
capacity for conducting a dialogue or information exchange between users and the interface”
including “e-mail links, feedback forms, chat rooms, and audio or video downloads” (Sundar et
al., 2003, p. 33). In the contingency view, interactivity refers to “a process involving users, media,
and messages, with an emphasis on how messages relate to one another” including the levels of
two-way/noninteractive, reactive, and responsive/interactive communication (2003, pp. 33-34).
Both the functional and contingency views suggest that interactivity promotes the user’s
participation in information content (Kim & Sundar, 2010).
Wu (2005) provided the concepts of actual interactivity and perceived interactivity. The
actual interactivity consists of “the features of a medium, or capabilities of creating interactive
content or messages, or potential for interaction in general,” while the perceived interactivity
refers to “user’s psychological state experienced by a site-visitor during the interaction process”
(p. 30).
Instead of classifying the concept of interactivity from a different view, some scholars
defined the concept based on the characteristics of interactivity. Ha and James (1998) suggested
five dimensions of interactivity to satisfy different communication needs: playfulness, choice,
connectedness, information collection, and reciprocal communication. Playfulness is related to
entertainment experience through Internet use, while choice is seen as unrestrained navigation
online. Internet users also feel connectedness by interacting with web contents based on
23
hyperlinks and images. Information collection refers to the communicator’s data gathering about
the audience to tailor messages to meet audience needs, while reciprocal communication is
related to a reciprocal communication loop to encourage feedback between the communicator
and the audience. Ha and James (1998) further noted that information collection and reciprocal
communication, as higher levels of interactivity, involve direct and two-way communication
between the communicator and the audience.
Liu and Shrum (2002) classified three dimensions of interactivity: active control, two-
way communication, and synchronicity. Active control refers to voluntary and instrumental
actions of users. Two-way communication includes not only “reciprocal communication between
companies and users and users and users” but also “the ability to make transactions directly
online,” while synchronicity is concerned with “the degree to which users’ input into a
communication and the response they receive from the communication are simultaneous” (p. 55).
Incorporating all three dimensions of interactivity, Liu and Shrum defined the concept of
interactivity as “the degree to which two or more communication parties can act on each other,
on the communication medium, and on the messages and the degree to which such influences are
synchronized” (p. 54). Jiang, Chan, Tan, and Chua (2010) also regarded the 'user’s active control
over the medium' as interactivity because this control represents “the ability to choose
information and guide an interaction” (p. 37).
Likewise, although no single standard for interactivity exists, it can be assumed that the
concept of interactivity is a multidimensional construct including the structure of a
communication channel, users, and communication process between the structure and users. A
communication channel includes a variety of interactive structures that facilitate active control,
two-way, reciprocal, direct communication between users. As Rafaeli (1988) noted that
24
“interactivity is potential adequacy, but it is up to the communicators to realize it” (p. 117), users
tend to perceive interactivity based on the communication channel that provides interactive
features and different communication needs. Thus, this study defines the concept of interactivity
as an interactive communication process between the structure of a communication channel and
user’s perception.
With regard to interactivity in social media, Lim, Hwang, Kim, and Biocca (2015)
suggested three dimensions of social media engagement: functional, emotional, and communal
engagement. First, functional engagement is relevant to “the online users’ real-time participation
in the social media platform to create, modify, and share the format and content” (p. 159). As the
concept of interactivity emphasizes relational phases between the organization and the public
beyond mere interface-based interactivity, functional engagement in the social media
environment refers to “social media user’s interactions with other users in the process of co-
creating, conversing and sharing the content” through interfaces such as hashtags, retweets, and
replies that encourage users’ engagement in conversation (Lim et al., 2015, p. 159). Second,
emotional engagement occurs when users feel emotions such as joy, excitement, amity,
frustration, disappointment, or animosity toward messages in the form of posts, opinions, or
comments on social media. Third, communal engagement is “online users’ involvement with
brand, cause, and programs of an organization and their feelings of belonging and social
connection” (Lim et al., 2015, p. 160). Interactive interfaces of social media such as hashtags
allow users to feel like part of a community and to be involved in interaction among users.
Although most social media platforms provide similar ways of communicating in
personalized ways with their interactive potential, the interactive nature of each communication
channel is distinct (Hughes et al., 2011). For example, direct and synchronized communication is
25
more highlighted on Twitter than on Facebook, while two-way, interpersonal, and reciprocal
social interaction among “friends” is more highlighted on Facebook than on Twitter. Davenport
et al. (2014) noted that Facebook, “the prototypical SNS,” includes a variety of services such as
“friend requests, tagging others, posting comments, posting pictures, and creating status updates”
that facilitate reciprocal social interaction between users and their friends (pp. 212-212).
Specifically, Facebook offers a profile on which users post information, web links, pictures, and
videos and a real-time instant messenger to send private and public messages to their “friends”
(Hughes et al., 2011). The size of the user’s friend network is “somewhat under the control of a
user” because it depends on whether the user accepts or denies friend requests from other users
(Davenport et al., 2014, p. 213). Facebook distributes users’ messages to their friends, who have
the chance to comment on the post (Murthy, 2011). In this way, users use Facebook as a
relational communication tool to create and maintain social capital based on their known social
connections such as friends, family, and co-workers (Bevan, Pfyl, & Barclay, 2012; Quercia et
al., 2011). Unlike Facebook, which is based on reciprocal social interaction among “friends,”
Twitter focuses on the sharing of opinion and information rather than on social pressure because
it is not required for users to make a profile or to send and accept friend requests in order to
communicate with each other (Huberman et al., 2009). Further, since Twitter allows users to
communicate short messages through “tweets” with a maximum length of 140 characters, it is
regarded as being designed for direct and speedy communication between users (Davenport et
al., 2014; Murthy, 2011; Shultz, Utz, & Göritz, 2011). For this reason, some scholars argue that
Twitter provides an opportunity to publicize information and therefore plays a role as a mass
communication channel (Kim et al., 2014).
26
Taken together, each social media platform has distinct interactive features that may
differently engender users’ cognitive, emotional, and behavioral responses. Hence, this study
looks at how certain interactive aspects of the social media platform as a tool in political
campaign affect users’ cognitive, emotional, and behavioral responses. Although the prior
literature suggests that not only the interactive nature of a communication channel but
individual's perceived interactivity engender the cognitive, emotional, and behavioral effects of
interactivity, this study focuses on the effects of interactive as the term referring to ‘an
interactive communication process between the structure of a communication channel and user’s
perception’ as noted earlier.
Modality
Social media are media platforms consisting of hard platforms such as message vehicles
with soft programming of contents in different modes. Politicians use social media to present
their messages orally, visually, and/or physically to potential voters. To compare the effects of
different modes of communication channels used in political campaign advertisements in social
media context, the present study focuses on modality.
The concept of modality refers to “the mode of presentation such as text, audio, picture,
video that corresponds to human senses used for processing the presented material” (Sundar et
al., 1998, p. 823). According to Conway (1967), the difference between the print channel and the
audio channel lies in whether it involves the visual or auditory modality. Unnava, Burnkrant and
Erevelles (1994) argued that people tend to encode modality-specific content such as text, audio,
and picture while processing information. In particular, modality has been recognized as a
perceptual variable related to particular associated memory retrieval processes (Russell, 2002;
Unnava, Agarwal, & Haugtvedt, 1996). Dijkstra et al. (2005) claimed that each communication
27
channel has not only different content but also different numbers of sensory modes. For example,
print has only text and visuals, television has moving visuals and audio, and Internet contains
digitalized textual information of print and/or audiovisuals of television (Dijkstra et al., 2005;
Snyder, 1996). Furthermore, although print media and the Internet are regarded as retrieval
channels, their mechanisms of information transfer to the audience are different (Sundar et al.,
1998). Sundar et al. (1998) noted that readers process print ads entirely, but they process Internet
ads through scrolling, and thus readers had a better memory for print ads than Internet ads.
Scholars also indicated that print media such as newspapers provide a variety of visual cues in
accordance with the relevance of topics, while Internet sources such as online newspapers offer a
smaller number of cues (Bucher & Schumacher, 2006; Schönbach, de Waal, & Lauf, 2005). This
is because online newspaper users are more likely to select news stories consistent with their
own interests and less likely to “follow the cues of news editors and producers” (Tewksbury,
2003, p. 694). The sensory modes influence the direct processing of cognitive and affective
reactions, or the indirect processing of reactions through other sensory modes (Dijkstra et al.,
2005).
Research on modality has explored how receivers processed information differently when
they were exposed to stimuli with different modes of presentation and how receivers change their
behaviors according to an interaction of more than two stimuli with different modalities (Sundar
et al., 1998). However, past studies on modality have provided inconsistent results. Some
scholars found that a larger number of sensory modes results in more effective processing of
information, while others have the opposite view because of the selective information processing
that occurs in accordance with the limited human perceptual system (Dijkstra et al., 2005).
Specifically, some researchers have demonstrated that a larger number of sensory modes (i.e.,
28
more visual elements, more pictorial elements, or multiple sensory modes) facilitate cognitive
responses such as learning, memory (recall), elaboration, support, attention, and interpretation of
meaning (Bagui, 1998; Craik & Lockhart, 1972; Deutsch & Deutsch, 1963; Jones &
Kalyanaraman, 2000; Kiousis, 2006; Kisielius & Sternthal, 1984; Paivio & Okovita, 1971;
Triesman, 1964). On the other hand, a larger number of sensory modes of the medium can also
interfere with cognitive elaboration and critical thinking because it requires more effort to
process information (Edell & Keller, 1989).
While numerous researchers have spent a great deal of time in studying the cognitive
processing of a communication channel with different sensory modes, past studies paid little
attention to emotional processing. However, some scholars suggest that difference in the
modality of the channel engenders individuals’ different emotional reactions. For example,
McClure and Patterson (1976) considered television news to be fascinating and highly
entertaining but not informative. Brader (2006) noted the general sentiment of political
practitioners that the audiovisual characteristics of television are associated with emotional
reactions. Daignault, Soroka, and Giasson (2013) also argued that the goal of an emotional cue
from audiovisual properties is to increase the effectiveness of advertising. Prior studies also
confirmed the interaction of source and media factors, showing that source characteristics create
different persuasive effects across media. Some findings demonstrated that a medium with
audiovisual modalities, such as television, is more persuasive for a likable or trustworthy source
than for an unlikable source, while print medium is more persuasive for an unlikable source
(Andreoli & Worchel, 1978).
Taken together, prior studies suggest that different modality of the communication
channel results in different cognitive, emotional, and persuasive effects, focusing mainly on
29
differences between print media and television. However, it is still unknown whether and how
different modality such as visuals and audiovisuals differently affect users’ cognitive, emotional,
and behavioral responses to political advertisement in the social media context. Although social
media transmit the visual and audiovisual ads that are used in newspapers and television, the
interactive natures of social media platforms containing the visual and audiovisual ads may
affect individual’s cognitive, emotional, and behavioral responses. To fill this gap, this study
reviews theories related to individual’s cognitive, emotional, and behavioral responses to
modality and interactivity of the channel.
Cognition, Emotion, and Behavior
Cognitive, Emotional, and Behavioral Effects of Interactivity
Prior findings suggested that the interactivity engenders users’ cognitive, emotional,
behavioral responses. In the political context, evidence demonstrated that interactivity stimulate
information-seeking behaviors, positive attitude toward the advertisement and the political
campaign website, issue learning, candidate responsiveness, positive candidate perceptions,
voting intention, and political engagement (Auter & Davis, 1991; Althaus & Tewksbury, 2000;
Corrado & Firestone, 1996; Cutbirth & Coombs, 1997; Jacque & Ratzan, 1997; Kaid, 2002,
2003; Lee & Shin, 2012; McKinney & Gaddie, 2000; Milbank, 1999; Sundar et al., 1998;
Sundar, Kalyanaraman, & Brown, 2003; Sundar & Kim, 2005; Warnick et al., 2005).
In light of the effects of the interactivity of a communication channel on the user’s
information processing, some scholars have argued that a more interactive channel requires
higher cognitive and emotional processing of information. Liu and Shrum (2002) noted that the
interactive communication channel demands higher cognitive involvement such as attention and
cognitive processing due to its qualities of active control and two-way communication, which
30
further leads to better user understanding and learning. Sicilia, Ruiz and Munuera (2005) also
argued that an interactive channel results in users’ high control and high involvement that lead to
more intensive information processing because users can select and organize the information
they want while using the interactive channel. Several theories explain why the interactivity of
the channel affects individuals’ cognition, emotion, and behaviors.
Since no single conceptualization of interactivity exists, the theories focused on certain
interactive features such as user control, hypertexuality, real-time interaction, and direct, two-
way, interpersonal, and reciprocal relationships. Specifically, the theory of user control, the
theory of structural isomorphism, the media richness theory and the social presence theory
suggest that the channel with certain interactive features engenders more cognitive, emotional,
and potential behavioral responses, while cognitive load and disorientation lead to less
information learning.
Theory of user control. Compared to information websites, social media such as
Facebook and Twitter provide an environment that facilitates the user’s voluntary and active
participation based on Web 2.0 applications. The theory of user control suggests that the
communication channel that provides more user control increases users’ motivation to learn.
Specifically, when user controllability is high, users are able to engage more with the content and
have an increased interest in learning because they tend to keep away from irrelevant, too
complex, and too simple information (Eveland & Dunwoody, 2001). This theory assumes that
“since an individual’s knowledge structure is unique, based upon his or her own set of
experiences and abilities, the ways that individuals prefer to access, interact with, and interrelate
information is also distinct” (Jonassen, 1988, p. 14). The theory is based on two assumptions: (1)
Individuals learn in idiosyncratic ways, and to force learning to take place in a single style will
31
reduce the learning of some individuals to less than their true potential; and (2) Providing control
over the learning process can increase the motivation of the learner (Eveland & Dunwoody,
2001, p. 54). User control ranges from complete user control to complete program control
(Eveland & Dunwoody, 2001; Milheim & Martin, 1991). Hypermedia systems provide a more
user-friendly learning environment because users can learn information by navigating freely
through the various links embedded in a text based on their “personal interests, abilities,
background knowledge, and learning style” (Eveland & Dunwoody, 2002, p. 37).
However, empirical studies about the influence of user control provide inconsistent
findings. For example, some studies found that more user control is better for learning than
program control, whereas others demonstrated opposite findings (Eveland & Dunwoody, 2002).
In a similar vein, in contrast to findings that the user’s selective scanning of content encouraged
by user control increases learning, some researchers found that selective scanning reduces
learning because the user’s ability to monitor his or her own learning is limited (Eveland &
Dunwoody, 2002). Some users often deselect the information that is most necessary for
understanding the content (Eveland & Dunwoody, 2002).
Theory of structural isomorphism. Social media has been regarded as interactive media
based on its hypertextual structures that provide information from many to many and two-way
communication between users (Lee & Cho, 2011; Liu & Shrum, 2002; Meijer & Thanes, 2013;
Rowley, 2004; Shultz, Utz, & Göritz, 2011). The theory of structural isomorphism in psychology
posits that the process of human learning creates and maintains meaningful links among concepts
(Eveland & Dunwoody, 2001). This definition of the process of learning is consistent with the
process of using hypermedia such as the Internet in that both human memory and hypermedia
consist of nodes of information linked to related concepts (Eveland & Dunwoody, 2001).
32
Specifically, hypermedia has the benefits of “mirroring the organization of human memory” and
“facilitating the processing of information through associative links as does human memory”
(Eveland et al., 2004, p. 213). Since the Internet is designed to mimic a human being’s
knowledge structure based on hyperlinks, users can learn more from the Internet than from
traditional print media (Eveland & Dunwoody, 2001). For example, hyperlinks facilitate learning
by indicating how the information is related to each other (Eveland & Dunwoody, 2002).
Eveland et al. (2004) found that being exposed to nonlinear Web presentations engenders less
factual learning of information, but more dense learning (e.g., learning the interconnectedness of
the information) than being exposed to traditional, print-like, linear Web presentations. In
relation to the blogosphere, the findings of Vraga et al. (2011) supported the theory of structural
isomorphism, demonstrating that unfamiliar story format reduces the user’s motivation to learn
the information.
Cognitive load and disorientation. By contrast, some scholars have argued that the
hypermedia structure and more user control reduce learning because it increases cognitive load
and disorientation in comparison with less interactive counterparts. Cognitive load can be
conceptualized as “the amount of mental effort required to locate specific information and
understand how this information is oriented within a larger information source” (Eveland &
Dunwoody, 2001, p. 56). The channel with a high degree of user control requires more cognitive
load from users than the channel with a low degree of user control. Disorientation can be
described as a “user’s feelings of being lost” caused by “inexperience with the medium or
content and site design” (Eveland & Dunwoody, 2001, p. 56). Users navigate unfamiliar spaces
using the medium, and the resulting feelings of disorientation increase cognitive load. In this
33
sense, the hypermedia structure and amount of user control in the Internet require more cognitive
demand than print media, which impedes information learning (Eveland & Dunwoody, 2001).
Media richness theory. Media richness theory (or information richness theory) holds
that “communication media vary in their richness” (Lu et al., 2014, p. 14), or the capacity to
convey messages and cues and enable users to communicate effectively (Daft & Lengel, 1984;
Ledford, 2012). Based upon the work of Bodensteiner (1970) and Lengel (1983), Lengel and
Daft (1984) first proposed the concept of information richness “to explain how organizations
meet the need for information amount and to reduce equivocality” (p. 5). Lengel and Daft argued
that communication medium varies by (1) feedback capability, (2) communication channels
utilized, (3) source, and (4) language.
Dennis and Kinney (1998) suggested two types of feedback: concurrent and sequential.
Concurrent feedback provides simultaneous interaction between the sender and the receiver,
while sequential feedback has pauses in interaction. A rich medium offers quicker feedback
between the sender and the receiver for interaction (Ledford, 2012).
The concept of communication channels utilized refers to the number of cues provided by
a medium. A rich medium has multiple cues representing “the number of ways in which
information can be communicated, such as text (e.g., the spoken or written words themselves),
verbal cues (e.g., tone of voice), or nonverbal cues (e.g., physical gestures)” (Dennis & Kinney,
1998, p. 333). Dennis and Kinney suggested four ways that multiple cues affect information
processing: (1) Verbal and nonverbal cues provide faster and more accurate information
processing than text cues. (2) Computer-mediated communication is less effective to transmit a
message than verbal or face-to-face interaction because typing is a longer process than speaking
the message. (3) Computer-mediated communication features immediate sequential feedback,
34
while media with verbal and nonverbal cues facilitate concurrent feedback. (4) The lack of
verbal and nonverbal cues reduces social presence, leading to a more self-centered perception,
more antisocial behavior, and more fact-based decision making (or less personality-based
decision making). In terms of the first and second elements of media richness, Simon and Peppas
(2004) explained that rich media provide not only greater feedback but “more social, non-verbal
and complex cues (e.g., gestures, vocal inflection, touch, stance)” (p. 272).
Source is related to “the infusion of personal feelings and emotions, as well as the
tailoring of messages to the needs and current situation of the receiver” (Lee, Kozar, & Larsen,
2009, p. 453). Ledford (2012) stated that a rich medium is tailored to “the receiver’s frame of
reference and needs” to communicate (p. 177). Language variety is the final factor. A rich
medium uses a wide variety of language and natural language to transmit a variety of meanings.
A medium with more attributes among these four aspects is considered a rich medium,
while a medium with fewer of the attributes is considered a lean medium (Sheer, 2011). Daft and
Lengel (1986) argued that a rich medium enables users to communicate more quickly and
understand ambiguous messages, whereas a lean medium is more appropriate for unambiguous
communications. They considered face-to-face communication including both verbal and non-
verbal cues with greater feedback to be the richest medium and email to be a relatively lean
medium. However, empirical findings about perceptions of richness and leanness in various
media have been inconsistent, especially in the digital age. For example, Internet media which
feature more modalities of text, picture, audio, and video are considered to be richer media than
print newspapers because of their textual modality (Lu et al., 2014). In this sense, it can be
inferred that television, with its modalities of audio and video, may be classified as a richer
medium than print, but a leaner medium than the Internet.
35
Evidence has shown that the extent of richness of media is associated with the receiver’s
emotional and behavioral effects. For example, Sundar (2000) found that people exposed to
online news with text and picture evaluated the news more favorably than those exposed to
online news with text only. Coyle and Thorson (2001) explained that increased vividness
including rich media tools such as audio, video, and animation results in more positive and
enduring attitudes toward websites. Simon and Peppas (2004) also found positive effects of rich
media on consumers’ evaluations toward products, and Lu et al. (2014) found that people who
viewed Websites with higher media richness had higher behavioral intentions than those who
viewed Websites with lower media richness.
Social presence theory. Originating from Goffman’s (1963) concept of co-presence,
which is defined as a sense of being together in a space, social presence theory has been widely
used to explain an individual’s perception and behaviors in computer-mediated communications
(Park & Lee, 2013). The social presence theory posits that people are social beings who want to
“perceive other people to be physically present when they interact with others” (Lee, Kozar, &
Larsen, 2009, p. 453). Social presence is “the degree to which a medium is perceived as
conveying the actual physical presence of the communicating participants” (Rice, 1992, p. 476).
Each medium has a different capacity to “convey differing degrees of social cues that lead
interacts to feel more or less psychologically present with one another” (Lyons, Reysen, & Pierce,
2012, p. 182). Short, Williams and Christie (1976) argued that some media are more appropriate
for personal interaction because different technologies provide different presence such as
salience or a state of “being there.”
The degree of social presence of a medium depends on the user’s “personal connection
with the other users” (Kuma & Benbasat, 2002, p. 9). Thus, a more interpersonal medium
36
increases social presence (Kruikemeier et al., 2013). A higher presence medium is more personal,
more sociable, more sensitive, and warmer than a lower presence medium (Kuma & Benbasat,
2002; Park & Lee, 2013). In this regard, a higher presence medium allows people to “be present
in communication,” “be personal in their communication,” and “engage more frequently in
socio-emotional communication,” while a lower presence medium results in a reverse effect on
communication performance (Lee, Kozar, & Larsen, 2009, p. 453). Although the definition of
social presence is somewhat difficult to apply to media platforms, Park and Lee (2013) noted that
different media platforms can be classified based on the extent of social presence; thus, blogs and
Wikipedia are regarded as relatively lower presence media than social networking sites.
Lim et al. (2015) suggested that increased interaction with social media results in greater
feelings of social presence. Specifically, they found that active participation in a channel of
social media yields a user’s collective consciousness with friends. Kruikemeier et al. (2013)
found that online political communication, which is similar to interpersonal communication due
to its two-way and personalized interaction, yields a strong perception of social presence of the
politician. Lee and Jang (2013) noted that social presence facilitates parasocial interaction with
the target which results in positive thoughts about the target. Other findings further demonstrated
that greater feelings of social presence increase perceived learning, perceived trustworthiness,
satisfaction, favorable attitudes toward the agent, and purchase intention (Cyr et al., 2007; Gefen
& Straub, 2004; Gunawardena & Zittle, 1997; Kaplan & Haenlein, 2010; Kelleher & Miller,
2006; Richardson & Swan, 2003; Skalski & Tamborini, 2007).
Cognitive, Emotional, and Behavioral Effects of Modality
Research on modality has frequently made use of the information processing approach.
Information processing theory explains how individuals deal with all sensory information (e.g.,
37
symbols, images, and sounds) including “information selection, attention, encoding, schema
activation, information retrieval, and information storage in memory” (Wicks, 2006, p. 85).
Scholars have investigated individuals’ information processing to explain how and to what extent
features of media capture individuals’ attention to and processing of media messages (Lang,
Borse, Wise, & David, 2002). The Internet incorporates multimedia functions such as “speech,
music, text, graphics, still, animation and video, used in an integrated manner” and generates a
more complex processing of information than traditional media (Eveland & Dunwoody, 1999;
Hoogeveen, 1997, p. 151; Jo & Kim, 2003). Social media also convey contents in different
modes.
Among theories of information processing, Kiousis (2006) noted that dual coding and cue
summation theories explain the positive effects of increased modality on audience cognitive and
emotional reactions by stimulating multiple senses in response to multiple cues. This is in
contrast to the limited-capacity information processing and multiple resource theories, which
explain that increased modality elicits difficulty in information processing by generating greater
cognitive overload. Specific information about how each theory is relevant to explain the
cognitive and emotional effects of modality of the medium are described below.
Dual coding theory. Paivio’s (1990) dual coding theory explains multimedia information
processing. The dual coding theory proposes two ways of information processing: verbal
processing of auditory information and non-verbal processing of imagery, visual, and spatial
information. According to the dual coding theory, audiovisual information through television is
stored in memory in one verbal code and one visual code, separately, while text information
through newspapers is stored in one verbal code only (Walma van der Molen & van der Voort,
2000). The theory posits that multimedia representations with both visual and auditory
38
information have more powerful effects on learning and recall than a single representation with
verbal information (Mayer, Heiser, & Lonn, 2001; Yadav et al., 2011). Gunter, Furnham and
Griffiths (2000) explained this is because the multimedia representations provide retrieval cues
during recall which mediate learning. However, dual coding of information increases recall only
when text information corresponds to visual information (Walma van der Molen & van der Voort,
2000). When verbal information does not convey the same meaning as visual information, it
draws attention away from the information due to the individual’s limited attentional capacity
(Walma van der Molen & van der Voort, 2000). Relying upon this reasoning, the theory further
claims that people remember more from television than from the same stories presented in
newspapers because the television presentation adds relevant visual to verbal information
(Walma van der Molen & van der Voort, 2000).
In terms of emotional processing of modality, the theory notes that non-verbal systems
elicit affective and emotional reactions so that visual information with pictures and video lead to
stronger emotional reactions than written words (Clark & Paivio, 1991; Sadoski, Goetz, &
Rodriguez, 2000). In particular, video of emotional interactions led to higher emotional
engagement in the object such as ratings of personal, warm, sensitive, and emotional (Koehler et
al., 2005). However, this emotional engagement is aroused only for human-interest narrative, not
for artistic or informational content (Koehler et al., 2005). Yadav and his colleagues (2011) also
found that video or video with text elicits more powerful emotional feelings towards the focal
person represented in the video than text only and that video promotes participants’ pro-social
behaviors because it is more realistic than text. However, they did not find a difference in
cognitive processing (e.g., recall) between video and text.
39
Cue summation theory. In reviewing past research on learning and retention, Severin
(1967) found the cue summation principle, which predicts that learning would be enhanced by an
increased number of cues or stimuli available in the communication. However, Severin also
noted that learning is increased only if the cues from different modalities are relevant to one
another. Miller (1957) supported this notion:
When cues from different modalities (or different cues within the same modality) are used simultaneously, they may either facilitate or interfere with each other. When the cues elicit the same responses simultaneously, or different responses in the proper succession, they should summate to yield increased effectiveness. When the cues elicit incompatible responses, they should produce conflict and interference. (p. 78) Severin (1967) further suggested four predictions based on the cue summation theory to
compare the effects of multichannel communications with those of single-channel
communications (p. 243): (1) Multichannel communications which combine words with related
or relevant illustrations will provide the greatest gain because of the summation of cues between
the channels; (2) Multichannel communications which combine words in two channels (words
aurally and visually in print) will not result in significantly greater gain than single-channel
communications since the added channel does not provide additional cues; (3) Multichannel
communications which contain unrelated cues in two channels will cause interference between
channels and result in less information gain than if one channel were presented alone; (4) Single-
channel communications will be superior to condition (3) (above), equal to condition (2), and
inferior to condition (1). The efficiency of single-channel communications when comparing
various channels will depend upon the complexity of the material presented for a given audience,
as already discussed.
Likewise, Severin (1967) argued that multiple channel communications are more
effective than single channel communications under the condition that relevant cues are
40
combined across different channels. Relying upon the cue summation theory, prior findings
demonstrated that pictures with audio cues produced greater learning than audio only (Day &
Beach, 1950; Hoban & Van Ormer, 1951). Findings also indicated that audio with visual cues
such as television are more effective for learning and recognition than audio only cues such as
radio or visual only cues such as pictures (Severin, 1967; Westley & Barrow, 1959). More
recently, Brashears, Fraze, Lawver and Baker (2005) found that multichannel communications
(e.g., text with audio or video or multiple cues) increased user’s knowledge, information
retention, and satisfaction more than single-channel communication (e.g., text only or a single
cue), but the knowledge decreased at the point of over-abundance of cues.
Limited capacity information-processing model. Drawing upon the information-
processing tradition of cognitive psychology (Lachman, Lachman, & Butterfield, 1979) and the
literature on mass communication effects (Berger & Chafee, 1989), Lang (2000) suggested a
limited capacity information-processing model of mediated message processing (LC3MP). Lang
et al. (2002) argued that the limited capacity information processing model is necessary to
examine “how viewers process computer presented information” and “how the computer as a
medium may be processed differently from other electronic media” (p. 243). This model explains
how people process messages and how messages cause intended and unintended effects. Lang
(2000) offered two main assumptions in this model. The first is that people are information
processors, and the second is that a person’s ability to process information is limited. These
assumptions include the idea that people tend to be involved in information processing by
perceiving stimuli, performing mental representation of stimuli, and reactivating mental
representations. This information processing of messages requires mental resources, but the
41
mental resources people have are limited. Hence, people may allocate fewer resources to process
a message than it requires, and the message will not be thoroughly processed.
Lang (2000) noted the application of the limited capacity information-processing model
to television viewing, in which the content, the medium, and the viewer all affect the information
processing of the message. As a psychological stimulus, television contains not only content
(e.g., news, sports, and violence) but also audio and visual structural information (e.g.,
luminance levels, cuts, slow motion, animation, zooms, pans, video graphics, frequency levels,
sound effects, music, rate of presentation, and narrative structure). The television viewer is an
information processor. The information processing of the message contains three subprocesses
(encoding, storage, and retrieval) as a result of automatic (unintentional) and controlled
(intentional) processes.
In a television viewing situation, Lang (2000) argued that recall increases depending on
the viewer’s goals and needs to remember the messages or on the arousing or emotional content
of the message. Specifically, because the television viewer has limited capacity to encode and
store all messages provided by television, the viewer is more likely to control “the allocation of
processing resources by making decisions about whether to watch, how carefully to watch, and
how hard to concentrate during viewing” (Grabe, Lang, & Zhao, 2003, p. 390). Based on prior
findings about recall through television viewing (Davidson et al., 1979; Dimmond, Farrington, &
Johnson, 1976; Gantz, 1978; Katz, Adoni, & Parness, 1977; Lang, Dhillon, & Dong, 1995; Lang,
1990; Lang & Friestad, 1993; Lang, Newhagen, & Reeves, 1996; Reeves et al., 1988; Thorson &
Friestad, 1989), Lang (2000) pointed out that an individual’s need to watch television news for
informational purposes increases recall more than for entertainment purposes. Further, Lang
42
(2000) argued that emotion-eliciting messages on television result in the viewer’s better recall
than less emotional counterparts.
Regarding structural features of computers, Lang et al. (2002) found that warnings and
animated banner advertisements elicit orienting responses in viewers, while text-based and
simple computer stimuli (e.g., plain-text, boxed text, and nonanimated banner advertisements) do
not. Therefore, they suggested that a computer is different from other electronic media in terms
of its interaction with the automatic attention system to elicit orienting responses.
Later, Lang (2006) presented the limited capacity model of motivated mediated message
processing (LC4MP) to create effective communication messages. This model explores how
messages are processed according to different media, contents, and goals. Based on the two main
assumptions in the LC3MP, Lang suggested five assumptions in the LC4MP: the nature of
cognition, the nature of motivation, the nature of media, the nature of time, and the nature of
communication. Similar to the assumptions of the LC3MP, first, the nature of cognition refers to
the fact that people are limited information processors. In other words, a person’s cognitive
resources to perceive, encode, understand, and remember are limited. Second, people have two
motivational systems, the appetitive (approach) and the aversive (avoidance) systems, that
activate automatically “in response to motivationally relevant stimuli in the environment and
influence ongoing cognitive processing” (p. 59). The appetitive motivational system evolved to
help the individual to ensure survival, while the aversive motivational system evolved to protect
the individual from dangerous situations. Third, media consist of “multiple sensory channels
(e.g., eyes, ears, and touch) and formats (e.g., words, text, pictures, and moving pictures)” (p.
59). Fourth, human cognition and behaviors are constantly, continuously, and simultaneously
43
changing over time. Fifth, communication is interaction between communication messages and
its cognitive and motivational information processing.
Further, Lang (2006) pointed out that four questions should be asked to create effective
messages (p. 62): (1) What is the message goal?; (2) Who is in the target market?; (3) What
medium will carry the message?; and (4) What is the motivational and personal relevance of the
main information in the message for the majority of people in the target audience? Of these
questions, the third asking about the medium implies that different structural features from
different media result in automatic allocation of resources. As noted by Lang, numerous
researchers (e.g., Bolls, Muehling, & Yoon, 2003; Hitchon, Thorson, & Duckler, 1994) have
shown that different structural features are associated with the amount of information encoded
and stored as well as “evaluation of the message arguments, overall attitude toward the message,
and intent to engage in the behavior being advocated regardless of the informational content of
the message” (p. 63).
Multiple resource theory. The multiple resource theory originated from the concept of a
“single channel bottleneck” that forms because human beings cannot effectively deal with two
high-speed tasks together because of their limited capacity to process (Wickens, 2002). The
multiple resource theory has been used to explain the notion that “performing two tasks
simultaneously impedes task performance” because cognitive capacity to process information is
limited (Van Cauwenberge, Schaap, & Roy, 2014, pp. 101-102). The multiple resource theory
posits that people are able to pay attention to resources as far as their mental capacity is available
(Wickens, 1980). Hence, when people encounter tasks that exceed their processing capacities
(i.e., cognitive overload), their task performance tends to decrease.
44
The multiple resource theory premises that “processing resources are defined by
processing stages at two levels (perceptual/central processing versus response processes),
processing codes (verbal versus spatial), and processing modalities (auditory versus visual input,
vocal versus manual response)” (Wickens, 1981, p. 32). Buchholz and Smith (1991) suggested
four relationships (p. 1): (1) To the extent that two tasks require separate attentional resources
they will be time shared efficiently (i.e., without significant crosstask interference); (2) To the
extent that two tasks require common attentional resources the performance on the tasks will
depend on how resources are allocated to the tasks; (3) The difficulty of a task is increased when
additional resources are needed for its performance. If two tasks fully use available common
resources, increasing the difficulty of one task will hamper performance on a concurrent task. If
two tasks use separate resources, increasing the difficulty of one task will not affect performance
on a concurrent task; and (4) Task priority determines how attentional resources are allocated,
and plays the biggest role when tasks are difficult and share resources.
Leigh (1991) suggested that message congruence across modalities, stimulus information
factors, and motivation to process information are key predictors that influence different
responses to the modalities of different channels. In terms of message congruence among
modalities, TV ads with congruent audiovisual stimuli produce more cognitive processing than
those with incongruent audiovisual stimuli. Further, TV ads which contain audiovisual messages
stimulate better memory and recognition than radio ads that contain audio-only messages,
irrespective of the level of message congruence (Buchholz & Smith, 1991; Leigh, 1991). Low-
involvement consumers are unlikely to engage in cognitive processing, while high-involvement
consumers are more likely to do so.
45
Effects of Political Advertising
Channel Differences in Political Advertising
Political advertising refers to “the communication process by which a source (usually a
political candidate or party) purchases the opportunity to expose receivers through mass channels
to political messages with the intended effect of influencing their political attitudes, beliefs,
and/or behaviors” (Kaid, 1981, p. 250). However, this definition is based on the traditional
communication models of Lasswell (1948) and Berlo (1960) which consist of source, message,
channel, receiver, and effects. This model has limited application to all forms of political
advertising in various contexts (Kaid, 2004). Kaid (1999, 2004) suggested a much broader
definition of modern political advertising featuring two characteristics: control of the message
and use of mass communication channels where the message appears. Kaid (2004, p. 156)
broadly defined political advertising as “any message primarily under the control of a source”
used to advocate political candidates, parties, propositions, policy issues, and/or stances on
political ideas through some type of mass channels. Notably, Kaid (2004) noted that the type of
political advertising provided by mass communication channels such as print media (e.g.,
posters, pamphlets, brochures, direct mail, newspaper, and magazine), broadcast media (e.g.,
television and cable television), and electronic media (e.g., Internet or other electronic systems)
should be distinguished from that provided through interpersonal and general public
communication (e.g., political speeches or rallies).
The 1960 Kennedy-Nixon debates ignited the channel comparison argument in political
advertising (Kaid, 2002). The presidential debates were broadcast on both television and radio,
and Katz and Feldman (1962) found that television viewers thought Kennedy was the winner of
the debate, while radio listeners believed Nixon was the winner. McKinnon, Tedesco, and Kaid
46
(1993) investigated the channel difference in the 1992 presidential debate between Clinton and
Bush. They found that Clinton received a significantly higher score from people who listened to
the debate on the radio than from people who watched it on television, while no significant
change for Bush was found in either group. McKinnon and Tedesco (1999) found similar results
for Clinton and Dole in the 1996 presidential debate.
According to Kaid (2002, 2003), however, media research on political advertising has
never been very successful in providing strong evidence for channel difference. This is because
interactions between source (candidate) and channel variables complicate the discernible effects
of different channels (Andreoli & Worchel, 1978; Cohen, 1976; Kaid, 2002). For example, some
candidates do well in one medium while others excel at using another medium (Kaid, 2002).
Recognizing this issue, Kaid (2002) compared exposure to presidential ads via the Internet
versus traditional channels in the 2000 presidential campaign. To avoid interactions between the
source and channel variables, Kaid exposed participants to the same set of political
advertisements on a television monitor (traditional media group) and on a computer monitor
(Internet group). The findings showed that undecided voters exposed to Internet ads changed
their vote choice to Al Gore, while those exposed to television ads changed their vote choice to
George W. Bush. In addition, although there was no significant effect of political ads on political
cynicism, there were significant differences in the types of information seeking and political
activity intentions between the Internet and television groups.
Cognitive, Emotional, and Behavioral Responses to Political Advertising
In the 1970s, social scientists began to question the tradition of believing in minimal and
limited effects of communication channels, especially with the advent of political television
advertising (Kaid, 2004). Perloff (1998) argued that political advertising can influence “voters’
47
evaluations of candidates and their interpretations of political events” (p. 374). Kaid (2004)
noted three dimensions of the effects of political advertising: 1) cognitive effects on voter
knowledge levels, 2) affective effects on voter perceptions of candidates, and 3) behavioral
effects including effects on voting preferences and information-seeking behaviors. Later, Kaid
and Holtz-Bacha (2006) suggested that such cognitive, affective, and behavioral effects of
political advertisements are related to different characteristics of the advertising content,
different receiver’s characteristics, and different formats and channels of the political
advertising.
The cognitive effects of political advertising are related to voter’s knowledge levels
(Kaid, 2004). Early studies of voting decision-making have not upheld the expectation of an
informed electorate about politics and politicians in the United States (Delli Carpini & Keeter,
1996; Klapper, 1960; Lazarsfeld, Berelson, & Gaudet, 1948). Unlike advertising in newspapers,
televised advertising was free of partisan selectivity of the channel; empirical studies indicated
that televised political advertising will be more helpful for voter’s learning about issues and
candidates than advertising in newspapers (Kaid & Holtz-Bacha, 2006). Evidence has
demonstrated that exposure to political advertising can influence voter recall about specific
campaign issues and candidate issue positions (Atkin et al., 1973; Atkin & Heald, 1976; Faber &
Storey, 1984; Freedman, Franz, & Goldstein, 2004; Hofstetter & Strand, 1983; Holbert, et al.,
2002; Kaid, 1982; Martinelli & Chaffee, 1995; Pfau, et al., 2002; Ridout, Shaw, Goldstein, &
Franz, 2004; Sulfaro, 2001; Valentino, Hutchings, & Williams, 2004; West, 1994). In particular,
televised political advertising often produces greater effects on voters’ recall of the candidate’s
information when the candidate is unfamiliar to the voters (Kaid, 1982; Schleuder, 1990; West,
1994).
48
In addition, scholars examined whether the type of ad content (e.g., issue ads vs. image
ads and negative ads vs. positive ads), receiver characteristics (e.g., levels of campaign
involvement, voting decision status, and information seeking tendency), and the structure and
design of the political ads can affect voter’s recall level or overall information acquisition in
political campaigns. For example, image advertising results in greater recall of information than
issue advertising (Kaid & Sanders, 1978). Receiver characteristics are another factor that affects
recall of information from advertising exposure. People with lower feelings of involvement tend
to gain more knowledge from advertising exposure than those with higher feelings of
involvement (Rothschild & Ray, 1974). In regard to the structure of political ads, the effects on
recall and recognition are maximized when the audio and the video carry well-matched
information (Gunter, 1987; Lang & Lanfear, 1990; Reese, 1984). Television advertising was
found to be superior in terms of knowledge gain compared to other formats such as television
news or debates, but few studies have examined different channel effects of political advertising
on voter’s information gain (Kaid & Holtz-Bacha, 2006).
The affective effects of political advertising are relevant to candidate image evaluations
(Kaid, 2004). Researchers found that political knowledge and information did not necessarily
substitute for voter evaluations about a candidate or party (Kaid & Holtz-Bacha, 2006). Instead,
viewing political advertising had a significant influence on candidate image evaluation and
information learning on candidate issues and traits (Atkin & Heald, 1976; Cundy, 1986; Kaid &
Noggle, 1998, Kaid & Johnston, 2001; Kaid & Chanslor, 1995; Kaid & Sanders, 1978; Kaid &
Tedesco, 1999; West, 1993). Specifically, issue advertising was found to elicit more positive
evaluations toward a candidate than image advertising (Geiger & Reeves, 1991; Kaid, Chanslor,
& Hovind, 1992; Kaid & Sanders, 1978; Thorson, Christ, & Caywood, 1991). Findings also
49
demonstrated that positive advertising results in more positive candidate evaluations than
negative or comparative advertising (Hill, 1989; Kahn & Geer, 1994; Shen & Wu, 2002). In
regard to receiver characteristics, low-involvement viewers (e.g., the least aware viewers) in the
campaign tend to be more persuaded by the advertisements than high-involvement viewers
(Valentino, Hutchings, & Williams, 2004). Effects of channel difference on voter’s candidate
evaluation have been more of interest than before with the spread of the Internet in the political
advertising context (Kaid & Holtz-Bacha, 2006). In particular, researchers have found that the
Internet is a superior medium to generate higher candidate evaluations than television (Kaid,
2003).
The behavioral effects of political advertising include political behaviors such as voting
for or against particular candidates and information seeking behaviors. Although the direct
effects of political ads on receiver behavior are somewhat difficult to substantiate, significant
effects of televised political advertising on viewer’s political behaviors have been found. There
have been two streams of research explaining these direct effects on candidate voting decisions.
The first is to investigate a relationship between voting outcomes and amount of expenditures in
political advertising. The second is to examine a relationship between the exposure to political
ads and voting decisions. Findings showed that medium (e.g., television and the Internet) for
political advertising influences voting behaviors, although specific advertising effects differ by
channel difference (Kaid, 2002). As with the cognitive and affective effects of political ads,
further, receiver variables such as levels of involvement are related to advertising effects on
voting behavior. Research suggests that low-involvement voters are more susceptible to
persuasion from political advertising than high-involvement voters (Rothschild & Ray, 1974).
50
Taken together, the literature on the effects of political advertising has suggested three
types of effects: cognitive effects on voter knowledge levels, affective effects on voter
perceptions of candidates, and behavioral effects on voting preferences. The formats and
channels of the political advertising as well as the advertising content and receiver’s
characteristics are factors to influence the three dimensions. Although advertising scholars use
the term “affect” rather than “emotion,” emotional effects are more appropriate to explain
candidate evaluation for the purpose of this study. Positive or negative evaluation toward a
candidate can be better explained as “emotion,” referring to valenced (positive or negative)
orientations toward the objects, than “affect,” which represents overall feelings toward the object
including emotion, mood, and attitude.
Covariates
Cognitive and Emotional Involvement with Communication Channel
In social psychology, advertising, and marketing research, scholars have paid attention to
the concept of involvement as a predictor or a mediator of consumer behaviors (Huang, Chou, &
Lin, 2010; Sood, 2002). Sood considered involvement as a mediator between predictors such as
individual needs, sources of communication, and environmental factors and responses to
advertisements. Zaichkowsky (1985) proposed three antecedents of involvement that affect
consumers’ behavioral decisions: the characteristics of the person, the physical characteristics of
the stimulus, and the situation. Personal variables include individual needs, importance, interest,
and values that may determine a person’s involvement with a particular object. The stimulus
variables include differentiation of alternatives in the product being advertised, source of
communication (e.g., print, television, or radio), and content of communication. The situational
variable includes the specific situation for purchase or use of the product. Zaichkowsky argued
51
that all three antecedents influence the level of involvement, which affects the individual’s
responses to advertisements, products being advertised, and purchase decisions. Te’eni-Harari,
Lehman-Wilzig and Lampert (2009) also stated that “involvement is created by the personal
significance that the individual ascribes to the features of the object (message, situation, and
product) because involvement is a matter of interpretation, rather than the stimulus itself” (p.
204).
Although numerous researchers have investigated the effects of involvement on the
individual’s reactions to advertisements, the results for the relationships between involvement
and the effectiveness of advertising have been mixed (Lee & Thorson, 2009). Some scholars
have found that a lower involvement condition results in more attention, better recall for
messages, and more favorable attitudes toward the brand than a higher involvement condition
(Anand & Sternthal, 1992; Soldow & Principe, 1981), while others found the reverse (Kennedy,
1971). Lee and Thorson (2009) pointed out that these mixed results are due to the fact that there
is no consistent definition of involvement. Although communication researchers commonly and
broadly define involvement as “an internal state activated by a stimulus” which is relevant to
“personal relevance or importance of the product to the consumer” (Gotlieb, Schlacter, & Louis,
1992, p. 193), involvement has been measured in several ways such as “engaging in fewer
distracting activities while watching television,” “attention to program content,” “thinking and
talking about program content,” “looking forward to programs,” “feeling part of program
action,” “a characteristic of issues, objects, or events that is associated with greater willingness to
attend to and process information,” “a characteristic of the medium,” and “a personality trait”
(Perse, 1990b, p. 18).
52
In the particular context of channel difference research, Perse (1990a) regarded media
involvement as audience participation during information processing of messages they are
exposed to via media. Perse’s conceptualization of involvement focused on its participatory or
interactive nature. Specifically, the concept of involvement as a dimension of audience activity
in media use is in line with Levy and Windahl’s (1985) notion of involvement as “first, the
degree to which an audience member perceives a connection between him or herself and mass
media content” and “second, the degree to which the individual interacts psychologically with a
medium or its messages” (p. 112). Perse (1990b) further argued that involvement is a key
element that influences media uses and effects based on prior findings showing that “first,
involvement influences the gratifications that people receive from media use,” “second,
involvement influences subsequent planned media exposure,” and “third, involvement has been
linked to higher levels of soap opera satisfaction” (p. 18). These findings are also aligned with
research findings showing the persuasive, political, advertising, and interpersonal effects of
involvement (Perse, 1990a).
This study also focused on involvement as one of the constructs describing audience
activity, adopting Rubin and Perse’s (1987) definition of media involvement as “cognitive,
affective, and behavioral participation during and because of exposure” (p. 247). Media
involvement stresses media users’ interaction with media or messages represented by media
(Sun, Rubin, & Haridakis, 2008). For example, Sun et al. found that audience activity is manifest
in several dimensions such as individual’s media motives and selectivity, psychological
involvement, and media use and effects across different time periods of media exposure.
Furthermore, involvement arises during media exposure, while motives and selectivity for using
53
media are elicited before exposure and media use and effects are elicited after exposure (Levy &
Windhal, 1984; Sun et al., 2008).
Involvement during media exposure takes the form of cognitive information processing
and emotional reaction (Perse, 1990c). Cognitive processing generally emphasizes the reason or
logic of information, while emotional processing emphasizes feelings or emotions (Sojka &
Giese, 1997). Cognitive involvement induces thought-related reactions including attention,
recognition, elaboration, memory, knowledge, and comprehension of the media content, and
emotional involvement induces feelings such as humor, amusement, happiness, sadness, disgust,
applause, cheers, sighs, and fright or image related-aspects of objects which are linked to
purchase intention (Mittal, 1987; Park & Young, 1986; Perse, 1990c). For example, high
cognitive involvement with segments of a television program decreases memory of embedded
advertisements because cognitive activity is more focused on the program than on the
advertisements (Bryant & Comisky, 1978; Perse, 1990c). On the other hand, positive emotional
involvement with a television program induces more positive emotions toward embedded
advertisements (Goldberg & Gorn, 1987).
As such, prior research on individual’s involvement suggests that being engaged in
different media channels at different levels of cognitive and emotional involvement with the
channels influences expected audience’s responses. This study examines the cognitive,
emotional, and behavioral effects of different communication channels in political advertising in
the social media context. If people are engaged in social media platforms such as Facebook and
Twitter at different levels of cognitive and emotional involvement, it would influence their
expected cognitive, emotional, and behavioral responses. Hence, the present study used cognitive
and emotional involvement with social media platform as covariates to further examine the
54
cognitive, emotional, and behavioral effects of different communication channels including
interactivity, social media platform, and modality in political advertising in the social media
context.
Political Involvement
While involvement with communication channel may affect the user’s cognitive,
emotional, and behavioral responses to the channel, involvement with messages can influence
cognitive, emotional, and behavioral responses to the political advertising. Since this study is
conducted in the political advertising context, an individual’s involvement with the message may
also be a covariate that affects the cognitive, emotional, and behavioral responses to
communication channels in political campaign advertisements.
Scholars have shown that high-involvement messages have a greater influence on
personal relevance and personal connections than low-involvement counterparts (Petty &
Cacioppo, 1979; Petty, Cacioppo, & Schumann, 1983; Schiffman & Kanuk, 1997; Zaichkowsky,
1985). According to this view, the greater the perception of relevance, the higher the individual’s
involvement, which results in more effortful processing or systematic information processing
(i.e., investing a considerable amount of time and effort in searching for information they want
(Shao, Baker, & Wagner, 2004). In particular, the effects of involvement on an individual’s
responses to advertisements have been investigated in dual-processing models such as the
Elaboration Likelihood Model (ELM) of attitude change and the heuristic-systematic information
processing model (HSM; Chaiken, 1980; Eagly & Chaiken, 1993; Petty, Cacioppo, & Schumann,
1983).
The ELM holds that people are persuaded by the “elaboration likelihood of the
communication situation (i.e., the probability of message- or issue-relevant thought occurring)”
55
(Petty et al., 1983, p. 137). Specifically, the central route to persuasion is effective in a high
elaboration likelihood situation, while the peripheral route to persuasion is effective in a low
elaboration likelihood situation (Petty et al., 1983). The central route to persuasion occurs “under
the conditions where a message receiver is both motivated to process message content (e.g., due
to the personal relevance of the issue) and has the ability to process the content” (Andrews &
Shimp, 1990, p. 196). On the other hand, the peripheral route to persuasion occurs when these
conditions are not satisfied, so that the message receiver focuses more on background cues such
as music, source characteristics, scenery, and the number of message arguments than the quality
of message arguments (Andrews & Shimp, 1990). Thus, people decide whether to accept or
reject the message after thinking rationally under the central route or based on extrinsic product
cues in the peripheral route (Petty & Cacioppo, 1981). Taken together, the ELM contends that
attitude formed through the central route (high-involvement condition) has a significant
influence on a receiver’s behaviors, while attitude formation through the peripheral route (low-
involvement condition) does not (Kokkinaki & Lunt, 1999).
Alternatively, the HSM assumes that the extent of one’s issue involvement or response
involvement affects persuasion (Chaiken, 1980). Specifically, high involvement (e.g., personally
important topics or judgments) leads to message-based cognitions that influence persuasion,
while low involvement (e.g., personally unimportant topics or judgments) is concerned with
simple decision cues that influence persuasion (Chaiken, 1980). The HSM provides two routes to
persuasion: systematic and heuristic processing of information. Systematic processing occurs
when people actively exert cognitive effort to process the information by comprehending and
evaluating the message’s arguments, while heuristic processing occurs when people put forth
little cognitive effort to evaluate the message and rely on non-content inferences or source
56
characteristics (Chaiken, 1980). Thus, in systematic processing, easily accessible information
such as source factors indirectly influences the acceptance of a message’s conclusion, while in
heuristic processing, this information directly influences reception of the conclusion (Chaiken,
1980).
Both dual-processing models suggest that people with high involvement tend to put forth
more cognitive effort to evaluate an issue or product than those with low involvement (Chaiken,
1980; Petty et al., 1983). In line with this, there is evidence that when involvement is high,
central cues such as the quality of the arguments (i.e., strong vs. weak arguments) have a greater
influence on persuasion. However, when involvement is low, peripheral cues such as the
attractiveness, credibility, or prestige of a message source have a greater influence on persuasion
(Chaiken, 1980; Petty & Cacioppo, 1979; Petty, Cacioppo, & Goldman, 1981; Petty, Cacioppo,
& Heesacker, 1981; Petty et al., 1983). This is because people with high involvement tend to be
motivated to engage in issue-relevant thought and so exert the cognitive effort to evaluate the
issue-relevant arguments, while those with low involvement do not (Petty et al., 1983).
Applying the influence of an individual’s involvement with messages to the political
context, the extent of an individual’s political involvement may influence persuasion. For
example, people who are exposed to more politically related topics or judgments may be more
susceptible to being persuaded by the quality of message arguments than those who are exposed
to less politically related topics or judgments. In contrast, people who are exposed to fewer
politically related topics or judgments may be more susceptible to being persuaded by the
background cues of the message than those who are exposed to more politically related topics or
judgments. In this regard, people with higher political involvement may be less susceptible to
being persuaded by the contents of political campaign advertisements than those with lower
57
political involvement. Given that political involvement can be a key variable that influences the
persuasive process of political advertisements, this study used an individual’s political
involvement variables such as political orientation and political participation as covariates.
Overall Summary, Experimental Hypotheses and Research Questions
To date, the literature on advertising has investigated the relationship between advertising
effectiveness and advertising channels such as print media, broadcast media, and the Internet.
However, theories and evidence explaining cognitive and emotional responses to advertisements
have been somewhat mixed and inconsistent. Further, there has been little empirical research on
the social media context, despite the explosive growth of social media in political campaign
advertisements. As such, this study will look at how different channels in political campaign
advertisements influence users’ cognitive, emotional, and behavioral reactions in the social
media context. Specifically, this study focuses mainly on the interactivity and modality of the
channel to compare the effects of Facebook and Twitter as well as the effects of print and
audiovisual advertising in the social media context.
The effects of interactivity were explained by the theory of user control, the theory of
structural isomorphism, media richness theory, social presence theory, and the concepts of
cognitive load and disorientation. The theory of user control, the theory of structural
isomorphism, and the concepts of cognitive load and disorientation suggest that a more
interactive communication channel increases users’ learning, while too much user control or
inexperience in a hypertextual environment yield cognitive overload that impedes learning.
Specifically, the theory of user control explains that higher user control of the communication
channel increases learning by encouraging motivation to learn. Structural isomorphism provides
the notion that hypermedia facilitate information processing by creating and maintaining
58
meaningful links among related concepts. The media richness theory and the social presence
theory further provide that a higher interactive communication channel increases the user’s
emotional and behavioral responses in addition to cognitive responses. The media richness
theory holds that a richer medium which has multiple cues facilitating faster and more accurate
communications engenders more favorable evaluations and behavioral intentions toward the
product the medium provides than a leaner medium. Similarly, the social presence theory posits
that a medium with stronger social presence (e.g., interpersonal medium) increases trust,
satisfaction, and favorable attitudes toward the product or service than a lower social presence
medium.
Taken together, these theories of interactivity suggest that a more interactive
communication channel enhances users’ cognitive (e.g., learning), emotional (e.g., favorable
attitude toward the object), and behavioral responses (e.g., behavioral intentions) better than a
lower counterpart when the context of the communication channel does not exceed the user’s
cognitive load and the user experiences the communication environment through the channel. In
the political advertising context, the literature on the effects of political advertising suggests how
cognitive, emotional, and behavioral responses should be measured. Previous researchers have
measured cognitive responses as voter knowledge levels (i.e., voter learning about campaign
issues and the candidate’s information), affective responses as voter perceptions of candidates
(i.e., voter evaluations toward the candidate, candidate’s issues, and the party), and behavioral
responses as voting preferences (i.e., voting decisions and information seeking behaviors).
Reflecting the expected cognitive, emotional, and behavioral responses from the literature of
political advertising, the following hypotheses were proposed:
59
H1: People who are exposed to campaign ads on a higher interactive channel will learn more about the campaign than those who are exposed to campaign ads on a lower counterpart. H2: People who are exposed to campaign ads on a higher interactive channel will evaluate the candidate-related information more favorably than those who are exposed to campaign ads on a lower counterpart. H3: People who are exposed to campaign ads on a higher interactive channel will be more likely to engage in political behaviors for the candidate than those who are exposed to campaign ads on a lower counterpart. Consistent with the theories of interactivity, social media such as Facebook and Twitter
are unique in that they provide a user-friendly environment, facilitate the search for information
based on hypertextuality, encourage immediate feedback between users as in face-to-face
interaction, and offer messages tailored to the user’s needs and interpersonal communication.
Compared to Twitter, Facebook requires higher user control of the channel in that the social
network of Facebook is under the control of users who can accept or reject friend requests.
Although both Facebook and Twitter are based on hypertext features, the extent of the number of
links in a hypertext is different between Facebook and Twitter. Facebook relies on links among
more related or pre-existing social networks, such as friends, family, and co-workers, while
Twitter focuses less on it based on lack of interpersonal relationship between users. Furthermore,
Facebook’s relational and reciprocal social interaction between users makes it more similar to
face-to-face communications than Twitter. However, based on its short and text-based messages,
Twitter is more appropriate for real-time interaction in comparison with Facebook. Applying the
theories of interactivity, Facebook and Twitter may trigger different cognitive (i.e., increasing
learning), emotional (i.e., favorable evaluations), and behavioral effects (e.g., behavioral
intentions) based on their different interactive natures. Given that Facebook has more interactive
60
elements than Twitter, Facebook may increase cognitive, emotional, and behavioral responses as
compared with Twitter. Based on this reasoning, the following hypotheses are posited:
H4: People who are exposed to campaign ads on Facebook will learn more about the campaign than those who are exposed to campaign ads on Twitter. H5: People who are exposed to campaign ads on Facebook will evaluate the candidate-related information more favorably than those who are exposed to campaign ads on Twitter. H6: People who are exposed to campaign ads on Facebook will be more likely to engage in political behaviors for the candidate than those who are exposed to campaign ads on Twitter. On the other hand, theories of modality including the dual coding theory, the cue
summation theory, the limited capacity information-processing model, and the multiple resource
theory suggest that increased modality (e.g., multiple cues) of the communication channel causes
more cognitive effects, but the effect size is limited according to users’ cognitive ability to
process information and consistency of the modality. For example, the dual coding theory
explains that multimedia representations increase the user’s learning and recall more than a
single representation, but inconsistency of information between multiple cues decreases the
user’s attention because of the individual’s limited cognitive processing capacity. The cue
summation theory indicates that an increased number of cues enhances learning only if the cues
from different modalities are relevant to each other, but impedes learning at the point of over-
abundance of cues. The limited capacity information-processing model suggests that cognitive
information processing differs by individual goals and needs and the characteristics of
information because a person’s cognitive resources are limited. The multiple resource theory
similarly proposes that people are more likely to engage in cognitive information processing
when the modality is composed of congruent multiple cues and the information is consistent with
the individual’s goals due to the individual’s limited capacity for information processing.
61
Although most theories do not focus on emotional and behavioral responses to the
modality of different communication channels, it can be inferred that increased modality elicits
more emotional and behavioral reactions. The dual coding theory indicates that audiovisual
information arouses stronger emotional (e.g., affection toward the object) and behavioral
reactions (e.g., pro-social behaviors) than visual only information. Although the cue summation
theory focuses primarily on the user’s cognitive reactions (i.e., learning) to the modality of the
communication channel, it also represents the possibility of emotional reactions to the modality
such as satisfaction with the object. Given that all four theories of modality assume the
individual’s limited capacity for information, it can be also assumed that increased modality
increases emotional reactions as far as the individual has the capacity to process the cues.
Likewise, information processing theories of modality suggest that increased modality of
the communication channel evokes more cognitive, emotional, and (potential) behavioral
responses. Unlike traditional media that features a single modality, social media can include a
number of modalities. Based on this characteristic, social media can be used as a political
campaign advertising tool with different modalities such as visual or audiovisual. Although the
theories of modality have examined the effects of different modality in traditional media such as
newspapers and television, how the modality of political advertising embedded in the social
media context plays a role in engendering users’ cognitive, emotional, and behavioral responses
is unexplored. Accordingly, this study investigates how the modality of political advertising
embedded in social media differently evokes users’ cognitive, emotional, and behavioral
responses. Given that prior studies of modality suggest that audiovisual advertising may be more
effective than visual advertising, this study assumes that audiovisual ads will increase more
62
cognitive, emotional, and behavioral responses than visual ads. In relation to the effects of
political advertising, the following hypotheses are developed:
H7: People who are exposed to audiovisual campaign ads will learn more about the campaign than those who are exposed to visual campaign ads. H8: People who are exposed to audiovisual campaign ads will evaluate the candidate-related information more favorably than those who are exposed to visual campaign ads. H9: People who are exposed to audiovisual campaign ads will be more likely to engage in political behaviors for the candidate than those who are exposed to visual campaign ads. Despite a number of empirical studies to explain the effects of interactivity and modality
of traditional communication channels, findings provide unexpected results in the social media
context not only because the definition of interactivity is blurred but because prior research
overlooks the unique characteristics of social media as interactive communication channels with
a number of modalities. Although prior theories of interactivity and modality suggest that a
relatively more interactive communication channel and a communication channel with more
modalities may arouse more cognitive, emotional, and behavioral responses than a less
interactive channel and a channel with fewer modalities, the cognitive, emotional, and behavioral
responses when modality is combined with interactivity of the channel in social media
environment remain unknown. Specifically, there is little research on the cognitive, emotional,
and behavioral effects when televised political advertising in a less interactive channel competes
with print-based political advertising in a more interactive channel. To investigate the combined
effects of different social media platform, interactivity, and modality, the following research
questions are raised:
RQ1. Under what combined conditions are ads most effective in increasing individual’s learning about campaign information?
63
RQ2. Under what combined conditions are ads most effective in increasing individual’s favorable evaluation toward the candidate-related information? RQ3. Under what combined conditions are ads most effective in increasing individual’s political behaviors for the candidate?
64
CHAPTER THREE
RESEARCH METHOD
Participants and Design
The purpose of this study is to investigate whether and/or how different channels of
political advertisements affect the audience’s cognitive, emotional, and behavioral responses in
the social media context. The study is designed to investigate the effects of different social media
with different levels of interactivity on users’ processing of embedded visual and audiovisual ads
in terms of cognition (i.e., learning about campaign information), emotion (i.e., favorable
evaluation toward the candidate-related information), and behavioral responses (i.e., political
behaviors). To answer a series of research questions and hypotheses, this study utilized a 2
(social media platform: Facebook, Twitter) X 2 (interactivity: high, low) X 2 (modality: visual,
audiovisual) between-subjects experimental design. Social media platforms, interactivity, and
modality are the manipulated variables, while learning about campaign information, favorable
evaluation toward the candidate-related information, and political behaviors are measured as
dependent variables. This study also measured other covariates that may influence these effects,
including an individual’s cognitive and emotional involvement with social media platform and
political involvement.
The theoretical population of this study is eligible voters who use Facebook and Twitter.
A total of 527 voters were recruited to complete the online experiment, using a convenience
sampling procedure. Invitations distributed via Amazon Mechanical Turk were employed to
recruit the participants. As compensation, about $1 was given to the participants recruited via
65
Amazon Mechanical Turk. After removing incomplete responses, 396 responses were extracted.
To control for the effects of participant’s prior knowledge about the campaign, participants (n =
79) who were familiar with the political campaign and the candidate used in the stimuli in this
study were excluded. As a result, data from a total of 317 participants were used for analysis in
this study.
Stimuli
For this study, political advertisements of a candidate were manipulated through the
creation of eight different political ads, varying in a 2 (social media platform: Facebook, Twitter)
X 2 (interactivity: high, low) X 2 (modality: visual, audiovisual) design. In 2015, general
elections in the United States were scheduled to be held on November 3. This is an off-year
election for filling vacant seats if incumbents resigned or died. Of three gubernatorial races that
will be held in Kentucky, Louisiana, and Mississippi, the Kentucky gubernatorial election was
selected. Among five candidates who ran for the election, Hal Heiner was selected because he
uses Facebook and Twitter as a tool of the campaign.
An actual 30-second political advertisement with audiovisuals was obtained from
Heiner’s Facebook and Twitter postings. To compare the visual and audiovisual advertisements
under the same condition except for elements such as music, voice, and moving pictures, the
visual advertisement was manipulated to show eight main frames that the audiovisual
advertisement provides. The time of exposure was also controlled in the visual ad condition to be
equal to the time of the audiovisual presentation condition. Under the visual ad condition,
participants were exposed to the eight frames (3 seconds per frame) of the visual ads one by one
as a slide show.
66
Further, the visual and audiovisual advertisements without interactive features (low
interactivity condition) and with interactive features such as “reply,” “retweet,” and “favorite”
for Twitter and “reply,” “share,” and “like” for Facebook (high interactivity condition) served as
the stimuli to manipulate interactivity. To compare the effects of Facebook and Twitter, other
conditions such as the advertisements and the number of interactive features were equally
manipulated in addition to the media platform. As a result, eight experimental treatments were
created: (1) a visual ad with low interactivity on Facebook, (2) an audiovisual ad with low
interactivity on Facebook, (3) a visual ad with high interactivity on Facebook, (4) an audiovisual
ad with high interactivity on Facebook, (5) a visual ad with low interactivity on Twitter, (6) an
audiovisual ad with low interactivity on Twitter, (7) a visual ad with high interactivity on Twitter,
and (8) an audiovisual ad with high interactivity on Twitter.
Procedure
After completing the participant information sheet, the participants were asked about
whether they are familiar with the Kentucky gubernatorial election and Hal Heiner to control the
effect of participants’ prior knowledge about the campaign. Only participants who agree with the
informed consent form and those who are not familiar with the election and the candidate
completed the survey. Familiarity with the election and the candidate was measured by asking
participants, “How familiar are you with the Kentucky gubernatorial election?” and “How
familiar are you with the candidate, Hal Heiner?” The answer to each of the statements was
ranked on a 5-point scale, with 1 = very unfamiliar and 5 = very familiar. Of the 396 respondents
who completed the survey, 216 respondents (54.5%) answered they are very unfamiliar with the
Kentucky gubernatorial election, 107 (27.0%) answered they are unfamiliar, 53 (13.4%) are
somewhat familiar, 11 (2.8%) are familiar, and 9 (2.3%) are very familiar. In terms of the
67
familiarity with the candidate, Hal Heiner, 242 respondents (61.1%) answered they are very
unfamiliar with the candidate, 102 (25.8%) answered they are unfamiliar, 32 (8.1%) are
somewhat familiar, 16 (4.0%) are familiar, and 4 (1.0%) are very familiar. Only responses of
participants who are very unfamiliar or unfamiliar with the campaign and the candidate (n = 317)
were included in the analysis of the study, so their familiarity with the Kentucky gubernatorial
election and the candidate, Hal Heiner, was assumed to be low.
Before exposing participants to a stimulus, participants’ involvement variables (e.g.,
cognitive and emotional involvement with social media platform and political involvement) were
measured. Participants then read the statement, “Now, you will see a political advertisement
related to the current gubernatorial election. Please watch this advertising carefully.” They were
then randomly exposed to one of eight political advertisements by different modality,
interactivity, and social media platform. After being exposed to the stimulus, participants were
asked questions for the manipulation check, the dependent measures of learning about the
campaign, candidate evaluation, voting intention, and other sociodemographic variables.
Dependent Variables
Campaign Learning
The scale of campaign learning, adapted and modified from Feldman and Price (2008),
was administered to assess participants’ learning about the issues and image of the candidate.
The scale was measured by asking participants, “Please mark whether each of the statements is
correct, incorrect, or “do not know”. Correct responses were coded as 1, and incorrect or “do not
know” responses were coded as 0. Four items about issue knowledge were used to ask about
whether the candidate 1) favors Obama’s policy, 2) supports drug testing for welfare recipients, 3)
has a plan to increase Kentucky jobs, and 4) suggests making e-verify immigration screening. In
68
a similar vein, four items about knowledge of the candidate’s personal and political background
were used to ask about whether the candidate 1) is the “Frankfort outsider,” 2) is a candidate for
Governor of Kentucky, 3) is liberal, and 4) was not a successful businessman. Responses to these
eight items were averaged to create a composite campaign learning scale (M = .80, SD = .22, α
= .694).
Candidate Evaluation
The candidate evaluation scale was adapted from a “feeling thermometer” scale that has
been used to measure perceptions of candidate image in several studies (e.g., Kaid, 2003; Kaid,
Fernandes, & Painter, 2011; Kaid & Postelnicu, 2005; McKinney, Houston, & Hawthorne, 2014;
Weber, Dunaway, & Johnson, 2011). Participants were asked to rank three statements “How do
you rate your feelings toward Hal Heiner?”, “How do you rate your feelings toward the
campaign issues of Hal Heiner?”, and “How do you rate your feelings toward the party to which
Hal Heiner belongs?” The answer to each of the statements was ranked on a 5-point scale, with 1
= unfavorable-cold and 5 = favorable-warm. Responses to these three items were averaged to
create a composite candidate evaluation scale (M = 2.40, SD = 1.15, α = .935).
Political Behavior
The political behavior scale includes voting intention and information-seeking behaviors.
Voting intention was measured by asking participants, “If the election were held tomorrow, how
likely or unlikely would you be to vote for the candidate in the advertising you just watched?” on
a 5-point scale, with 1 = very unlikely and 5 = very likely. The scale of information-seeking
behaviors, adapted and modified from Kaid (2002), featured 12 items about which participants
were asked, “How likely or unlikely would you be engage in each of the following activities?”
on a 5-point scale, with 1 = very unlikely and 5 = very likely. Responses to the voting intention
69
and information-seeking behaviors (13 items) were averaged to create a composite information-
seeking behavior scale (M = 2.02, SD = .81, α = .909).
Covariates
Individual variables may affect cognitive, emotional, and behavioral responses to
different channels in political campaign advertisements. In particular, individual’s involvement
variables have been recognized as mediators and predictors of behavioral responses (Huang,
Chou, & Lin, 2010; Sood, 2002). In this regard, individual’s involvement variables were
measured as covariates for cognitive, emotional, and behavioral responses for additional analysis
in this study. These covariates were treated as exogenous to explain the effects of different
channels in political advertisements. Involvement variables include cognitive and emotional
involvement with social media platform, political orientation, and political participation.
Cognitive Involvement with Social Media Platform
The cognitive involvement with social media platform scale, adapted and modified from
Perse (1990a & 1990b), was asked participants “How much do you agree or disagree with each
of the following statements?” on a five-point scale, with 1 = strongly disagree and 5 = strongly
agree. A total of 15 items was used to measure three dimensions of cognitive involvement with
channel such as attention, recognition, and elaboration. First, attention was measured with five
items to assess an individual’s attention to Facebook and Twitter. After recoding some
negatively worded items, responses to these five items were averaged to create a composite
attention scale (M = 3.15, SD = 1.09, α = .897 for Facebook, M = 2.64, SD = 1.05, α = .858 for
Twitter, and M = 2.89, SD = 1.10 for social media). Second, recognition was measured with five
items to assess an individual’s understanding about new information on Facebook and Twitter.
After recoding some negatively worded items, responses to these five items were averaged to
70
create a composite recognition scale (M = 4.21, SD = .72, α = .832 for Facebook, M = 3.64, SD =
1.02, α = .889 for Twitter, and M = 3.91, SD = .93 for social media). Third, elaboration was
measured with five items to assess how an individual relates new information on Facebook and
Twitter to previous knowledge. Responses to these five items were averaged to create a
composite elaboration scale (M = 2.54, SD = 1.10, α = .939 for Facebook, M = 1.96, SD = 1.00, α
= .930 for Twitter, and M = 2.24, SD = 1.08 for social media). Response to these three items
were averaged to create a composite cognitive involvement with Facebook scale (M = 3.30, SD
= .82, α = .924), a composite cognitive involvement with Twitter scale (M = 2.75, SD = .85, α
= .928), and a composite cognitive involvement with social media scale (M = 3.01, SD = .88).
Emotional Involvement with Social Media Platform
The emotional involvement with social media platform scale, adapted from Perse (1990b),
uses 15 items to measure the intensity of emotions while using Facebook and Twitter,
respectively. Along with the factor analysis employed by Perse (1990b), a principal factor
analysis with oblique rotation with a three-factor solution was employed. As a result, three
distinct emotions such as happiness, anger, and sadness were extracted. Five items were included
in each emotion. In terms of Facebook use, Factor 1, Happiness (eigenvalue = 5.63), accounted
for 37.52% of the total variance. Factor 2, Anger (eigenvalue = 4.01), accounted for 26.76% of
the total variance. Factor 3, Sadness (eigenvalue = 1.64), accounted for 10.95% of the total
variance. Similarly, regarding Twitter use, Factor 1, Happiness (eigenvalue = 5.98), accounted
for 39.89% of the total variance. Factor 2, Anger (eigenvalue = 4.06), accounted for 27.07% of
the total variance. Factor 3, Sadness (eigenvalue = 1.73), accounted for 11.53% of the total
variance. Responses to the items from each emotion were averaged to a create composite
happiness scores (M = 2.15, SD = 1.02, α = .950 for Facebook, M = 1.83, SD = .93, α = .937 for
71
Twitter, and M = 1.98, SD = .99 for social media), anger scores (M = 1.37, SD = .61, α = .895 for
Facebook, M = 1.36, SD = .72, α = .935 for Twitter, and M = 1.37, SD = .67 for social media),
and sadness scores (M = 1.23, SD = .42, α = .858 for Facebook, M = 1.13, SD = .37, α = .888 for
Twitter, and M = 1.18, SD = .40 for social media). Response to these 15 items were averaged to
create a composite emotional involvement with Facebook scale (M = 1.58, SD = .48, α = .861), a
composite emotional involvement with Twitter scale (M = 1.44, SD = .47, α = 862), and a
composite emotional involvement with social media scale (M = 1.51, SD = .48).
Political Orientation
The political orientation scale asked participants “Please mark which best identifies your
political views” on a five-point scale, with 1 = very liberal and 5 = very conservative (M = 2.59,
SD = 1.05).
Political Participation
The political participation scale, adapted from Valenzuela, Park and Kee (2009), has nine
items asking participants, “Have you ever done any of the following activities within the last 12
months?” with 0 = never, 0.5 = yes, but not within the last 12 months, and 1 = yes, within the last
12 months. Responses to these nine items were summed to create a composite civic and political
participation scale (M = 1.67, SD = 1.86, α = .89).
72
CHAPTER FOUR
RESULTS
Study Participants
Of the 317 participants used in the study, 153 (48.3%) were men and 164 (51.75) were
women. The mean (M) age of the participants was 37.03 (SD = 11.72), ranging from 18 (n = 1,
0.3%), the youngest age of U.S. voters, to 69 (n = 1, 0.3%). Other demographic characteristics of
the participants are provided in Table 1.
Table 1. Demographics of Study Participants
Characteristic N %
Education level
Less than high school 3 0.9
High school/GED 44 13.9
Some college 116 36.6
College degree 115 36.3
Some graduate work 12 3.8
Master’s degree 21 6.6
PhD / Professor 6 1.9
Race
White-Non Hispanic 245 77.3
Asian or Asian American 25 7.9
73
Latino or Hispanic 22 6.9
African American or Black 16 5.0
Biracial (any combination or the above) 5 1.6
Other 4 1.3
Income
Less than $20,000 67 21.1
$20,000 to $39,999 77 24.3
$40,000 to $59,999 79 24.9
$60,000 to 79,999 50 15.8
$80,000 to $99,999 25 7.9
$100,000+ 19 6.0
Region
South 81 25.6
Midwest 77 24.3
Northeast 74 23.3
West 63 19.9
Mid-Atlantic 16 5.0
I do not live in the U.S. 6 1.9
Partisan
Democrat 146 46.1
Republican 52 16.4
Independent 99 31.2
74
Other 14 4.4
Do not know 6 1.9
In addition to demographic characteristics of the participants, the study tested whether
political orientation and political participation were distributed evenly into eight groups that were
exposed to each of eight different stimuli. Results of univariate analysis of variance (ANOVA)
with a post-hoc Bonferroni test showed that there were no significant differences between any of
the groups regarding political orientation (F(7, 309) = 1.578, η2 = .035, p > .05) and political
participation (F(7, 309) = .829, η2 = .018, p > .05).
Manipulation Check
The study manipulated social media platform (i.e., Facebook and Twitter), interactivity
(high and low), and modality (visual and audiovisual ads) to examine the cognitive, emotional,
and behavioral responses to different versions of a political ad. While different types of social
media platform and modality are noticeable and recognizable to the general user, there is no
easily identifiable manipulation of interactivity. For this reason, a manipulation check of
interactivity was included to determine whether participants perceived the different interactive
nature of advertisements via social media platform as they were intended. Nine items revised and
adapted from Thorson and Rogers’ (2010) scale of perceived interactivity were used on the post-
questionnaire.2 The scale asked participants to rank how much they agreed or disagreed with
items including: 1) Interacting with Facebook (or Twitter) was like having a personal
conversation with a representative of the campaign; 2) I felt as if the candidate’s Facebook (or
Twitter) talked back to me while I was navigating; 3) I perceive the candidate’s Facebook (or
2 Thorson and Rogers (2006) modified McMillan and Hwang’s (2002) scale of perceived interactivity for the political campaign context.
75
Twitter) to be sensitive to my needs for information; 4) The candidate’s Facebook (or Twitter) is
interpersonal; 5) I could communicate directly to the candidate if I wanted to using the
candidate’s Facebook (or Twitter); 6) I could communicate in real time with other users who
shared my interest in the campaign through the candidate’s Facebook (or Twitter); 7) The
candidate’s Facebook (or Twitter) had the ability to respond to my specific questions quickly and
efficiently; 8) It is easy to find my way through the candidate’s Facebook page (or Twitter); and
9) The candidate’s Facebook (or Twitter) has a variety of information I want to see. Answers
were given on a 5-point scale anchored by 1 = strongly disagree and 5 = strongly agree.
Responses to these nine items were averaged to create a composite interactivity of Facebook
scale (M = 2.28, SD = .82, α = .892), a composite interactivity of Twitter scale (M = 2.27, SD
= .83, α = .888), and composite interactivity of social media scale (M = 2.27, SD = .82, α = .885).
The findings indicate a successful manipulation (𝑀𝑀high= 2.42, 𝑆𝑆𝑆𝑆high= .83 vs. 𝑀𝑀low= 2.14,
𝑆𝑆𝑆𝑆low= .80; t [315] = 3.11, p < .01). Despite the successful manipulation, the mean values of
both high and low interactivity are still low, which indicates that participants did not pay
attention to the stimuli they were exposed to or they perceived the stimuli as less interactive
regardless of whether they were exposed to the manipulated high or low interactive channels.
Data Analysis
To test the nine research hypotheses, a series of two-tailed Mann-Whitney U tests and a
series of independent samples t-tests were employed. The campaign learning variable is highly
skewed due to a large number of participants having higher scores of campaign learning after
being exposed to one of eight political advertisements. The mean score of a composite campaign
learning scale was .80 of 1.00. For that reason, the campaign learning variable cannot be
normalized by data formation (Skewness = -1.427 and Kurtosis = 2.051). Because use of
76
independent samples t-tests is not recommended for data that are not normalized, nonparametric
tests that do not require normality, such as two-tailed Mann-Whitney U tests, were used to test
effects on campaign learning. A series of two-tailed Mann-Whitney U tests were thus carried out
to find out different effects of interactivity, modality, and social media platform on campaign
learning. A series of independent samples t-tests were used to test different effects of
interactivity, modality, and social media platform on candidate evaluation and political behavior.
H1, H2, and H3 predicted different effects of interactivity on campaign learning,
candidate evaluation, and political behavior. H1 predicted that people who were exposed to
campaign ads on a higher interactive channel learned more about the campaign than those who
were exposed to lower interactive campaign ads. The results of the two-tailed Mann-Whitney U
test revealed that there was no significant difference between high and low interactive channels
in relation to campaign learning, U = 12108.500, p > .05 (see Table 2). Therefore, H1 was not
supported.
Table 2. Mann-Whitney U Test of High and Low Interactive Channels on Campaign Learning
N M SD Mean Ranks Sum of Ranks
High interactivity 154 .80 .22 161.87 24928.50
Low interactivity 163 .79 .22 156.29 25474.50
Note: Mean Difference = .00; Mann-Whitney U = 12108.500; Z = -.559; p > .05
H2 predicted that people who were exposed to campaign ads on a higher interactive
channel evaluated the candidate-related information more favorably than those who were
exposed to campaign ads on a lower counterpart. The independent samples t-tests showed that
77
candidate evaluation was not significantly different when participants were exposed to campaign
ads on a higher interactive channel (M = 2.44, SD = 1.17) compared to a lower counterpart (M =
2.37, SD = 1.14), t (315) = .599, p > .05 (see Table 3). Hence, H2 was not supported.
H3 predicted that people who are exposed to campaign ads on a higher interactive
channel were more likely to intend to engage in political behaviors for the candidate than those
who are exposed to campaign ads on a lower counterpart. According to the results of the
independent samples t-tests, political behavior was not significantly different when participants
were exposed to campaign ads on a higher interactive channel (M = 2.08, SD = .79) compared to
a lower counterpart (M = 1.97, SD = .82), t (315) = 1.162, p > .05 (see Table 3). Thus, H3 was
not supported.
Table 3. Comparing Candidate Evaluation and Political Behavior between High and Low
Interactive Channels
N M SD t d.f. p
Candidate Evaluation High interactivity 154 2.44 1.17 .599 315 .549
Low interactivity 163 2.36 1.14
Political Behavior High interactivity 154 2.08 .79 1.162 315 .246
Low interactivity 163 1.97 .82
H4, H5, and H6 predicted different effects of social media platform on campaign learning,
candidate evaluation, and political behavior. H4 predicted that people who were exposed to
campaign ads on Facebook learned more about the campaign than those who were exposed to
campaign ads on Twitter. A two-tailed Mann-Whitney U test was used for the analysis. The
78
result revealed that campaign learning was significantly different when participants were
exposed to campaign ads on Facebook compared to Twitter. Specifically, participants who were
exposed to campaign ads on Facebook (M = .83, SD = .20) obtained significantly higher scores
of campaign learning than those who were exposed to campaign ads on Twitter (M = .77, SD
= .23), U = 10721.500, p < .05 (see Table 4). Therefore, H4 was supported.
Table 4. Mann-Whitney U Test of Facebook and Twitter on Campaign Learning
N M SD Mean Ranks Sum of Ranks
Facebook 152 .83 .20 170.96 25986.50
Twitter 165 .77 .23 147.98 24416.50
Note: Mean Difference = .06; Mann-Whitney U = 10721.500; Z = -2.298; p < .05
H5 predicted that people who were exposed to campaign ads on Facebook evaluated the
candidate-related information more favorably than those who were exposed to campaign ads on
Twitter. Independent samples t-tests were employed. The results revealed that candidate
evaluation was not significantly different when participants were exposed to campaign ads on
Facebook (M = 2.38, SD = 1.15) compared to Twitter (M = 2.43, SD = 1.16), t (315) = -.426, p
> .05 (see Table 5). Hence, H5 was not supported.
H6 predicted that people who were exposed to campaign ads on Facebook were more
likely to engage in political behaviors for the candidate than those who were exposed to
campaign ads on Twitter. The analysis of independent samples t-tests revealed that political
behavior was not significantly different when participants were exposed to campaign ads on
79
Facebook (M = 1.97, SD = .79) compared to Twitter (M = 2.07, SD = .82), t (315) = -1.08, p
> .05 (see Table 5). Thus, H6 was not supported.
Table 5. Comparing Candidate Evaluation and Political Behavior between Facebook and Twitter
N M SD t d.f. p
Candidate Evaluation Facebook 152 2.38 1.15 -.426 315 .670
Twitter 165 2.43 1.16
Political Behavior Facebook 152 1.97 .79 -1.076 315 .283
Twitter 165 2.07 .82
H7, H8 and H9 predicted different effects of modality on campaign learning, candidate
evaluation, and political behavior. H7 predicted that people who were exposed to audiovisual
campaign ads learned more about the campaign than those who were exposed to visual campaign
ads. The results of a two-tailed Mann-Whitney U test demonstrated that campaign learning was
significantly different when participants were exposed to audiovisual campaign ads compared to
visual campaign ads. Participants who were exposed to audiovisual campaign ads (M = .87, SD
= .16) obtained significantly higher scores of campaign learning than those who were exposed to
visual campaign ads (M = .73, SD = .24), U = 8096.000, p < .001 (see Table 6). This result
supported H7.
80
Table 6. Mann-Whitney Test of Visual and Audiovisual Advertisements on Campaign Learning
N M SD Mean Ranks Sum of Ranks
Visual Ads 160 .73 .24 131.10 20976.00
Audiovisual Ads 157 .87 .16 187.43 29427.00
Note: Mean Difference = .14; Mann-Whitney U = 8096.000; Z = -5.638; p < .001
H8 predicted that people who were exposed to audiovisual campaign ads evaluated the
candidate-related information more favorably than those who were exposed to visual campaign
ads. Results from independent samples t-tests confirmed this prediction. Findings showed that
candidate evaluation was significantly higher when participants were exposed to audiovisual
campaign ads (M = 2.55, SD = 1.25) than visual campaign ads (M = 2.26, SD = 1.03), t (315) = -
2.318, p < .05 (see Table 7). Thus, H8 was supported.
H9 predicted that people who were exposed to audiovisual campaign ads were more
likely to engage in political behaviors for the candidate than those who were exposed to visual
campaign ads. However, the results of the independent samples t-tests revealed that political
behavior was not significantly different when participants were exposed to audiovisual campaign
ads (M = 2.03, SD = .83) compared to visual campaign ads (M = 2.02, SD = .78), t (315) = -.091,
p > .05 (see Table 7). This result means that H9 was not supported.
81
Table 7. Comparing Candidate Evaluation and Political Behavior between Visual and
Audiovisual Advertisements
N M SD t d.f. p
Candidate Evaluation Visual Ads 160 2.26 1.03 -2.318 315 .021
Audiovisual Ads 157 2.55 1.25
Political Behavior Visual Ads 160 2.02 .78 -.091 315 .928
Audiovisual Ads 157 2.03 .83
RQ1, RQ2, and RQ3 asked about the effects of interactivity, modality, and social media
platform on campaign learning, candidate evaluation, and political behavior. RQ1 asked under
what combined conditions ads are most effective in increasing an individual’s learning about
campaign information. Univariate analysis of covariance (ANCOVA) with post-hoc Bonferroni
test was conducted with interactivity (high vs. low), modality (audiovisual vs. visual), and social
media platform (Facebook vs. Twitter) as fixed factors predicting campaign learning, controlling
for political orientation, political participation, cognitive involvement with social media platform,
and emotional involvement with social media platform as covariates.
The result showed that covariates of political orientation, political participation,
cognitive involvement with social media platform, and emotional involvement with social media
platform were not significantly related to increased political learning. Controlling for four
covariates, the result indicated a significant two-way interaction between modality and social
media platform, F(1, 305) = 11.163, η2 = .035, p < .001, explaining about 3.5% of variance.
Participants who are exposed to audiovisual campaign ads on Twitter (M = .88, SE = .02) learned
more about the campaign than those who are exposed to audiovisual campaign ads on Facebook
82
(M = .85, SE = .02). In contrast, participants who are exposed to visual campaign ads on
Facebook (M = .80, SE = .02) learned more about the campaign than those who are exposed to
visual campaign ads on Twitter (M = .67, SE = .02).
Table 8. Tests of Between-Subjects Main and Interaction Effects: Interactivity, Modality and
Social media as Predictors of Campaign Learning
Campaign Learning
d.f. F η2
Corrected model 11 6.077*** .180
Covariates
Political orientation 1 .879 .003
Political participation 1 3.654 .012
Cognitive involvement with social media platform 1 2.236 .007
Emotional involvement with social media platform 1 3.155 .010
Main effects
Interactivity 1 .363 .001
Modality 1 33.592*** .099
Social media platform 1 4.963* .016
Interactions
Interactivity * Modality 1 .005 .000
Interactivity * Social media platform 1 .015 .000
Modality * Social media platform 1 11.163*** .035
83
Interactivity * Modality * Social media platform 1 1.604 .005
Error 305
*p < .05; **p < .01; ***p < .001
Figure 1. Interaction of Modality and Social Media Platform on Campaign Learning
RQ2 asked under what combined conditions ads are most effective in increasing
individual’s favorable evaluation toward candidate-related information. Univariate analysis of
covariance (ANCOVA) with post-hoc Bonferroni test was conducted with interactivity (high vs.
low), modality (audiovisual vs. visual), and social media platform (Facebook vs. Twitter) as
fixed factors predicting candidate evaluation, controlling for political orientation, political
participation, cognitive involvement with social media platform, and emotional involvement with
social media platform as covariates.
The result showed that the covariates of political orientation and cognitive involvement
with social media platform were positively related to increased candidate evaluation. Participants
00.10.20.30.40.50.60.70.80.9
1
Facebook Twitter
Campaign Learning
Visual Audiovisual
84
who perceived themselves as conservative reported a more favorable evaluation toward the
candidate than those who perceived themselves as liberal, F(1, 305) = 327.969, η2 = .518, p
< .001, explaining about 51.8% of variance. Participants with higher levels of cognitive
involvement with the social media platform reported a more favorable evaluation toward the
candidate than those with lower cognitive involvement, F(1, 305) = 5.900, η2 = .019, p < .05,
explaining about 1.9% of variance.
Controlling for four covariates, the result indicated a significant two-way interaction
between modality and social media platform, F(1, 305) = 4.731, η2 = .015, p = .030, explaining
about 1.5% of variance. In the Facebook condition, participants who were exposed to audiovisual
campaign ads (M = 2.47, SE = .09) reported a more favorable evaluation toward the candidate
than those who are exposed to visual campaign ads (M = 2.13, SE = .09). In the Twitter condition,
however, participants who were exposed to visual campaign ads (M = 2.52, SE = .09) reported a
more favorable evaluation toward the candidate than those who are exposed to audiovisual
campaign ads (M = 2.48, SE = .09). These results are presented in Table 9 and Figure 2.
85
Table 9. Tests of Between-Subjects Main and Interaction Effects: Interactivity, Modality and
Social media as Predictors of Candidate Evaluation
Candidate Evaluation
d.f. F η2
Corrected model 11 36.648*** .569
Covariates
Political orientation 1 327.969*** .518
Political participation 1 .852 .003
Cognitive involvement with social media platform 1 5.900* .019
Emotional involvement with social media platform 1 1.814 .006
Main effects
Interactivity 1 .023 .000
Modality 1 2.745 .009
Social media platform 1 4.775* .015
Interactions
Interactivity * Modality 1 .353 .000
Interactivity * Social media platform 1 .014 .000
Modality * Social media platform 1 4.731* .015
Interactivity * Modality * Social media platform 1 .490 .002
Error 305
*p < .05; **p < .01; ***p < .001
86
Figure 2. Interaction of Modality and Social Media Platform on Candidate Evaluation
RQ3 asked under what combined conditions ads are most effective in increasing an
individual’s political behaviors on behalf of the candidate. Univariate analysis of covariance
(ANCOVA) with post-hoc Bonferroni test was conducted with interactivity (high vs. low),
modality (audiovisual vs. visual), and social media platform (Facebook vs. Twitter) as fixed
factors predicting political behavior. Political orientation, political participation, cognitive
involvement with social media platform, and emotional involvement with social media platform
were entered as covariates to control for these factors.
As expected, the covariates of political orientation, political participation, cognitive
involvement with social media platform, and emotional involvement with social media platform
were positively related to increased political behavior. Specifically, participants who perceived
themselves as liberal engaged more in political behaviors (i.e. voting behavior or political
information seeking behaviors) than those who perceived themselves as conservative, F(1, 305)
= 96.898, η2 = .241, p < .001, explaining about 24.1% of variance. Participants who have more
1.9
2
2.1
2.2
2.3
2.4
2.5
2.6
Facebook Twitter
Candidate Evaluation
Visual Audiovisual
87
actively participated in politics in the past engaged more in political behaviors than those who
less actively participated in politics, F(1, 305) = 27.298, η2 = .082, p < .001, explaining about
8.2% of variance. Participants with higher levels of cognitive involvement with the social media
platform engaged more in political behaviors than those with lower cognitive involvement, F(1,
305) = 12.337, η2 = .039, p < .001, explaining about 3.9% of variance. Participants with higher
levels of emotional involvement with the social media platform engaged more in political
behaviors than those with lower emotional involvement, F(1, 305) = 6.306, η2 = .020, p < .05,
explaining about 2.0% of variance.
Controlling for the four covariates, although no interactive effects were found for
interactivity, modality, and social media, there was a significant main effect of social media
platform on political behavior, F(1, 305) = 11.374, η2 = .036, p < .001, explaining about 3.6% of
variance. Post-hoc pairwise comparison tests showed that social media platform positively
affected political behavior, indicating that participants who are exposed to campaign ads on
Twitter (M = 2.14, SE = .05) engaged more in political behaviors than those who were exposed
to campaign ads on Facebook (M = 1.89, SE = .06). See Table 10.
88
Table 10. Tests of Between-Subjects Main and Interaction Effects: Interactivity, Modality and
Social Media as Predictors of Political Behavior
Political Behavior
d.f. F η2
Corrected model 11 15.357*** .356
Covariates
Political orientation 1 96.898*** .241
Political participation 1 27.298*** .082
Cognitive involvement with social media platform 1 12.337*** .039
Emotional involvement with social media platform 1 6.306* .020
Main effects
Interactivity 1 .343 .001
Modality 1 .239 .001
Social media platform 1 11.374*** .036
Interactions
Interactivity * Modality 1 .026 .000
Interactivity * Social media platform 1 .755 .002
Modality * Social media platform 1 1.574 .005
Interactivity * Modality * Social media platform 1 .516 .002
Error 305
*p < .05; **p < .01; ***p < .001
89
Table 11 summarizes the findings of this study.
Table 11. Results of Hypotheses and Research Questions
Hypotheses and Research Questions Result
H1: People who are exposed to campaign ads on a
more interactive channel will learn more about the
campaign than those who are exposed to campaign
ads on a less counterpart.
Not supported. No significant
difference between high and low
interactive channels in campaign
learning.
H2: People who are exposed to campaign ads on a
higher interactive channel will evaluate the
candidate-related information more favorably than
those who are exposed to campaign ads on a lower
counterpart.
Not supported. No significant
difference between high and low
interactive channels in candidate
evaluation.
H3: People who are exposed to campaign ads on a
higher interactive channel will be more likely to
engage in political behaviors for the candidate than
those who are exposed to campaign ads on a lower
counterpart.
Not supported. No significant
difference between high and low
interactive channels in political
behavior.
H4: People who are exposed to campaign ads on
Facebook will learn more about the campaign than
those who are exposed to campaign ads on Twitter.
Supported. Participants who were
exposed to campaign ads on Facebook
(M = .83, SD = .20) obtained
significantly higher scores of
campaign learning than those who
90
were exposed to campaign ads on
Twitter (M = .77, SD = .23), U =
10721.500, p < .05.
H5: People who are exposed to campaign ads on
Facebook will evaluate the candidate-related
information more favorably than those who are
exposed to campaign ads on Twitter.
Not supported. No significant
difference between Facebook and
Twitter in candidate evaluation.
H6: People who are exposed to campaign ads on
Facebook will be more likely to engage in political
behaviors for the candidate than those who are
exposed to campaign ads on Twitter.
Not supported. No significant
difference between Facebook and
Twitter in political behavior.
H7: People who are exposed to audiovisual campaign
ads will learn more about the campaign than those
who are exposed to visual campaign ads.
Supported. Participants who were
exposed to audiovisual campaign ads
(M = .87, SD = .16) obtained
significantly higher scores of
campaign learning than those who
were exposed to visual campaign ads
(M = .73, SD = .24), U = 8096.000, p
< .001.
H8: People who are exposed to audiovisual campaign
ads will evaluate the candidate-related information
more favorably than those who are exposed to visual
campaign ads.
Supported. Candidate evaluation was
significantly higher when participants
were exposed to audiovisual campaign
ads (M = 2.55, SD = 1.25) than visual
91
campaign ads (M = 2.26, SD = 1.03), t
(315) = -2.318, p < .05.
H9: People who are exposed to audiovisual campaign
ads will be more likely to engage in political
behaviors for the candidate than those who are
exposed to visual campaign ads.
Not supported. No significant
difference between audiovisual and
visual campaign ads in political
behavior.
RQ1. Under what combined conditions are ads most
effective in increasing individual’s learning about
campaign information?
There is a significant two-way
interaction between modality and
social media platform (F(1, 305) =
11.163, η2 = .035, p < .001).
Audiovisual campaign ads on Twitter
were most effective and visual
campaign ads on Twitter were least
effective.
RQ2. Under what combined conditions are ads most
effective in increasing individual’s favorable
evaluation toward the candidate-related information?
There is a significant two-way
interaction between modality and
social media platform (F(1, 305) =
4.731, η2 = .015, p < .05). Visual
campaign ads on Twitter were most
effective and visual campaign ads on
Facebook were least effective.
RQ3. Under what combined conditions are ads most
effective in increasing individual’s political behaviors
There was a significant effect of social
media platform on political behavior
92
for the candidate? (F(1, 305) = 11.374, η2 = .036, p
< .001). Campaign ads on Twitter
were more effective than campaign
ads on Facebook.
93
CHAPTER FIVE
DISCUSSION
Since McLuhan (1964, p. 7) made the claim that “the medium is the message,” studies on
channels of communication have proliferated. Researchers have been especially curious about
the role channels can play in the receiver’s interpretation of a message (Kaid & Postelnicu,
2005). In relation to political campaigns, candidates employ a new communication channel when
it appears to be able to help them spread their messages. Before the emergence of the Internet,
scholars regarded massive campaign coverage by media as a type of political campaign that is
different from interpersonal campaigning such as coffees or door-to-door canvassing. Further,
scholars have investigated the power of each medium in political campaigns, focusing on
modalities that correspond to human senses (i.e., print media for seeing, radio for hearing, and
television for both seeing and hearing). Scholars are currently investigating the different effects
of the Internet compared to other traditional media based on its massive and interpersonal
campaigning characteristics in different modes such as oral and/or visual presentation. Since the
successful use of Facebook and Twitter in the 2008 presidential election, many studies have paid
attention to the role of social media in individual’s political responses as a more personalized and
dialogical campaigning tool compared to websites and online communities.
However, studies on the influence of social media have produced mixed results, with
some demonstrating that political cognition, emotion, and behaviors were enhanced by exposure
to social media in political campaigns and others finding no significant effect. Further, few
studies have looked at the different characteristics of social media in consideration of its
94
dialogical and interactive features nor investigated the different uses of each of social media
platform with the same contents in different modes of presentation. Hence, the present study
examines the effects of social media on political outcomes (i.e. campaign learning, candidate
evaluation and political behavior) while considering different characteristics of social media,
including social media platform (i.e., Facebook vs. Twitter), interactivity (i.e., high vs. low) and
modality (i.e. visual vs. audiovisual).
The overall findings supported the proposed hypotheses about different effects of social
media platform on campaign learning and different effects of modality on campaign learning and
candidate evaluation. Specifically, Facebook and Twitter differently influence campaign learning
(H4), and visual and audiovisual advertisements differently influence campaign learning (H7)
and candidate evaluation (H8). However, no significant effects of interactivity on political
outcomes were found. Further, the findings did not support the hypotheses about different effects
of interactivity, social media platform, and modality on political behaviors. Interestingly,
however, under combined conditions of social media platform and modality when involvement
variables were used as covariates, a significant interaction was found in campaign learning and
candidate evaluation. An effect of social media platform on political behavior was also found,
controlling for involvement variables. The specific findings are discussed below.
No Significant Effects of Interactivity on Political Outcomes
The results for the first three hypotheses and the three research questions indicated no
significant effect of interactivity on campaign learning, candidate evaluation, and political
behavior. Several factors may influence this insignificant effect. In this study, interactivity is
manipulated by varying the interface’s capacity for communication between users and the
interface including “reply,” “retweet,” “favorite,” “share,” and “like”. However, it still remains
95
unknown which technological elements of social media platforms contribute to interactivity.
According to Sundar et al. (2003), not only the interface’s interactive features but the user’s
involvement in interactive communication should be included in the categorization of
interactivity. In order to fill the gap between the interface’s and user’s side of interactivity, the
present study measured perceived interactivity and found that the manipulation was successful.
However, the scores of participants’ self-reported answers were still low even for high
interactive channels so that it is still unclear whether participants really paid attention to the
interactive nature of social media or engaged in interactive use of the social media platform.
Unfortunately, the experiment and survey method used in this study do not examine the
relationship between the features of interactivity and the user’s participation in interactive
communication. Additional in-depth interview methods or a simulated online experiment may be
able to cover this topic more thoroughly in future studies.
Theories related to interactivity such as the theory of user control, the theory of structural
isomorphism, the media richness theory, and the social presence theory note that channels with
certain interactive features (e.g., user controllability, hypertextual structures, feedback capability,
and social presence) enhance cognitive, emotional, and potential behavioral responses under the
condition that individual’s cognitive demand and disorientation are low. According to this
notion, individual’s cognitive demand and disorientation may influence the results. Further, as
suggested in the theory of user control, some variables such as users’ personal interests, abilities,
background knowledge, and learning style may influence the expected outcome, especially in
relation to campaign learning (Eveland & Dunwoody, 2002). However, it is impossible to
consider all confounding variables in an experimental setting. To systematically examine
interactivity effects on political outcomes, future researchers should include covariate measures
96
such as cognitive demand, disorientation, personal interests, abilities, background knowledge,
and learning style.
Interaction Effects of Modality and Social Media on Campaign Learning
As expected, the results for H4 and H7 revealed different effects of social media platform
(Facebook vs. Twitter) and modality (visual vs. audiovisual advertisements). Specifically, the
results for H4 revealed that Facebook viewers had greater learning about the campaign than did
Twitter viewers, while the results for H7 also showed that viewers of audiovisual ads had greater
campaign learning than viewers of visual ads. These findings are hardly surprising, given that
each social media platform has distinct interactive features such as more direct and synchronized
communication on Twitter and more two-way, interpersonal, and reciprocal communication on
Facebook (Davenport et al., 2014; Hughes et al., 2011), and the result of the manipulation check
of the perceived interactivity scale of the social media platform showed that participants
perceived Facebook as more interactive than Twitter. This is consistent with the theory of user
control and the theory of structural isomorphism, which focus on the influence of interactivity on
user’s cognitive outcomes such as learning.
However, it is interesting to find a significant effect of the interaction of modality and
social media platform on campaign learning shown in the result of the analysis for RQ1. The
findings indicate that while audiovisual ads serve to increase campaign learning for the Twitter
condition, visual ads tend to increase the level of campaign learning for the Facebook condition.
These findings reveal that while audiovisual ads uniformly positively influence individual’s
learning about the campaign, the effects of social media platform are not so uniform. The two-
way interactions between modality and social media platform render additive consequences of
97
the main effects of modality and social media platform, with audiovisual ads on Twitter being
most effective in contributing to campaign learning and visual ads doing the same for Facebook.
The findings confirmed that social media play an important role in providing political
information based on its interactive nature, as noted in prior studies (see Delacourt, 2014;
Gottfried, 2014; Rutledge, 2013). However, the findings additionally demonstrated that different
use of the social media platforms with different cues may result in different effects on political
learning. For example, prior research has shown that Facebook use is more linked to involving
users in various social and informational activities, while Twitter use is more linked to engaging
politically interested users in gaining political information (Delacourt, 2014; Gottfried, 2014;
Rutledge, 2013). According to the theories of interactivity (e.g., the theory of user control, the
theory of structural isomorphism, the media richness theory, the social presence theory, cognitive
load, and disorientation), a more interactive communication channel increases users’ learning,
but cognitive overload impedes learning. The theories of modality (e.g., dual coding and cue
summation theories, the limited capacity information-processing model, and the multiple
resource theory) also suggest that stimuli with more cues (e.g., visual and audio information)
evoke more powerful effects on learning and emotional reactions than those with a single cue
(e.g. visual information), but only as far as the individual has the capacity to process the cues.
Based on these suggestions, these findings may also be due to individual differences in the
usability, structure, and design of Facebook and Twitter, as noted in prior studies (see Kaid,
2004; Kaid & Holtz-Bacha, 2006; Sundar, 2004). However, this assumption is beyond the topical
and methodological areas of this study, and thus future research into the effects of modality,
individual’s mental capacity, and message congruency will be needed to provide more
theoretical elaboration.
98
Interaction Effects of Modality and Social Media on Candidate Evaluation
Inconsistent with expectations, the results for H5 demonstrated no significant difference
in candidate evaluation between Facebook and Twitter users. However, as expected, the results
for H8 showed that viewers of audiovisual ads had more favorable candidate evaluation than
viewers of visual ads. The results of the analysis for RQ2 indicated a significant effect of the
interaction between modality and social media platform on candidate evaluation. Specifically,
audiovisual ads on Facebook tend to increase favorable candidate evaluation more than visual
ads on Facebook, while visual ads on Twitter had more favorable candidate evaluation than
audiovisual ads on Twitter.
These inconsistencies in the results between H5 and RQ2 may be due to the powerful
influence of a covariate such as political orientation on candidate evaluation. Findings
demonstrated that an individual’s political orientation explains 51.8% of variance, while
interaction between modality and social media platform explains only 1.5% of variance on the
effects on candidate evaluation. These findings are hardly surprising, given that dual-processing
models such as the Elaboration Likelihood Model (ELM) and the heuristic-systematic
information processing model (HSM) suggest that people who have higher involvement with the
messages tend to be more susceptible to being persuaded by the content of advertisements than
those who have lower involvement with the messages. Since this study used advertisement
stimuli of a Republican candidate, people who perceived themselves as conservative (higher
involved condition) are likely to be more persuaded by the contents of the advertisements than
those who perceived themselves as liberal (lower involved condition).
Given that social media are based on interactivity, the interaction effects of modality and
social media platform on candidate evaluation can be explained by Sundar’s (2004) argument
99
about the combined effects of interactivity and modality on users’ attitudinal responses to the
interface. Sundar noted that interactivity may additively combine with a technological variable
such as modality to increase favorable attitudes, but a plethora of such cues may evoke negative
attitudes. Sundar provided several propositions to explain the reason for this argument. First, the
abundance of cues through combination impedes gaining information that requires users’
evaluation. Second, the appearance of such cues alters the way users perceive interactivity.
Third, more cues lead to more distraction because the process of perception is attentional.
The media richness theory and the social presence theory also provide possible reasons
for the combined effects of modality and social media platform on candidate evaluation based on
the effects of the number of cues provided by a medium. According to the media richness theory,
a richer medium which provides more attributes (e.g., quicker feedback between users and more
social, non-verbal, and complex cues) and satisfies the user’s need to communicate, leading to
more favorable evaluations toward the products the medium advertises than a lean medium
(Coyle & Thorson, 2001; Dennis & Kinney, 1998; Ledford, 2012; Sheer, 2011; Simon & Peppas,
2004; Sundar, 2000). The social presence theory posits that a higher-presence medium perceived
by users as more personal, more sociable, more sensitive, and warmer facilitates more favorable
attitudes toward the agent than a lower presence medium (Kuma & Benbasat, 2002; Park & Lee,
2013; Skalski & Tamborini, 2007).
However, although these theories suggest an association between the extent of richness or
social presence of a medium and users’ emotional responses, the particular attributes the medium
holds that create the individual’s perception of the cues of the richness or social presence of a
medium are unclear. Future research on individuals’ perceptions of the richness and social
presence of a medium should consider the degree of complex cues in accordance with
100
interactivity, modality, and social media platform. This detailed investigation should give more
insight into the emotional responses to the combination of interactivity, modality, and social
media platform.
Effect of Social Media on Political Behavior
The results for H3, H6, and H9 represent insignificant effects of interactivity, modality,
and social media platform on political behavior. This insignificant effect is not shocking, given
that research on political advertising has hardly substantiated the direct effects of political ads on
receiver behavior (Kaid, 2004; Kaid & Holtz-Bacha, 2006).
However, the result of the analysis for RQ3 revealed a significant effect of social media
on political behavior when an individual’s political involvement variables such as political
orientation and prior participatory behaviors and an individual’s cognitive and emotional
involvement with social media platform were used as covariates. Prior findings related to the
effects of social media on political outcomes have been mixed (Bollen, Mao, & Pepe, 2010;
Hesseldahl, MacMillan, & Kharif, 2008; Marchese, 2008; Owen, 2008; Ancu & Cozma, 2009;
Gil de Zúñiga, Puig, & Rojas, 2009). The findings of this study support the idea that involvement
affects the relationship between exposure to social media use and political behavior. Specifically,
the findings indicated that Twitter tends to be more effective for increasing political behavior
than Facebook under the condition that involvement variables were used as covariates. In
particular, an individual’s political orientation explains 24.1% of variance, prior participatory
behaviors explains 8.2% of variance, cognitive involvement with social media platform explains
3.9% of variance, and emotional involvement with social media platform explains 2.0% of
variance, while the effect of social media platform explains 3.6% of variance on political
behavior. These findings suggest that an individual’s political involvement variables such as
101
political orientation and prior participatory behaviors and the individual’s cognitive and
emotional involvement with social media platform are significant variables that influence
political behaviors. These findings are in line with prior research demonstrating significant
effects of political advertising on political behaviors, especially focusing on channel difference
and receiver variables such as involvement (Kaid, 2002; Rothschild & Ray, 1974).
However, questions about how political involvement and cognitive and emotional
involvement with Facebook and Twitter interactively and/or differently influence the relationship
between the exposure of each social media and political behavior are still unsolved and beyond
the scope of this study. Future research focusing on the effects of individuals’ cognitive and
emotional involvement with channels on political outcomes in various situations will provide
more solid theoretical explication of the effects of individual involvement.
102
CHAPTER SIX
CONCLUSION
The aim of the present study is to examine whether and/or how different channels of
political advertisements affect the audience’s cognitive, emotional, and behavioral responses in
the social media context. In order to investigate the effects of different channels of political
advertisements, different channels such as social media platform (Facebook vs. Twitter), the
level of interactivity (high vs. low) and modality (visual vs. audiovisual advertisements) were
used as predictors.
Previous studies of the relationship between media use and political outcomes have often
used the terms “medium,” “channel,” and “modality” interchangeably or have not clearly
differentiated between them. With the emergence of the Internet, the boundaries of these terms
are more blurred. In order to resolve the mixed use of concept, the present study started by
reexamining different conceptualizations of medium, channel, mode of presentation, and media
platform (e.g., Berlo, 1960; Chong, 2012; Lasswell, 1948; McLuhan, 1964; Shannon & Weaver,
1949) and conceptualized each of the medium, channel, modality, and media platform. Further,
by reviewing unique characteristics of social media, the study included interactivity and modality
in addition to different social media platforms such as Facebook and Twitter. These aspects of
this study provide a better understanding of the validity of traditional theories about medium and
communication channel, especially in the social media context.
In terms of cognitive, emotional, and behavioral effects of interactivity and modality,
prior studies have provided mixed results based on different theoretical frameworks. Since a
103
single conceptualization of interactivity does not exist, the study further investigated the theories
related to certain interactive features of social media such as user control, hypertexuality, real-
time interaction, and direct, two-way, interpersonal, and reciprocal relationships. Regarding the
effects of interactivity, for example, the theory of user control, the theory of structural
isomorphism, the media richness theory and the social presence theory suggest that interactive
channels provoke more cognitive, emotional, and potential behavioral responses, but cognitive
load and disorientation impede cognitive responses. Similarly, in terms of the effects of modality,
dual coding and cue summation theories suggest that channels with multiple cues increase
cognitive, emotional, and potential behavioral responses in contrast to the limited-capacity
information processing and multiple resource theories. Given that lack of explanation of the
effects of interactivity and modality in relation to social media, especially in the political
campaign context, this study attempts to theorize the cognitive, emotional, and behavioral effects
of interactivity, modality, and social media platform based on theories of interactivity and
modality in addition to empirical studies on the effects of political advertising. Therefore, this
study will provide theoretical contributions to the understanding of individuals’ cognitive,
emotional, and behavioral reactions to different channels in the political campaign context.
The findings of this study suggest that interactivity, modality, and social media platform
differently influence campaign learning, candidate evaluation, and political behavior.
Specifically, Facebook generated greater campaign learning than Twitter without consideration
of other factors such as interactivity and modality, and audiovisual advertisements generated
greater campaign learning than visual counterparts. However, there were interactive effects on
campaign learning and candidate evaluation when social media are combined with modality. In
terms of responses to political behavior, significant effects of social media platform were found
104
when political orientation, political participation, and cognitive and emotional involvement with
social media were added as covariates, showing that Twitter generated more political behaviors
than Facebook.
Practically, the findings of this study may help politicians and political campaign
practitioners build appropriate strategies for political campaigns through social media platforms
to encourage their target publics to engage in their expected political activities including political
learning, candidate evaluation, and political behaviors. For example, campaign strategies should
be designed differently according to the public they wish to target, the social media they wish to
use, and the goal they wish to meet.
Study Limitations
The findings of this study should be interpreted with caution. First, this study obtained
participants from the subject pool of Amazon.com’s Mechanical Turk (MTurk). Regarding the
quality of data using MTurk, Buhrmester, Kwang and Gosling (2011, p. 3) found that 1) MTurk
participants are slightly more demographically diverse than are standard Internet samples and are
significantly more diverse than typical American college samples; 2) participation is affected by
compensation rate and task length, but participants can still be recruited rapidly and
inexpensively; 3) realistic compensation rates do not affect data quality; and 4) the data obtained
are at least as reliable as those obtained via traditional methods. Although the MTurk has the
potential to recruit a diverse subject pool rapidly and inexpensively, this population has different
levels of gender, age, education, political knowledge, political ideology, political interest, and
use of social media compared with the general public; these may influence the outcomes of this
study (Berinsky, Huber, & Lenz, 2012). However, unlike a survey method, an experiment does
not require the generalizability of the findings but focuses on testing causal effects of two
105
variables in certain theoretical condition. Nevertheless, replication of the findings in various
settings will provide broader knowledge of the effects of political advertisements.
Second, the measurement of key variables such as interactivity and cognitive, emotional,
and behavioral responses may also pose a limitation. With regard to the measure of interactivity,
the study manipulated high and low interactivity conditions based on the number of interactive
features such as “reply,” “retweet,” and “favorite” for Twitter and “reply,” “share,” and “like”
for Facebook. Given that the concept of interactivity is defined in this study as an interactive
communication process between the structure of a communication channel and user’s perception,
the best way to measure high and low interactivity condition would be to measure the interactive
communication process between the structure of a communication channel and the user’s
perception in real time. The measurement used in this study focused on the structure of a
communication channel rather than the user’s perception of interactivity. To overcome this
drawback, the study used participants’ self-reported answers about the different interactive
nature of the advertisements they watched via social media platform. However, the scores of
participants’ self-reported answers were still low for both high and low interactive channels.
Thus, it is still unclear whether participants are really engaged in the interactive use of the social
media platform. Future researchers should replicate the findings of this study using in-depth
interview or a simulated online experiment to validate the measurement of interactivity. In a
similar vein, this study measured cognitive response as “campaign learning,” emotional response
as “candidate evaluation,” and behavioral response as “political behavior” based on Kaid’s (2004)
conceptualization of the effects of political advertising. However, cognitive, emotional, and
behavioral responses can be measured as more than the three variables used in this study. For
example, some cognitive responses such as attention, recognition, awareness, and thoughts are
106
still untested. Future investigation of the instruments to measure untested variables of cognitive,
emotional, and behavioral responses is needed to ensure the validity of the measurement.
The lack of other variables related to cognitive, emotional, and behavioral outcomes in
the political campaign context is the third limitation of this study. In addition to involvement
with media channels, prior studies suggested that sociodemographic variables (e.g., gender, age,
race, income, years of education, and region), media engagement variables (e.g., news media use
and interactive media use), political variables (e.g., political interest, political efficacy, political
trust, strength of party identification, and interpersonal political discussion) can be expected to
be mediators between exposure to political advertising and expected political outcomes,
especially voting behaviors (Cho et al., 2009; Jung, Kim, & Gil de Zúñiga, 2011; Lee, Shah, &
McLeod, 2012; Shah, Cho, Eveland, & Kwak, 2005; Shah et al., 2007). In this sense, future
researchers that are considering the effects of demographic, personality, political, and media use
variables should review and replicate the findings of this study.
Finally, this study used political campaign advertisements of one of the candidates in the
2015 Kentucky gubernatorial election as the stimuli. The time frame and campaign stimuli may
have been beneficial to the study in that this study intends to control the effects of participant’s
prior knowledge about the campaign by removing data from participants who are familiar with
the political campaign and the candidate. However, future studies in other election settings or the
non-election setting would broaden the knowledge of expected cognitive, emotional, and
behavioral responses to channel differences in political campaign advertisements.
107
REFERENCES
ABC news (2013, March 25). How Social Media Could Impact 2016 Presidential Election. ABC news. Retrieved from http://abcnewsradioonline.com/politics-news/how-social-media-could-impact-2016-presidential-election.html#ixzz3UCXZjkTx
Althaus, S. L., & Tewksbury, D. (2000). Patterns of Internet and traditional news media use in a networked community. Political Communication, 17(1), 21-45. doi: 10.1080/105846000198495.
Altheide, D. L.,& Snow, R. P. (1991). Media worlds in the postjournalism era. New York: Aldine de Gruyter.
Anand, P., & Sternthal, B. (1992). The effects of program involvement and ease of message counterarguing on advertising persuasiveness. Journal of Consumer Psychology, 1(3), 225-238. doi:10.1016/S1057-7408(08)80037-X
Ancu, M., & Cozma, R. (2009). MySpace politics: Uses and gratifications of befriending candidates. Journal of Broadcasting & Electronic Media, 53(4), 567-583. doi: 10.1080/08838150903333064
Andreoli, V., & Worchel, S. (1978). Effects of Media, Communicator, and Message Position on Attitude Change. Public opinion quarterly, 42(1), 59-70. doi: 10.1086/268429
Andrews, J. C., & Shimp, T. A. (1990). Effects of involvement, argument strength, and source characteristics on central and peripheral processing of advertising. Psychology & Marketing, 7(3), 195-214. doi: 10.1002/mar.4220070305
Andrews, N. (2015, February 17). On Twitter, 2016 Rivals Let the Jabs Fly. The Wall Street Journal. Retrieved from http://www.wsj.com/articles/on-twitter-2016-rivals-let-the-jabs-fly-1424201759
Ansolabehere, S., Behr, R., & Iyengar, S. (1993). The media game: American politics in the television age. New York: Macmillan.
Atkin, C. K., Bowen, L., Nayman, O. B., & Sheinkopf, K. G. (1973). Quality versus quantity in televised political ads. Public Opinion Quarterly, 37(2), 209-224. doi: 10.1086/268078
Atkin, C., & Heald, G. (1976). Effects of political advertising. Public Opinion Quarterly, 40(2), 216-228. doi: 10.1086/268289
Auter, P. J., & Davis, D. M. (1991). When characters speak directly to viewers: Breaking the fourth wall in television. Journalism & Mass Communication Quarterly, 68(1-2), 165-171. doi: 10.1177/107769909106800117
108
Bagui, S. (1998). Reasons for increased learning using multimedia. Journal of educational multimedia and hypermedia, 7(1), 3-18.
Berger, C. R., & Chaffee, S. H. (1989). Handbook of Communication Science. Newbury Park, CA: Sage.
Berinsky, A. J., Huber, G. A., & Lenz, G. S. (2012). Evaluating online labor markets for experimental research: Amazon. com's Mechanical Turk. Political Analysis, 20(3), 351-368. doi:10.1093/pan/mpr057
Berlo, D. K. (1960). Process of Communication: An Introduction to Theory and Practice. New York, NY: Holt, Rinehart, and Winston.
Bertot, J. C., Jaeger, P. T., & Grimes, J. M. (2012). Promoting transparency and accountability through ICTs, social media, and collaborative e-government. Transforming Government: People, Process and Policy, 6(1), 78-91.
Bertot, J. C., Jaeger, P. T., & Hansen, D. (2012). The impact of polices on government social media usage: Issues, challenges, and recommendations. Government Information Quarterly, 29(1), 30-40. doi:10.1016/j.giq.2011.04.004
Bevan, J. L., Pfyl, J., & Barclay, B. (2012). Negative emotional and cognitive responses to being unfriended on Facebook: An exploratory study. Computers in Human Behavior, 28(4), 1458-1464. doi:10.1016/j.chb.2012.03.008
Bodensteiner, W. D. (1970). Information channel utilization under varying research and development project conditions: an aspect of inter-organizational communication channel usage. The University of Texas at Austin
Bollen J, Mao H, Pepe, A. (2010). Determining the public mood state by analysis of microblogging posts. Paper presented at the 12th International Conference on the Synthesis and Simulation of Living Systems. Odense, Denmark, October 2010. Retrieved from: http://pti.iu.edu/pubs/determining-public-mood-state-analysis-microblogging-posts.
Bolls, P. D., Muehling, D. D., & Yoon, K. (2003). The effects of television commercial pacing on viewers' attention and memory. Journal of Marketing Communications, 9(1), 17-28. doi: 10.1080/1352726032000068032
Boyd, D. M., & Ellison, N. B. (2007). Social network sites: Definition, history, and scholarship. Journal of Computer‐Mediated Communication, 13(1), 210-230. doi: 10.1111/j.1083-6101.2007.00393.x
Brader, T. (2006). Campaigning for hearts and minds: How emotional appeals in political ads work. Chicago: University of Chicago Press.
Brashears, M. T., Fraze, S., Lawver, D., & Baker, M. (2005). The effects of cue summation on student cognition and satisfaction in an electronically-delivered secondary agriculture sciences
109
unit of instruction. Paper presented at the 2005 AAAE research conference, May 25-28, San Antonio, TX.
Bryant, J., & Comisky, P. W. (1978). The effect of positioning a message within differentially cognitively involving portions of a television segment on recall of the message. Human Communication Research, 5(1), 63-75. doi: 10.1111/j.1468-2958.1978.tb00623.x
Bucher, H. J., & Schumacher, P. (2006). The relevance of attention for selecting news content. An eye-tracking study on attention patterns in the reception of print and online media. Communications, 31(3), 347-368. doi: 10.1515/COMMUN.2006.022
Buchholz, L. M., & Smith, R. E. (1991). The role of consumer involvement in determining cognitive response to broadcast advertising. Journal of Advertising, 20(1), 4-17. doi: 10.1080/00913367.1991.10673202
Bucy, E. P., & Newhagen, J. E. (1999). The emotional appropriateness heuristic: Processing televised presidential reactions to the news. Journal of Communication, 49(4), 59-79. doi: 10.1111/j.1460-2466.1999.tb02817.x
Buhrmester, M., Kwang, T., & Gosling, S. D. (2011). Amazon's Mechanical Turk a new source of inexpensive, yet high-quality, data?. Perspectives on psychological science, 6(1), 3-5. doi: 10.1177/1745691610393980
Capobres, K. (2012, November 7). Election 2012: President Obama, Twitter Break Records. Fox News. Retrieved from http://latino.foxnews.com/latino/news/2012/11/07/election-2012-president-obama-twitter-break-records/
Chaiken, S. (1980). Heuristic versus systematic information processing and the use of source versus message cues in persuasion. Journal of personality and social psychology, 39(5), 752-766. doi: http://dx.doi.org/10.1037/0022-3514.39.5.752
Chi, F., & Yang, N. (2011). Twitter adoption in Congress. Review of Network Economics, 10(1), 1446-9022. doi: 10.2202/1446-9022.1255
Cho, J. (2005). Media, interpersonal discussion, and electoral choice. Communication Research, 32(3), 295-322. doi: 10.1177/0093650205275382
Cho, J., Shah, D. V., McLeod, J. M., McLeod, D. M., Scholl, R. M., & Gotlieb, M. R. (2009). Campaigns, Reflection, and Deliberation: Advancing an O‐S‐R‐O‐R Model of Communication Effects. Communication Theory, 19(1), 66-88. doi: 10.1111/j.1468-2885.2008.01333.x
Chong, E. (2012). The emotions social media bring to news: The emergence of empathy and compassion as elements in news messages. University of Maryland.
Clark, J. M., & Paivio, A. (1991). Dual coding theory and education. Educational psychology review, 3(3), 149-210.
110
Cohen, A. A. (1976). Radio vs. TV: The effect of the medium. Journal of Communication, 26(2), 29-35. doi: 10.1111/j.1460-2466.1976.tb01376.x
Conway, J. K. (1967). Multiple-sensory modality communication and the problem of sign types. Educational Technology Research and Development, 15(4), 371-383.
Corrado, A. & Firestone, C. M. (1996). Elections in cyberspace: Toward a new era in American politics. Washington, DC: Aspen Institute.
Coyle, J. R., & Thorson, E. (2001). The effects of progressive levels of interactivity and vividness in web marketing sites. Journal of advertising, 30(3), 65-77. doi: 10.1080/00913367.2001.10673646
Craik, F. I., & Lockhart, R. S. (1972). Levels of processing: A framework for memory research. Journal of verbal learning and verbal behavior, 11(6), 671-684. doi:10.1016/S0022-5371(72)80001-X
Cundy, D. T. (1986). Political commercials and candidate image: The effect can be substantial. In L. L. Kaid, D. Nimmo, & K. R. Sander (Eds.), New perspectives on political advertising (pp. 210-234). Carbondale: Southern Illinois University Press.
Cutbirth, C.W., & Coombs, T. (1997). The Coming Electronic Democracy: The Internet, Political Communication, and the Duties of Citizenship. Paper presented at the National Communication Association Convention, Chicago.
Cyr, D., Hassanein, K., Head, M., & Ivanov, A. (2007). The role of social presence in establishing loyalty in e-service environments. Interacting with Computers, 19(1), 43-56. doi:10.1016/j.intcom.2006.07.010
Dadashzadeh, M. (2010). Social media in government: From eGovernment to eGovernance. Journal of Business & Economics Research (JBER), 8(11), 81-86.
Daft, R. L., & Lengel, R. H. (1986). Organizational information requirements, media richness and structural design. Management science, 32(5), 554-571.
Daignault, P., Soroka, S., & Giasson, T. (2013). The Perception of Political Advertising During an Election Campaign: A Measure of Cognitive and Emotional Effects. Canadian Journal of Communication, 38(2), 167-186.
Davenport, S. W., Bergman, S. M., Bergman, J. Z., & Fearrington, M. E. (2014). Twitter versus Facebook: Exploring the role of narcissism in the motives and usage of different social media platforms. Computers in Human Behavior, 32, 212-220. doi:10.1016/j.chb.2013.12.011
Davidson, R. J., Schwartz, G. E., Saron, C., Bennett, J., & Goleman, D. J. (1979). Frontal versus parietal EEG asymmetry during positive and negative affect. Psychophysiology, 16(2), 202-203.
Davis, R., & Owen, D. (1998). New media and American politics. New York: Oxford University Press.
111
Day, W. F., & Beach, B. R. (1950). A survey of the research literature comparing the visual and auditory presentation of information. Charlortesville: University of Virginia.
Delacourt, S. (2014, October 31). Facebook is personal, Twitter is political. thestar.com. Retrieved from http://www.thestar.com/news/insight/2014/10/31/facebook_is_personal_twitter_is_political.html
Delli Carpini, M. X., & Keeter, S. (1996). What Americans Know about Democracy and Why it Matters. New Haven, CT, Yale University Press.
Dennis, A. R., & Kinney, S. T. (1998). Testing media richness theory in the new media: The effects of cues, feedback, and task equivocality. Information systems research, 9(3), 256-274.
Deutsch, J. A., & Deutsch, D. (1963). Attention: some theoretical considerations. Psychological review, 70(1), 80-90. doi: http://dx.doi.org/10.1037/h0039515
Dijkstra, M., Buijtels, H. E. J. J. M., & van Raaij, W. F. (2005). Separate and joint effects of medium type on consumer responses: a comparison of television, print, and the Internet. Journal of Business Research, 58(3), 377-386. doi: 10.1016/s0148-2963(03)00105-x
Dimond, S. J., Farrington, L., & Johnson, P. (1976). Differing emotional response from right and left hemispheres. Nature, 261, 690-692.
Eagly, A. H., & Chaiken, S. (1993). The psychology of attitudes. Fort Worth, TX: Harcourt. Brace, Jovanovich.
eBizMBA (2015). Top 15 Most Popular Social Networking Sites. eBizMBA. Retrieved from http://www.ebizmba.com/articles/social-networking-websites
Edell, J. A., & Keller, K. L. (1989). The information processing of coordinated media campaigns. Journal of Marketing Research, 26(2), 149-163. doi: http://dx.doi.org/10.2307/3172602
Enli, G. S., & Skogerbø, E. (2013). Personalized campaigns in party-centred politics: twitter and Facebook as arenas for political communication. Information, Communication & Society, 16(5), 757-774. doi: 10.1080/1369118X.2013.782330
Enli, G. S., & Thumim, N. (2012). Socializing and self-representation online: exploring Facebook. Observatorio (OBS*), 6(1), 87-105.
Eveland, Jr, W. P., & Dunwoody, S. (2002). An investigation of elaboration and selective scanning as mediators of learning from the Web versus print. Journal of Broadcasting & Electronic Media, 46(1), 34-53. doi: 10.1207/s15506878jobem4601_3
Eveland, W. P., & Dunwoody, S. (1999). Using Hypermedia Research To Advance the Study of Learning on the World Wide Web. Research Monograph, 15, 1-36.
112
Eveland, W. P., & Dunwoody, S. (2001). User control and structural isomorphism or disorientation and cognitive load? Learning from the Web versus print. Communication research, 28(1), 48-78. doi: 10.1177/009365001028001002
Eveland, W. P., Cortese, J., Park, H., & Dunwoody, S. (2004). How Web site organization influences free recall, factual knowledge, and knowledge structure density. Human Communication Research, 30(2), 208-233. doi: 10.1111/j.1468-2958.2004.tb00731.x
Faber, R. J., & Storey, M. C. (1984). Recall of information from political advertising. Journal of Advertising, 13(3), 39-44. doi: 10.1080/00913367.1984.10672900
Fouhy, B. (2012, May 7). Twitter Plays Outsize Role In 2012 Campaign. Huffington Post. Retrieved from http://www.huffingtonpost.com/2012/05/07/twitter-2012-campaign-outsize-role_n_1494797.html
Freedman, P., Franz, M., & Goldstein, K. (2004). Campaign advertising and democratic citizenship. American Journal of Political Science, 48(4), 723-741. doi: 10.1111/j.0092-5853.2004.00098.x
Gantz, W. (1978). How Uses and Gratifications Affect Recall of Television News. Journalism Quarterly, 55(4), 664-672.
Gaudin, S. (2012, November 7). Twitter a big winner in 2012 presidential election. Computerworld. Retrieved from http://www.computerworld.com/article/2493332/social-media/twitter-a-big-winner-in-2012-presidential-election.html
Gefen, D., & Straub, D. W. (2004). Consumer trust in B2C e-Commerce and the importance of social presence: experiments in e-Products and e-Services. Omega, 32(6), 407-424. doi:10.1016/j.omega.2004.01.006
Geiger, S. F., & Reeves, B. (1991). The effects of visual structure and content emphasis on the evaluation and memory for political candidates. In F. Biocca (Eds.), Television and political advertising (pp. 125-143). Hillsdale, NJ: Lawrence Erlbaum.
Gil de Zúñiga, H. G., Puig-I-Abril, E., & Rojas, H. (2009). Weblogs, traditional sources online and political participation: an assessment of how the internet is changing the political environment. New media & society, 11(4), 553-574. doi: 10.1177/1461444809102960
Goffman, E. (1963). Behavior in public places: Notes on the social organization of gatherings. New York: The Free Press.
Goldberg, M. E., & Gorn, G. J. (1987). Happy and sad TV programs: How they affect reactions to commercials. Journal of Consumer Research, 14(3), 387-403.
Gotlieb, J. B., Schlacter, J. L., & Louis, R. D. S. (1992). Consumer decision making: A model of the effects of involvement, source credibility, and location on the size of the price difference required to induce consumers to change suppliers. Psychology & Marketing, 9(3), 191-208. doi: 10.1002/mar.4220090303
113
Gottfried, J. (2014, November 12). Facebook and Twitter as political forums: Two different dynamics. Pew Research Center. Retrieved from http://www.pewresearch.org/fact-tank/2014/11/12/facebook-and-twitter-as-political-forums-two-different-dynamics/
Grabe, M. E., Lang, A., & Zhao, X. (2003). News content and form implications for memory and audience evaluations. Communication Research, 30(4), 387-413. doi: 10.1177/0093650203253368
Gunawardena, C. N., & Zittle, F. J. (1997). Social presence as a predictor of satisfaction within a computer‐mediated conferencing environment. American journal of distance education, 11(3), 8-26. doi: 10.1080/08923649709526970
Gunter, B. (1987). Poor Reception: Misunderstanding and forgetting broadcast news. Erlbaum: Hillsdale, NJ.
Gunter, B., Furnham, A., & Griffiths, S. (2000). Children's memory for news: A comparison of three presentation media. Media Psychology, 2(2), 93-118. doi: 10.1207/S1532785XMEP0202_1
Ha, L., & James, E. L. (1998). Interactivity reexamined: A baseline analysis of early business web sites. Journal of Broadcasting & Electronic Media, 42(4), 457-474. doi: 10.1080/08838159809364462
Havick, J. (2000). The impact of the Internet on a television-based society. Technology in Society, 22(2), 273-287. doi:10.1016/S0160-791X(00)00008-7
Hesseldahl, A., MacMillan, D., & Kharif, O. (2008). The vote: A victory for social media, too. Business Week. Retrieved from http://www.businessweek.com/technology/content/nov2008/tc2008115_988160. htm.
Hill, R. P. (1989). An exploration of voter responses to political advertisements. Journal of Advertising, 18(4), 14-22. doi: 10.1080/00913367.1989.10673163
Hitchon, J., Duckler, P., & Thorson, E. (1994). Effects of ambiguity and complexity on consumer response to music video commercials. Journal of Broadcasting & Electronic Media, 38(3), 289-306. doi: 10.1080/08838159409364266
Hoban, C. F., & Van Ormer, E. B. (1951). Instructional film research 1918-1950 (rapid mass learning). Port Washington, L. I., New York: U.S. Navy Special Devices Center.
Hofstetter, C. R., & Strand, P. J. (1983). Mass media and political issue perceptions. Journal of Broadcasting & Electronic Media, 27(4), 345-358. doi: 10.1080/08838158309386501
Holbert, R. L., Benoit, W., Hansen, G., & Wen, W. C. (2002). The role of communication in the formation of an issue-based citizenry. Communication Monographs, 69(4), 296-310. doi: 10.1080/03637750216549
114
Hong, S., & Nadler, D. (2012). Which candidates do the public discuss online in an election campaign?: The use of social media by 2012 presidential candidates and its impact on candidate salience. Government Information Quarterly, 29(4), 455-461. doi:10.1016/j.giq.2012.06.004
Hoogeveen, M. (1997). Toward a theory of the effectiveness of multimedia systems. International journal of human-computer interaction, 9(2), 151-168. doi: 10.1207/s15327590ijhc0902_4
Huang, C. Y., Chou, C. J., & Lin, P. C. (2010). Involvement theory in constructing bloggers' intention to purchase travel products. Tourism Management, 31(4), 513-526. doi:10.1016/j.tourman.2009.06.003
Huberman, B. A., Romero, D. M., & Wu, F. (2009). Crowdsourcing, attention and productivity. Journal of Information Science, XX (X), 1-8. doi: 10.1177/0165551509346786
Hughes, D. J., Rowe, M., Batey, M., & Lee, A. (2012). A tale of two sites: Twitter vs. Facebook and the personality predictors of social media usage. Computers in Human Behavior, 28(2), 561-569. doi:10.1016/j.chb.2011.11.001
Jacques, W. W., & Ratzan, S. C. (1997). The Internet's World Wide Web and Political Accountability New Media Coverage of the 1996 Presidential Debates. American Behavioral Scientist, 40(8), 1226-1237. doi: 10.1177/0002764297040008019
James, M. (2010). TV still the favored medium for political ad spending. Los Angeles Times. Retrieved from http://articles.latimes.com/2010/oct/29/business/la-fi-ct-political-ads-20101029
Jiang, Z., Chan, J., Tan, B. C., & Chua, W. S. (2010). Effects of interactivity on website involvement and purchase intention. Journal of the Association for Information Systems, 11(1), 34-59.
Jo, S., & Kim, Y. (2003). The effect of web characteristics on relationship building. Journal of Public Relations Research, 15(3), 199-223. doi: 10.1207/S1532754XJPRR1503_1
Jokinen, K., & Raike, A. (2003). Multimodality–technology, visions and demands for the future. Paper presented at the 1st Nordic Symposium on Multimodal Interfaces. Copenhagen, September, 2003. Retrieved from http://www.ling.helsinki.fi/~kjokinen/Publ/200309MMNordic.pdf
Jonassen, D. H. (1988). Designing Structured Hypertext and Structuring Access to Hypertext. Educational technology, 28(11), 13-16.
Jones, E., & Kalyanaraman, S. (2000). Modality Effects on Memory for Multimedia Messages. Paper presented at the General Assemble and Scientific Conference of the International Association for Media and Communication Research, Singapore.
Jung, N., Kim, Y., & Gil de Zúñiga, H. (2011). The mediating role of knowledge and efficacy in the effects of communication on political participation. Mass Communication and Society, 14(4), 407-430. doi: 10.1080/15205436.2010.496135
115
Kahn, K. F., & Geer, J. G. (1994). Creating impressions: An experimental investigation of political advertising on television. Political Behavior, 16(1), 93-116. doi: 10.1007/BF01541644
Kaid, L. L. (1981). Political advertising. In Dan Nimmo & Keith R. Sanders (Eds.), The handbook of political communication (pp. 249-271). Beverly Hills, CA: Sage Publications.
Kaid, L. L. (1982). Paid television advertising and candidate name identification. Campaigns and Elections, 3, 34-36.
Kaid, L. L. (1999), Political Advertising: A Summary of Research Findings. In B. Newman (Eds.), The Handbook of Political Marketing (pp. 423-438). Thousand Oaks, CA: Sage Publications.
Kaid, L. L. (2002). Political advertising and information seeking: Comparing exposure via traditional and Internet channels. Journal of Advertising, 31(1), 27-35. doi: 10.1080/00913367.2002.10673658
Kaid, L. L. (2003). Effects of Political Information in the 2000 Presidential Campaign: Comparing Traditional Television and Internet Exposure. American Behavioral Scientist, 46(5), 677-691. doi: 10.1177/0002764202238492
Kaid, L. L. (2004). Handbook of political communication research. Mahwah, New Jersey: Routledge.
Kaid, L. L., & Chanslor, M. (1995). Changing candidate images: The effects of political advertising. In K. L. Hacker (Eds.), Candidate images in presidential elections (pp. 83-97). Westport, CT: Praeger Publishers.
Kaid, L. L., & Holtz-Bacha, C. (2006). The Sage handbook of political advertising. Thousand Oaks, CA: Sage Publications.
Kaid, L. L., & Johnston, A. (2001). Videostyle in presidential campaigns. Westport, CT: Praeger.
Kaid, L. L., & Noggle, G. (1998). Televised political advertising in the 1992 and 1996 elections: Using technology to manipulate voters. Southeastern Political Review, 26(4), 889-906. doi: 10.1111/j.1747-1346.1998.tb00514.x
Kaid, L. L., & Postelnicu, M. (2005). Political advertising in the 2004 election comparison of traditional television and internet messages. American Behavioral Scientist, 49(2), 265-278. doi: 10.1177/0002764205279421
Kaid, L. L., & Sanders, K. R. (1978). Political Television Commercials An Experimental Study of Type and Length. Communication Research, 5(1), 57-70. doi: 10.1177/009365027800500103
Kaid, L. L., & Tedesco, J. C. (1999). Tracking reactions to the television advertising. In L. L. Kaid & D. Bystrom (Eds.), The electronic election (pp. 233-245). Mahwah, NJ: Lawrence Erlbaum Associates.
116
Kaid, L. L., Chanslor, M., & Hovind, M. (1992). The influence of program and commercial type on political advertising effectiveness. Journal of Broadcasting & Electronic Media, 36(3), 303-320. doi: 10.1080/08838159209364179
Kaplan, A. M., & Haenlein, M. (2010). Users of the world, unite! The challenges and opportunities of Social Media. Business horizons, 53(1), 59-68. doi:10.1016/j.bushor.2009.09.003
Katz, E., & Feldman, J. J. (1962). The debates in the light of research: A survey of surveys. In S. Kraus (Eds.), The great debates (pp. 173-223). Bloomington, IN.: Indiana University.
Katz, E., Adoni, H., & Parness, P. (1977). Remembering the news: What the picture adds to recall. Journalism Quarterly, 54, 231-239.
Keating, J. P., & Latane, B. (1976). Politicians on TV: The image is the message. Journal of Social Issues, 32(4), 116-132.
Kelleher, T., & Miller, B. M. (2006). Organizational blogs and the human voice: Relational strategies and relational outcomes. Journal of Computer‐Mediated Communication, 11(2), 395-414. doi: 10.1111/j.1083-6101.2006.00019.x
Kennedy, J. R. (1971). How program environment affects TV commercials. Journal of Advertising Research, 11(1), 33-38.
Kennedy, K. (2013, February 22). Use it or Lose it: Social Media in the 2012 US Election. Pulitzer Center. Retrieved from http://pulitzercenter.org/reporting/social-media-role-young-voters-increase-future-US-elections-Obama-Facebook-Twitter
Kietzmann, J. H., Hermkens, K., McCarthy, I. P., & Silvestre, B. S. (2011). Social media? Get serious! Understanding the functional building blocks of social media. Business horizons, 54(3), 241-251. doi:10.1016/j.bushor.2011.01.005
Kim, D., Chun, H., Kwak, Y., & Nam, Y. (2014). The Employment of Dialogic Principles in Website, Facebook, and Twitter Platforms of Environmental Nonprofit Organizations. Social Science Computer Review, 1-16. doi: 10.1177/0894439314525752
Kim, N. Y., & Sundar, S. S. (2010). Relevance to the rescue: Can “smart ads” reduce negative response to online ad clutter?. Journalism & Mass Communication Quarterly, 87(2), 346-362. doi: 10.1177/107769901008700208
Kim, Y., & Khang, H. (2014). Revisiting civic voluntarism predictors of college students’ political participation in the context of social media. Computers in Human Behavior, 36, 114-121. doi:10.1016/j.chb.2014.03.044
Kiousis, S. (2002). Interactivity: a concept explication. new media & society, 4(3), 355-383.
Kiousis, S. (2006). Exploring the impact of modality on perceptions of credibility for online news stories. Journalism Studies, 7(2), 348-359. doi: 10.1080/14616700500533668
117
Kisielius, J., & Sternthal, B. (1984). Detecting and explaining vividness effects in attitudinal judgments. Journal of Marketing Research, 21(1), 54-64.
Klapper, J. T. (1960). The effects of mass communication. Glencoe, IL: The Free Press.
Koehler, M. J., Yadav, A., Phillips, M., & Cavazos-Kottke, S. (2005). What is video good for? Examining how media and story genre interact. Journal of Educational Multimedia and Hypermedia, 14(3), 249-272.
Kokkinaki, F., & Lunt, P. (1999). The effect of advertising message involvement on brand attitude accessibility. Journal of Economic Psychology, 20(1), 41-51. doi:10.1016/S0167-4870(98)00042-7
Kruikemeier, S., van Noort, G., Vliegenthart, R., & de Vreese, C. H. (2013). Getting closer: The effects of personalized and interactive online political communication. European Journal of Communication, 0(0), 1-14. doi: 10.1177/0267323112464837
Kumar, N., & Benbasat, I. (2002). Para-social presence and communication capabilities of a web site: a theoretical perspective. E-service Journal, 1(3), 5-24.
Kushin, M. J., & Kitchener, K. (2009). Getting political on social network sites: Exploring online political discourse on Facebook. First Monday, 14(11). Retrieved from http://firstmonday.org/ojs/index.php/fm/article/view/2645/2350 doi: http://dx.doi.org/10.5210/fm.v14i11.2645
Lachman, R., Lachman, J. L., & Butterfield, E. C. (1979). Cognitive psychology and information processing; An introduction. Hillsdale, N.J.: Erlbaum.
Lang, A. (1990). Involuntary attention and physiological arousal evoked by structural features and emotional content in TV commercials. Communication Research, 17(3), 275-299. doi: 10.1177/009365090017003001
Lang, A. (2000). The limited capacity model of mediated message processing. Journal of communication, 50(1), 46-70. doi: 10.1111/j.1460-2466.2000.tb02833.x
Lang, A. (2006). Using the limited capacity model of motivated mediated message processing to design effective cancer communication messages. Journal of Communication, 56(1), 57-80. doi: 10.1111/j.1460-2466.2006.00283.x
Lang, A., & Friestad, M. (1993). Emotion, hemispheric specialization, and visual and verbal memory for television messages. Communication Research, 20(5), 647-670. doi: 10.1177/009365093020005002
Lang, A., & Lanfear, P. (1990). The Information Processing of Televised Political Advertising: Using Theory to Maximize Recall. Advances in Consumer Research, 17, 149-158.
118
Lang, A., Borse, J., Wise, K., & David, P. (2002). Captured by the World Wide Web orienting to structural and content features of computer-presented information. Communication Research, 29(3), 215-245. doi: 10.1177/0093650202029003001
Lang, A., Dhillon, K., & Dong, Q. (1995). The effects of emotional arousal and valence on television viewers’ cognitive capacity and memory. Journal of Broadcasting & Electronic Media, 39(3), 313-327. doi: 10.1080/08838159509364309
Lang, A., Newhagen, J., & Reeves, B. (1996). Negative video as structure: Emotion, attention, capacity, and memory. Journal of Broadcasting & Electronic Media, 40(4), 460-477. doi: 10.1080/08838159609364369
Lasswell, H. D. (1948) The Structure and Function of Communication in Society. In L. Bryson (Eds.), The Communication of Ideas (pp. 37-51). New York: IRSS.
Lazarsfeld, P. F., Berelson, B., & Gaudet, H. (1968). The people's choice: how the voter makes up his mind in a presidential campaign. New York: Columbia University Press.
Ledford, C. J. (2012). Changing Channels A Theory-Based Guide to Selecting Traditional, New, and Social Media in Strategic Social Marketing. Social Marketing Quarterly, 18(3), 175-186. doi: 10.1177/1524500412460671
Lee, E.-J. (2013). Effectiveness of Politicians' Soft Campaign on Twitter Versus TV: Cognitive and Experiential Routes. Journal of Communication, 14(7), 1128-1146. doi: 10.1111/jcom.12049
Lee, E. J., & Jang, J. W. (2011). Not so imaginary interpersonal contact with public figures on social network sites: How affiliative tendency moderates its effects. Communication Research, XX(X) 1-25. doi: 10.1177/0093650211431579
Lee, E. J., & Shin, S. Y. (2012). Are they talking to me? Cognitive and affective effects of interactivity in politicians' Twitter communication. Cyberpsychology, Behavior, and Social Networking, 15(10), 515-520. doi: 10.1089/cyber.2012.0228.
Lee, J. G., & Thorson, E. (2009). Cognitive and emotional processes in individuals and commercial web sites. Journal of Business and Psychology, 24(1), 105-115. doi: 10.1007/s10869-008-9087-8
Lee, N. J., Shah, D. V., & McLeod, J. M. (2012). Processes of Political Socialization A Communication Mediation Approach to Youth Civic Engagement. Communication Research, 40(5), 669-697. doi: 10.1177/0093650212436712
Lee, S., & Cho, M. (2011). Social Media Use in a Mobile Broadband Environment. Examination of Determinants of Twitter and Facebook Use. International Journal of Mobile Marketing, 6(2), 71-87.
Lee, Y., Kozar, K. A., & Larsen, K. R. (2009). Avatar e-mail versus traditional e-mail: Perceptual difference and media selection difference. Decision Support Systems, 46(2), 451-467. doi:10.1016/j.dss.2007.11.008
119
Leigh, J. H. (1991). Information processing differences among broadcast media: Review and suggestions for research. Journal of Advertising, 20(2), 71-75. doi: 10.1080/00913367.1991.10673214
Lengel, R. H. (1983). Managerial information processing and media selection behavior. Texas A&M University.
Lengel, R. H., & Daft, R. L. (1984). An exploratory analysis of the relationship between media richness and managerial information processing. Texas A&M University.
Levinson, P. (2003). Digital McLuhan: A guide to the information millennium. New York, NY: Routledge.
Levy, M. R., & Windahl, S. (1984). Audience activity and gratifications: A conceptual clarification and exploration. Communication Research, 11,51-78. doi: 10.1177/009365084011001003
Lim, J. S., Hwang, Y., Kim, S., & Biocca, F. A. (2015). How social media engagement leads to sports channel loyalty: Mediating roles of social presence and channel commitment. Computers in Human Behavior, 46, 158-167. doi:10.1016/j.chb.2015.01.013
Liu, Y., & Shrum, L. J. (2002). What is interactivity and is it always such a good thing? Implications of definition, person, and situation for the influence of interactivity on advertising effectiveness. Journal of advertising, 31(4), 53-64. doi: 10.1080/00913367.2002.10673685
Logan, R. K. (2010). Understanding new media extending Marshall McLuhan. New York, NY: Peter Lang Publishing, Inc.
Lu, Y., Kim, Y., Dou, X. Y., & Kumar, S. (2014). Promote physical activity among college students: Using media richness and interactivity in web design. Computers in Human Behavior, 41, 40-50. doi:10.1016/j.chb.2014.08.012
Lyons, A., Reysen, S., & Pierce, L. (2012). Video lecture format, student technological efficacy, and social presence in online courses. Computers in Human Behavior, 28(1), 181-186. doi:10.1016/j.chb.2011.08.025
Marchese, J. (2008, January 8). Social media’s election effect. Online Spin. Retrieved from http://www.mediapost.com/publications/?fa=Articles.showArticle&art_aid=73844
Martinelli, K. A., & Chaffee, S. H. (1995). Measuring new-voter learning via three channels of political information. Journalism & Mass Communication Quarterly, 72(1), 18-32. doi: 10.1177/107769909507200103
Mayer, R. E., Heiser, J., & Lonn, S. (2001). Cognitive constraints on multimedia learning: When presenting more material results in less understanding. Journal of educational psychology, 93(1), 187-198. doi: http://dx.doi.org/10.1037/0022-0663.93.1.187
Mayfield, T. D. (2011). Commander’s Strategy for Social Media. JFQ, 60(1), 79-83.
120
McClure, R. D., & Patterson, T. E. (1976). Print vs. network news. Journal of Communication, 26(2), 23-28. doi: 10.1111/j.1460-2466.1976.tb01375.x
McKinney, M. S., & Banwart, M. C. (2011). Civic engagement through presidential debates: Young citizens’ attitudes of political engagement throughout the 2008 election. In M. S. McKinney & M. C. Banwart (Eds.), Communication in the 2008 U.S. election: Digital natives elect a president (pp. 1-9). New York, NY: Peter Lang.
McKinney, M.S., & Gaddie, K.C. (2000). The Medium and the Message: TV vs. Internet Viewing of a Presidential Primary Debate. Paper presented at the National Communication Association Convention, Seattle.
McKinnon, L. M., & Tedesco, J. C. (1999). The influence of medium and media commentary on presidential debate effects. In L. L. Kaid & D. G. Bystrom (Eds.), The electronic election: Perspectives on the 1996 campaign communication (pp. 191-206). Mahwah, NJ: Lawrence Erlbaum.
McKinnon, L. M., Tedesco, J. C., & Kaid, L. L. (1993). The third 1992 presidential debate: Channel and commentary effects. Argumentation and Advocacy, 30(2), 106-113.
McLuhan, M. (1964). Understanding media: The extensions of man. New York: McGraw-Hill.
McLuhan, M. (2011). The Gutenberg galaxy: The making of typographic man. Toronto, Buffalo, London: University of Toronto Press.
McMillan, S. J. (2002). A four-part model of cyber-interactivity Some cyber-places are more interactive than others. New Media & Society, 4(2), 271-291. doi: 10.1177/146144480200400208
Medvic, S. K. (2011). New directions in campaigns and elections. New York, NY: Routledge.
Meijer, A., & Thaens, M. (2010). Alignment 2.0: Strategic use of new internet technologies in government. Government Information Quarterly, 27(2), 113-121. doi:10.1016/j.giq.2009.12.001
Meijer, A., & Thaens, M. (2013). Social media strategies: Understanding the differences between North American police departments. Government Information Quarterly, 30(4), 343-350. doi:10.1016/j.giq.2013.05.023
Milbank, D. (1999). Virtual Politics. The New Republic, 5, 22-27.
Milheim, W. D., & Martin, B. L. (1991). Theoretical bases for the use of learner control: Three different perspectives. Journal of Computer-Based Instruction, 18(3), 99-105.
Miller, N. E. (1957). Graphic communication and the crisis in education. Audio Visual Communication Review, 1-120.
Mittal, B. (1987). A framework for relating consumer involvement to lateral brain functioning. Advances in Consumer Research, 14(1), 41-45.
121
Morris, M., & Ogan, C. (1996). The Internet as mass medium. Journal of Computer‐Mediated Communication, 1(4), 0-0. doi: 10.1111/j.1083-6101.1996.tb00174.x
Murthy, D. (2011). Twitter: Microphone for the masses?. Media Culture and Society, 33(5), 779-789. doi: 10.1177/0163443711404744
Nardi, B., & O'Day, V. (1999). Information Ecologies: Using Technology with Heart: Chapter Two: Framing Conversations about Technology. First Monday, 4(5). Retrieved from http://pear.accc.uic.edu/ojs/index.php/fm/article/view/671/581
NPR. (2014, October 16). Political TV Ad Spending Expected To Top $1 Billion. NPR. Retrieved from http://www.npr.org/2014/10/16/356728094/political-tv-ad-spending-expected-to-top-1-billion
O'Reilly, T. (2007). What is Web 2.0: Design patterns and business models for the next generation of software. Communications & strategies, 65(1), 17-37.
Owen, S. (2008, November 3). Citizens, media use social media to monitor election. PBS. Retrieved from http://www.pbs.org/mediashift/2008/11/citizens-media-use-social-mediato-monitor-election308.html
Paivio, A. (1990). Mental representations. New York: Oxford University Press.
Park, C. W., & Young, S. M. (1986). Consumer response to television commercials: The impact of involvement and background music on brand attitude formation. Journal of marketing research, 23(1), 11-24. doi: 10.2307/3151772
Park, H., & Lee, H. (2013). Show Us You Are Real: The Effect of Human-Versus-Organizational Presence on Online Relationship Building Through Social Networking Sites. Cyberpsychology, Behavior, and Social Networking, 16(4), 265-271. doi:10.1089/cyber.2012.0051.
Pavlik, J.V. (1996). New media technology: Cultural and commercial perspectives. Boston, MA: Allyn & Bacon.
Perloff, R. A. (1998). Political communication: Politics, press, and public in America. Mahwah, NJ: Lawrence Erlbaum.
Perse, E. M. (1990a). Audience selectivity and involvement in the newer media environment. Communication Research, 17(5), 675-697. doi: 10.1177/009365090017005005
Perse, E. M. (1990b). Media involvement and local news effects. Journal of Broadcasting & Electronic Media, 34(1), 17-36. doi: 10.1080/08838159009386723
Perse, E. M. (1990c). Involvement with local television news cognitive and emotional dimensions. Human Communication Research, 16(4), 556-581.doi: 10.1111/j.1468-2958.1990.tb00222.x
122
Petty, R. E., & Cacioppo, J. T. (1979). Issue involvement can increase or decrease persuasion by enhancing message-relevant cognitive responses. Journal of personality and social psychology, 37(10), 1915-1926. doi: http://dx.doi.org/10.1037/0022-3514.37.10.1915
Petty, R. E., Cacioppo, J. T., & Goldman, R. (1981). Personal involvement as a determinant of argument-based persuasion. Journal of personality and social psychology, 41(5), 847-855. doi: http://dx.doi.org/10.1037/0022-3514.41.5.847
Petty, R. E., Cacioppo, J. T., & Heesacker, M. (1981). Effects of rhetorical questions on persuasion: A cognitive response analysis. Journal of personality and social psychology, 40(3), 432-440. doi: http://dx.doi.org/10.1037/0022-3514.40.3.432
Petty, R. E., Cacioppo, J. T., & Schumann, D. (1983). Central and peripheral routes to advertising effectiveness: The moderating role of involvement. Journal of consumer research, 10(2), 135-146.
Pfau, M., Holbert, R. L., Szabo, E. A., & Kaminski, K. (2002). Issue‐Advocacy Versus Candidate Advertising: Effects on Candidate Preferences and Democratic Process. Journal of Communication, 52(2), 301-315. doi: 10.1111/j.1460-2466.2002.tb02546.x
Poor, J. (2013). Drum roll: The biggest media impact maker on the 2012 Presidential election was… Local market broadcast television. (again!). TVB Local Media Marketing Solutions. Retrieved from http://www.tvb.org/media/file/Drum-Roll-Political-2013.pdf
Purvis, H. (2001). Media, politics and government. New York: Harcourt Inc.
Putnam, R. D. (1995). Bowling alone: America's declining social capital. Journal of democracy, 6(1), 65-78.
Quercia, D., Kosinski, M., Stillwell, D., & Crowcroft, J. (2011, October). Our Twitter profiles, our selves: Predicting personality with Twitter. In Privacy, Security, Risk and Trust (PASSAT) and 2011 IEEE Third Inernational Conference on Social Computing (SocialCom), 2011 IEEE Third International Conference on (pp. 180-185). IEEE. doi: 978-0-7695-4578-3/11
Rafaeli, S. (1988). Advancing Communication Science: Merging Mass and Interpersonal Processes. In R. P. Hawkinds, J. M. Wiemann, & S. Pingree (Eds.), Interactivity: From new media to communication (pp. 110-134). Beverly Hills, CA: Sage.
Rafaeli, S., & Sudweeks, F. (1997). Networked interactivity. Journal of Computer‐Mediated Communication, 2(4), 0-0. doi: 10.1111/j.1083-6101.1997.tb00201.x
Reese, S. D. (1984). Visual‐verbal redundancy effects on television news learning. Journal of Broadcasting & Electronic Media, 28(1), 79-87. doi: 10.1080/08838158409386516
Reeves, B., Lang, A., Thorson, E., & Rothschild, M. (1988). Hemispheric lateralization and the processing of emotional television scenes. Human Communication Research, 15, 493-508.
123
Rice, R. E. (1992). Task analyzability, use of new media, and effectiveness: A multi-site exploration of media richness. Organization science, 3(4), 475-500. doi: http://dx.doi.org/10.1287/orsc.3.4.475
Richardson, J. C., & Swan, K. (2003). Examining social presence in online courses in relation to students' perceived learning and satisfaction. JALN, 7(1), 68-88.
Ridout, T. N., Shah, D. V., Goldstein, K. M., & Franz, M. M. (2004). Evaluating measures of campaign advertising exposure on political learning. Political Behavior, 26(3), 201-225. doi: 10.1023/B:POBE.0000043453.96025.a8
Rogers, E. M. (1995). Discussion of innovations (4th ed.). NY: Free Press.
Rothschild, M. L., & Ray, M. L. (1974). Involvement and political advertising effect An exploratory experiment. Communication Research, 1(3), 264-285. doi: 10.1177/009365027400100302
Rowley, J. (2004). Partnering paradigms? Knowledge management and relationship marketing. Industrial Management & Data Systems, 104(2), 149-157.
Rubin, A. M., & Perse, E. M. (1987). Audience activity and soap opera involvement a uses and effects investigation. Human Communication Research, 14(2), 246-268. doi: 10.1111/j.1468-2958.1987.tb00129.x Russell, C. A. (2002). Investigating the effectiveness of product placements in television shows: The role of modality and plot connection congruence on brand memory and attitude. Journal of consumer research, 29(3), 306-318. doi: 10.1086/344432
Rutledge, P. (2013, January 25). How Obama Won the Social Media Battle in the 2012 Presidential Campaign. The Media Psychology Blog. Retrieved from http://mprcenter.org/blog/2013/01/how-obama-won-the-social-media-battle-in-the-2012-presidential-campaign/
Sadoski, M., Goetz, E. T., & Rodriguez, M. (2000). Engaging texts: Effects of concreteness on comprehensibility, interest, and recall in four text types. Journal of Educational Psychology, 92(1), 85-95. doi: http://dx.doi.org/10.1037/0022-0663.92.1.85
Schiffman, L. G., & Kanuk, L. L. (1997). Consumer Behavior (6th ed). NJ: Prentice Hall.
Schleuder, J. (1990). Effects of Commercial Complexity, the Candidates, and Issue vs. Image Strategies in Political Ads. Advances in Consumer Research, 17, 159-168.
Schönbach, K., de Waal, E., & Lauf, E. (2005). Research note: Online and print newspapers their impact on the extent of the perceived public agenda. European Journal of Communication, 20(2), 245-258. doi: 10.1177/0267323105052300
124
Schultz, F., Utz, S., & Göritz, A. (2011). Is the medium the message? Perceptions of and reactions to crisis communication via twitter, blogs and traditional media. Public relations review, 37(1), 20-27. doi:10.1016/j.pubrev.2010.12.001
Severin, W. (1967). Another look at cue summation. AV Communication Review, 15(3), 233-245. doi: 10.1007/BF02768608
Sevin, E., Kimball, S., & Khalil, M. (2011). Listening to President Obama: A Short Examination of Obama's Communication Practices. American Behavioral Scientist, 55(6), 803-812. doi: 10.1177/0002764211406847
Shah, D. V., Cho, J., Eveland, W. P., & Kwak, N. (2005). Information and expression in a digital age modeling Internet effects on civic participation. Communication research, 32(5), 531-565. doi: 10.1177/0093650205279209
Shah, D. V., Cho, J., Nah, S., Gotlieb, M. R., Hwang, H., Lee, N. J., ... & McLeod, D. M. (2007). Campaign ads, online messaging, and participation: Extending the communication mediation model. Journal of Communication, 57(4), 676-703. doi: 10.1111/j.1460-2466.2007.00363.x
Shannon, C. E., & Weaver, W. (1949). The mathematical theory of communication. Urbana, IL: University of Illinois Press.
Shao, C. Y., Baker, J. A., & Wagner, J. (2004). The effects of appropriateness of service contact personnel dress on customer expectations of service quality and purchase intention: The moderating influences of involvement and gender. Journal of Business Research, 57(10), 1164-1176. doi:10.1016/S0148-2963(02)00326-0
Sheer, V. C. (2011). Teenagers' use of MSN features, discussion topics, and online friendship development: The impact of media richness and communication control. Communication Quarterly, 59(1), 82-103. doi: 10.1080/01463373.2010.525702
Shen, F., & Wu, H. D. (2002). Effects of soft-money issue advertisements on candidate evaluation and voting preference: An exploration. Mass Communication & Society, 5(4), 395-410. doi: 10.1207/S15327825MCS0504_02
Shogan, C. J. (2010). Blackberries, tweets, and YouTube: Technology and the future of communicating with Congress. Political Science & Politics, 43(2), 231-233. doi: http://dx.doi.org/10.1017/S1049096510000041
Short, J., Williams, E., & Christie, B. (1976). The social psychology of telecommunications. London: Wiley.
Sicilia, M., Ruiz, S., & Munuera, J. L. (2005). Effects of interactivity in a web site: The moderating effect of need for cognition. Journal of Advertising, 34(3), 31-44.
Simon, S. J., & Peppas, S. C. (2004). An examination of media richness theory in product Web site design: an empirical study. info, 6(4), 270-281. doi: http://dx.doi.org/10.1108/14636690410555672
125
Skalski, P., & Tamborini, R. (2007). The role of social presence in interactive agent-based persuasion. Media psychology, 10(3), 385-413. doi: 10.1080/15213260701533102
Snyder, C. (2014, November 4). Social media is make or break for midterm candidates. Fox news. Retrieved from http://www.foxnews.com/tech/2014/11/04/social-media-is-make-or-break-for-midterm-candidates/
Snyder, L. (1996). Integrating computers into the literacy curriculum: More difficult than we first imagined. Australian Journal of Language and Literacy, 19(4), 330-344.
Sojka, J. Z., & Giese, J. L. (1997). Thinking and/or feeling: an examination of interaction between processing styles. Advances in Consumer Research, 24, 438-442.
Soldow, G. F., & Principe, V. (1981). Response to commercials as a function of program context. Journal of Advertising Research, 21(2), 59-65.
Sood, S. (2002). Audience involvement and entertainment—education. Communication Theory, 12(2), 153-172. doi: 10.1111/j.1468-2885.2002.tb00264.x
Steuer, J. (1992). Defining virtual reality: Dimensions determining telepresence. Journal of communication, 42(4), 73-93. doi: 10.1111/j.1460-2466.1992.tb00812.x
Sulfaro, V. A. (2001). Political advertisements and decision-making shortcuts in the 2000 election. Contemporary Argumentation and Debate, 22, 80-99.
Sun, S., Rubin, A. M., & Haridakis, P. M. (2008). The role of motivation and media involvement in explaining Internet dependency. Journal of Broadcasting & Electronic Media, 52(3), 408-431. doi: 10.1080/08838150802205595
Sundar, S. S. (2000). Multimedia effects on processing and perception of online news: A study of picture, audio, and video downloads. Journalism & Mass Communication Quarterly, 77(3), 480-499. doi: 10.1177/107769900007700302
Sundar, S. S. (2004). Theorizing Interactivity's Effects. The Information Society, 20(5), 385-389. doi: 10.1080/01972240490508072
Sundar, S. S., & Kim, J. (2005). Interactivity and persuasion: Influencing attitudes with information and involvement. Journal of interactive advertising, 5(2), 5-18. doi: 10.1080/15252019.2005.10722097
Sundar, S. S., Hesser, K. M., Kalyanaraman, S., & Brown, J. (1998). The Effect of Website Interactivity on Political Persuasion. Paper presented at the meeting of the International Association for Media and Communication Research, Glasgow, UK.
Sundar, S. S., Kalyanaraman, S., & Brown, J. (2003). Explicating Web Site interactivity impression formation effects in political campaign sites. Communication research, 30(1), 30-59. doi: 10.1177/0093650202239025
126
Sundar, S. S., Narayan, S., Obregon, R., & Uppal, C. (1998). Does Web Advertising Work? Memory for Print vs. Online Media. Journalism & Mass Communication Quarterly, 75(4), 822-835. doi: 10.1177/107769909807500414
Te’eni-Harari, T., Lehman-Wilzig, S. N., & Lampert, S. I. (2009). The importance of product involvement for predicting advertising effectiveness among young people. International Journal of Advertising, 28(2), 203-229. doi: 10.2501/S0265048709200540
Tedesco, J. C., McKinnon, L. M., & Kaid, L. L. (1996). Advertising Watchdogs: A Content Analysis of Print and Broadcast Ad Watches. The Harvard International Journal of Press/Politics, 1(4), 76-93. doi: 10.1177/1081180x96001004007
Tewksbury, D. (2003). What do Americans really want to know? Tracking the behavior of news readers on the Internet. Journal of Communication, 53(4), 694-710. doi: 10.1111/j.1460-2466.2003.tb02918.x
The Associate Press. (2013, August 17). Election 2016: A look at possible candidates and who could run in the next presidential race. Daily News. Retrieved from http://www.nydailynews.com/news/politics/election-2016-closer-candidates-article-1.1429652
Thorson, E., & Friestad, M. (1989). The Effects of Emotion on Episodic Memory for Television Commercials. In P. Cafferata & A. M. Tybout (Eds.), Cognitive and Affective Responses to Advertising (pp. 305-325). Lexington, MA: D.C. Heath and Company.
Thorson, E., Christ, W. G., & Caywood, C. (1991). Effects of issue‐image strategies, attack and support appeals, music, and visual content in political commercials. Journal of Broadcasting & Electronic Media, 35(4), 465-486. doi: 10.1080/08838159109364141
Trent, J. S., & Friednberg, R. V. (2008). Political campaign communication: principles & Practices. (6th ed.). Maryland: Rowman & Littlefield Publishers, Inc.
Triesman, A. M. (1964). Selective attention in man. British Medical Bulletin, 20(1), 12-16.
Tsukayama, H. (2012, November 7). Obama victory tweet becomes his most-retweeted ever. The Washington Post. Retrieved from http://www.washingtonpost.com/business/technology/obama-victory-tweet-becomes-his-most-retweeted-ever/2012/11/07/47793922-28f0-11e2-96b6-8e6a7524553f_story.html
Unnava, H. R., Agarwal, S., & Haugtvedt, C. P. (1996). Interactive effects of presentation modality and message-generated imagery on recall of advertising information. Journal of Consumer Research, 23(1), 81-88.
Unnava, H. R., Burnkrant, R. E., & Erevelles, S. (1994). Effects of presentation order and communication modality on recall and attitude. Journal of Consumer Research, 21(3), 481-490.
Valentino, N. A., Hutchings, V. L., & Williams, D. (2004). The impact of political advertising on knowledge, Internet information seeking, and candidate preference. Journal of Communication, 54(2), 337-354. doi: 10.1111/j.1460-2466.2004.tb02632.x
127
Van Cauwenberge, A., Schaap, G., & Van Roy, R. (2014). “TV no longer commands our full attention”: Effects of second-screen viewing and task relevance on cognitive load and learning from news. Computers in Human Behavior, 38, 100-109. doi:10.1016/j.chb.2014.05.021
van Koten, H. (2009). The digital image and the pleasure principle: The consumption of realism in the age of stimulation. In A. Bentkowska-Kafel, T. Cashen & H. Gardiner (Eds.), Digital visual culture: Theory and practice (Vol. 3). Chicago, IL: The University of Chicago Press.
Vraga, E. K., Edgerly, S., Wang, B. M., & Shah, D. V. (2011). Who taught me that? Repurposed news, blog structure, and source identification. Journal of Communication, 61(5), 795-815. doi: 10.1111/j.1460-2466.2011.01581.x
Walma van der Molen, J. H., & van der Voort, T. H. A. (2000). Children’s and adults’ recall of television and print news in children’s and adult news formats. Communication Research, 27, 132-160. doi: 10.1177/009365000027002002
Warnick, B., Xenos, M., Endres, D., & Gastil, J. (2005). Effects of Campaign‐to‐User and Text‐Based Interactivity in Political Candidate Campaign Web sites. Journal of Computer‐Mediated Communication, 10(3), 00-00. doi: 10.1111/j.1083-6101.2005.tb00253.x
West, D. M. (1993). Air wars: Television advertising in election campaigns, 1952-1992. Washington, DC: Congressional Quarterly Press.
West, D. M. (1994). Political advertising and news coverage in the 1992 California US Senate campaigns. The Journal of Politics, 56(4), 1053-1075. doi: http://dx.doi.org/10.2307/2132073
Westley, B. H., & Barrow, L. C. (1959). An investigation of news-seeking behavior. Journalism & Mass Communication Quarterly, 36(4), 431-438. doi: 10.1177/107769905903600405
Wickens, C. D. (1981). Processing Resources in Attention, Dual Task Performance, and Workload Assessment. Champaign, IL: University of Illinois.
Wickens, C. D. (2002). Multiple resources and performance prediction. Theoretical issues in ergonomics science, 3(2), 159-177. doi: 10.1080/14639220210123806
Wickens, C.D. (1980). The structure of attentional resources. In R. Nickerson (Eds.), Attention and Performance VIII. Hillsdale, NJ: Erlbaum.
Wicks, R. H. (2006). Media information processing. In J. Bryant & P. Vorderer (Eds.), Psychology of entertainment (pp. 85-102). New York: Routledge.
Williams, R., & Edge, D. (1996). The social shaping of technology. Research policy, 25(6), 865-899. doi:10.1016/0048-7333(96)00885-2
Woolley, J. K., Limperos, A. M., & Oliver, M. B. (2010). The 2008 presidential election, 2.0: A content analysis of user-generated political Facebook groups. Mass Communication and Society, 13(5), 631-652. doi: 10.1080/15205436.2010.516864
128
Wu, G. (2005). The mediating role of perceived interactivity in the effect of actual interactivity on attitude toward the website. Journal of Interactive advertising, 5(2), 29-39. doi: 10.1080/15252019.2005.10722099
Yadav, A., Phillips, M. M., Lundeberg, M. A., Koehler, M. J., Hilden, K., & Dirkin, K. H. (2011). If a picture is worth a thousand words is video worth a million? Differences in affective and cognitive processing of video and text cases. Journal of Computing in Higher Education, 23(1), 15-37. doi: 10.1007/s12528-011-9042-y
Zaichkowsky, J. L. (1985). Measuring the involvement construct. Journal of consumer research, 12(3), 341-352.
Zhang, W., Seltzer, T., & Bichard, S. L. (2013). Two sides of the coin assessing the influence of social network site use during the 2012 US presidential campaign. Social Science Computer Review, 31(5), 542-551. doi: 10.1177/0894439313489962
129
APPENDIX A
Institutional Review Board (IRB) Approval Form
130
131
132
APPENDIX B
Survey Questionnaire
[Study Information]
Julie Kim, Principal Investigator from the University of Alabama, is conducting a study called “a political campaign study.” She wishes to find out your views on political issues.
Taking part in this study involves completing a web survey that will take about 15 minutes. This survey contains questions about a potential voter's perspective on the current U.S. election.
No individual information or identifying information that could trace your answers back to you will ever be reported, and your answers will be kept strictly confidential. Only the researchers have access to the data collected from this research. The surveys themselves do not ask for your name or any clearly identifiable information. Data such as age, sex, and race are used only to provide general descriptions of the data and never to provide the answers of any particular participant. Only summarized data will be presented at meetings or in publications.
Although benefits cannot be promised in research, it is possible/likely that you will find a few moments to reflect on your views on a political campaign. This may provide you with an opportunity to assess your political attitude and behaviors.
As researchers, we do not foresee any risks. Remember, though, it is perfectly within your rights to stop taking the survey if you do not feel comfortable with the content.
When you finish this survey, Amazon Mechanical Turk System will offer you $1 for your participation. If you choose not to participate in this study, however, you can stop at any time.
If you have questions regarding this study, please email the investigator Ms. Julie Kim at [email protected]. If you have questions about your rights as a person taking part in a research study, or you wish to make suggestions or file complaints and concerns, you may call Ms. Tanta Myles, the Research Compliance Officer of the University at (205)-348-8461 or toll-free at 1-877- 820-3066. You may also ask questions, make suggestions, or file complaints and concerns through the IRB Outreach Website at http://osp.ua.edu/site/PRCO_Welcome.html. You may email us at [email protected]. This helps UA improve its protection of human research participants.
YOUR PARTICIPATION IS COMPLETELY VOLUNTARY. You are free not to participate or stop participating any time before you submit your answers.
*I understand the information sheet and wish to continue
133
YES, I AGREE TO PARTICIPATE.
NO, I DECIDED NOT TO PARTICIPATE.
[Familiarity with the Kentucky Gubernatorial Election]
Q. How familiar are you with the Kentucky gubernatorial election?
Very unfamiliar
Unfamiliar
Somewhat familiar
Familiar
Very familiar
[Familiarity with the Candidate]
Q. How familiar are you with the candidate, Hal Heiner?
Very unfamiliar
Unfamiliar
Somewhat familiar
Familiar
Very familiar
[Political Orientation]
Q. Please mark which best identifies your political views.
Very Liberal
Liberal
Moderate
Conservative
Very Conservative
[Partisanship]
134
Q. Generally speaking, do you usually think of yourself as a Republican, a Democrat, an
Independent, or what?
Democrat
Independent
Republican
Other
Do not know
[Political Participation]
Q. Have you ever done any of the following activities within the last 12 months?
(I have ...................)
Never Yes, but not within the last 12 months
Yes, within the last 12 months
Worked or volunteered for political groups or candidates
Called or sent a letter to an elected public official
Contacted a public official in person
Attended a political rally or political meetings (e.g., board or council)
Participated in any demonstrations, protests, or marches on some national or local issue
Voted in the 2012 presidential election
Participated in organizations or groups that took any local action for social or political reform
Been involved in political groups or candidates in person
Donated money to political groups or candidates
135
[Cognitive Involvement with Facebook]
Q. How much do you agree or disagree with each of the following statements? 1
Strongly disagree
2 3 4 5
Strongly agree
I pay close attention when I use Facebook.
I carefully read the content when I use Facebook.
I pay close attention to the content when I use Facebook.
I miss parts of the content when I use Facebook.
I put a lot of mental effort into using Facebook.
I think about what the content means to me and my family when I use Facebook.
I think about how the content is related to other things I know when I use Facebook.
I think about what the content means to other people when I use Facebook.
I think about the content over and over again when I use Facebook.
I think about what should be done when I use Facebook.
I understand Facebook.
I recognize the places of content on Facebook.
I recognize the names of content on
136
Facebook.
I get confused by the content on Facebook.
The content seems too complicated to understand on Facebook.
[Cognitive Involvement with Twitter]
Q. How much do you agree or disagree with each of the following statements? 1
Strongly disagree
2 3 4 5
Strongly agree
I pay close attention when I use Twitter.
I carefully read the content when I use Twitter.
I pay close attention to the content when I use Twitter.
I miss parts of the content when I use Twitter.
I put a lot of mental effort into using Twitter.
I think about what the content means to me and my family when I use Twitter.
I think about how the content is related to other things I know when I use Twitter.
I think about what the content means to other people when I use Twitter.
I think about the content over and over again when I use Twitter.
137
I think about what should be done when I use Twitter.
I understand Twitter.
I recognize the places of content on Twitter.
I recognize the names of content on Twitter.
I get confused by the content on Twitter.
The content seems too complicated to understand on Twitter.
[Emotional Involvement with Facebook]
Q. While using Facebook, how much do you agree or disagree with each of the following
statements?
1
Strongly disagree
2 3 4 5
Strongly agree
I feel a warm inner glow.
I feel like smiling.
I feel a warm excitement.
I feel happy.
I feel like laughing.
I feel my blood pressure go up.
I feel like clenching my fists.
I feel like clenching my teeth.
I feel my face get tight, tense, and hard.
138
I get angry.
I feel a lump in my throat.
I feel myself getting choked up.
My heart seems to ache.
Tears come to my eyes.
I feel sad.
[Emotional Involvement with Twitter]
Q. While using Twitter, how much do you agree or disagree with each of the following
statements?
1
Strongly disagree
2 3 4 5
Strongly agree
I feel a warm inner glow.
I feel like smiling.
I feel a warm excitement.
I feel happy.
I feel like laughing.
I feel my blood pressure go up.
I feel like clenching my fists.
I feel like clenching my teeth.
I feel my face get tight, tense, and hard.
I get angry.
I feel a lump in my throat.
139
I feel myself getting choked up.
My heart seems to ache.
Tears come to my eyes.
I feel sad.
[Stimuli for Visual Advertisements]
Q. Now, you will see a political advertisement related to the current gubernatorial election.
Please watch this advertising carefully.
[Stimuli for Audiovisual Advertisements]
Q. Now, you will see a political advertisement related to the current gubernatorial election.
Please adjust the speaker volume, then click the following link and watch this advertising
carefully.
[Visual Ad with High Interactivity on Facebook]
140
This question lets you record and manage how long a participant spends on this page. This question will not be displayed to the participant.
00
141
This question lets you record and manage how long a participant spends on this page. This question will not be displayed to the participant.
00
142
This question lets you record and manage how long a participant spends on this page. This question will not be displayed to the participant.
00
143
This question lets you record and manage how long a participant spends on this page. This question will not be displayed to the participant.
00
144
This question lets you record and manage how long a participant spends on this page. This question will not be displayed to the participant.
00
145
This question lets you record and manage how long a participant spends on this page. This question will not be displayed to the participant.
00
146
This question lets you record and manage how long a participant spends on this page. This question will not be displayed to the participant.
00
147
This question lets you record and manage how long a participant spends on this page. This question will not be displayed to the participant.
00
148
[Visual Ad with Low Interactivity on Facebook]
This question lets you record and manage how long a participant spends on this page. This question will not be displayed to the participant.
00
149
This question lets you record and manage how long a participant spends on this page. This question will not be displayed to the participant.
00
150
This question lets you record and manage how long a participant spends on this page. This question will not be displayed to the participant.
00
151
This question lets you record and manage how long a participant spends on this page. This question will not be displayed to the participant.
00
152
This question lets you record and manage how long a participant spends on this page. This question will not be displayed to the participant.
00
153
This question lets you record and manage how long a participant spends on this page. This question will not be displayed to the participant.
00
154
This question lets you record and manage how long a participant spends on this page. This question will not be displayed to the participant.
00
155
This question lets you record and manage how long a participant spends on this page. This question will not be displayed to the participant.
00
156
[Visual Ad with High Interactivity on Twitter]
This question lets you record and manage how long a participant spends on this page. This question will not be displayed to the participant.
00
157
This question lets you record and manage how long a participant spends on this page. This question will not be displayed to the participant.
00
158
This question lets you record and manage how long a participant spends on this page. This question will not be displayed to the participant.
00
159
This question lets you record and manage how long a participant spends on this page. This question will not be displayed to the participant.
00
160
This question lets you record and manage how long a participant spends on this page. This question will not be displayed to the participant.
00
161
This question lets you record and manage how long a participant spends on this page. This question will not be displayed to the participant.
00
162
This question lets you record and manage how long a participant spends on this page. This question will not be displayed to the participant.
00
163
This question lets you record and manage how long a participant spends on this page. This question will not be displayed to the participant.
00
164
[Visual Ad with Low Interactivity on Twitter]
This question lets you record and manage how long a participant spends on this page. This question will not be displayed to the participant.
00
165
This question lets you record and manage how long a participant spends on this page. This question will not be displayed to the participant.
00
166
This question lets you record and manage how long a participant spends on this page. This question will not be displayed to the participant.
00
167
This question lets you record and manage how long a participant spends on this page. This question will not be displayed to the participant.
00
168
This question lets you record and manage how long a participant spends on this page. This question will not be displayed to the participant.
00
169
This question lets you record and manage how long a participant spends on this page. This question will not be displayed to the participant.
00
170
This question lets you record and manage how long a participant spends on this page. This question will not be displayed to the participant.
00
171
This question lets you record and manage how long a participant spends on this page. This question will not be displayed to the participant.
00
172
[Audio-visual Ad with High Interactivity on Facebook]
http://fille222.wix.com/halheiner1
173
[Audio-visual Ad with Low Interactivity on Facebook]
http://fille222.wix.com/halheiner2
174
[Audio-visual Ad with High Interactivity on Twitter]
http://fille222.wix.com/halheiner4
175
[Audio-visual Ad with Low Interactivity on Twitter]
http://fille222.wix.com/halheiner3
176
[Perceived Interactivity of Facebook]
Q. How much do you agree or disagree with each of the following statements?
Strongly disagree Disagree
Neither Agree nor Disagree Agree
Strongly Agree
Interacting with Facebook was like having a conversation with a representative of the campaign.
I felt as if the candidate's Facebook talked back to me while I was navigating.
I perceive the candidate's Facebook to be sensitive to my needs for information.
The candidate's Facebook is interpersonal.
I could communicate directly to the candidate if I wanted to using the candidate's Facebook.
The candidate's Facebook is interactive.
The candidate's Facebook had the ability to respond to my specific questions quickly and efficiently.
It is easy to find my way through the candidate’s Facebook page.
The candidate’s Facebook has a variety of information.
177
[Perceived Interactivity of Twitter]
Q. How much do you agree or disagree with each of the following statements?
Strongly disagree Disagree
Neither Agree nor Disagree Agree
Strongly Agree
Interacting with Twitter was like having a conversation with a representative of the campaign.
I felt as if the candidate's Twitter talked back to me while I was navigating.
I perceive the candidate's Twitter to be sensitive to my needs for information.
The candidate's Twitter is interpersonal.
I could communicate directly to the candidate if I wanted to using the candidate's Twitter.
The candidate's Twitter is interactive.
The candidate's Twitter had the ability to respond to my specific questions quickly and efficiently.
It is easy to find my way through the candidate’s Twitter page.
The candidate’s Twitter has a variety of information.
178
[Campaign Learning]
Q. Please mark whether each of the statements is correct, incorrect, or "do not know".
Incorrect Correct “Do not know”
The candidate favors Obama's policy.
The candidate supports drug testing for welfare recipients.
The candidate has a plan to increase Kentucky jobs.
The candidate suggests making verify immigration screening.
The candidate is the "Frankfort outsider."
The candidate is a candidate for Governor of Kentucky.
The candidate is liberal.
The candidate was not a successful businessman.
[Candidate Evaluation toward the Candidate]
Q. How do you rate your feelings toward Hal Heiner?
1= Unfavorable (Cold),
5 = Favorable (Warm)
1 2 3 4 5
179
[Candidate Evaluation toward the Campaign Issues]
Q. How do you rate your feelings toward the campaign issues endorsed by Hal Heiner?
1= Unfavorable (Cold),
5 = Favorable (Warm)
1 2 3 4 5
[Candidate Evaluation toward the Party]
Q. How do you rate your feelings toward the party to which Hal Heiner belongs?
1= Unfavorable (Cold),
5 = Favorable (Warm)
1 2 3 4 5
[Voting Intention]
Q. Assuming you were a resident of Kentucky, if the election were held tomorrow, how likely would you be to vote for the candidate in the advertising you just watched?
Very unlikely
Unlikely
Undecided
Likely
Very likely
180
[Information Seeking Behaviors]
Q. How likely or unlikely would you be engage in each of the following activities?
1
Very unlikely
2 3 4 5
Very likely
Watch more ads
Watch news about the candidate
Volunteer to work in the candidate's campaign
Participate in online chat about the candidate
Talk with friends about the candidate
Read newspaper articles about the candidate
Contact the candidate directly
Use the Internet for general campaign information about the candidate
Vote in the next election
Contribute money to the candidate's campaign
Use the internet to get more information on issues of the candidate
Go to the candidate's Web page
[Gender]
Q. What is your gender?
181
Male
Female
[Age]
Q. How old are you?
[Education]
Q. What is the highest grade of regular school that you have completed and gotten credit for?
Grammar and Less
Some HS
HS Grad/GED
Some College
College Grad
Some Grad Work
Master's
PhD/Prof
[Race and/or Ethnicity]
Q. Please identify your race and/or ethnicity.
African American or Black
Asian or Asian American
Latino or Hispanic
Pacific Islander or Native Hawaiian
White-Non Hispanic
182
Biracial (any combination or the above)
Other
[Income]
Q. What is your household's level of income?
Less than $20,000
$20,000 to $39,999
$40,000 to $59,999
$60,000 to $79,999
$80,000 to $99,999
$100,000+
[Region]
Which region of the country do you live in?
Northeast
Mid-Atlantic
North
South
Midwest
West
I do not live in the U.S.
[Debriefing Statement]
Thank you for participating in this research. Your responses may be used to understand potential voters' perspective on the current U.S. election.
The political advertising you watched during the survey was used to examine how voters respond to different communication channels in political advertisements in social media context. The advertising was edited to remind you of campaign information for the Kentucky Gubernatorial
183
Election. This study is designed to investigate under what conditions (print-based ad on Facebook, audiovisual ad on Facebook, print-based ad on Twitter, and audiovisual ad on Twitter), issue ads are most effective in 1) individuals' learning campaign information, 2) positive candidate evaluation, and 3) voting intention. For this purpose, you were randomly assigned to watch one of four conditions to answer the questions. Results from this research will be useful to academic development in social science and your responses will be protected.
You can withdraw your answers if you do not want to participate in this study by checking the box saying “please withdraw my data from this study.” Researchers must destroy your answers as soon as possible and it must not be included in the final research analysis.
If you have any questions or concerns about any aspect of this research, please contact the research director, Ms. Julie Kim, a PhD candidate, [email protected], (205) 765-3519.
Thank you for your participation.
Comment:
[Withdrawal of Data]
Q. Please withdraw my data from this study
Yes
No