Seeing, Listening, Reading: Comparing Campaign Communication
Channels in the 2002 Florida Gubernatorial Election
By
Lynda Lee Kaid
Andrew Paul Williams
Kaye D. Trammell
College of Journalism and Communications
University of Florida
Gainesville, FL 32611
352-392-6557
[log in to unmask]
Paper submitted to the Radio-Television Journalism Division of Association
for Education in Journalism and Mass Communication Conference, Kansas City,
July-August, 2003
Kaid is Professor and Senior Associate Dean, College of Journalism and
Communications
University of Florida and may be contacted at PO Box 118400, Gainesville,
FL 32611-8400, Telephone: 352-392-6557, email: [log in to unmask]
Williams is a Ph.D. Student, College of Journalism and Communications
University of Florida and may be contacted at the PO Box 118400,
Gainesville, FL 32611-8400, Telephone: 352-392-6557, email:
[log in to unmask]
Trammell is a Ph.D. Student, College of Journalism and Communications
University of Florida and may be contacted at PO Box 118400, Gainesville,
FL 32611-8400, Telephone: 352-392-6557, email: [log in to unmask]
The authors would like to express appreciation to Kristen Landreville,
Susan Floyd, Laura Rowe, Drew Bagley, Meredith Barrios, Spiro Kiousis,
Daniela Dimitrova, Heather French and other members of the 2002 Election
Research Team at the University of Florida for their assistance with
various aspects of this project.
Please direct all correspondence to the first author.
Abstract
This experiment investigated debate affect based on exposure channels (TV,
radio, Internet video, or Internet text) among respondents (n = 136)
through a pretest/posttest design. Debate exposure across channels
resulted in significant positive change for one candidate and negative
change for the opponent, significant decline in feelings of political
cynicism, limited support for change in the levels of voter information
seeking and participation across channels, and no influence on viewer
agenda of issues across channels.
Seeing, Listening, Reading: Comparing Campaign Communication
Channels in the 2002 Florida Gubernatorial Election
The race for the Florida governor's house was heated in 2002. Jeb Bush,
the incumbent, and the current President's brother, fought an uphill battle
by most media accounts. Bush had several points against him, including
backlash from the 2000 election voting debacle in the state, the economy,
the voter problems that recurred in the 2002 primary election, various
children's welfare issues in the state, attacks by Rosie O'Donnell
regarding Bush's stance on gay adoptions rights, and the historical record
that no Florida Republican had ever won re-election in a gubernatorial
election. Indeed, just as the Republican Party spent a record-breaking
amount of money on the Florida governor's campaign, there was equal support
from the Democrats in favor of their party's challenger, Bill McBride
(Walsh, 2002).
McBride, a politically unknown lawyer from Tampa, Florida, became the
unlikely opponent when he upset Janet Reno in the primary. It seemed Reno,
the former US Attorney General, who ran her campaign from her "Red Truck
Tour," traveling across the state and speaking with voters, failed to get
her Red Truck over the bridge of the primary and was forced to concede to
McBride (Floyd, Trammell, & Williams, 2002). Throughout the race, McBride
was criticized for his lack of specific plans but ran neck in neck with
Bush in several pre-election polls.
While both Bush and McBride relied on the usual arsenal of campaign media,
the televised debate encounters between the two candidates became a
decisive factor in the election (Davis, 2002). Research on these important
state debates, which were available throughout the Florida on radio,
television, and media Web outlets, has shown that Bush was able to improve
his overall image and percentage of vote choice as a result of his
performance in these media events (Trammell, Kaid, Williams, & Landreville,
2003). With the growing significance of media debates in state races and
the increasing availability of debates to viewers on multiple channels, the
research reported here sought to evaluate the potential differences in
effects resulting from exposure to debates via differing communication
channels, comparing exposure via radio, television, Internet video, and
Internet text.
Theoretical Perspectives
Political Debate Effects
Although candidates have a variety of venues to get their message out to
the public (Weaver & Drew, 1995), the media find debates more attractive
than single candidate campaign stops (Kraus, 1996; Sears & Chaffee,
1979). Despite the campaign planning and prepping, debates remain one of
the most spontaneous campaign events, offering media outlets and the public
a chance to catch a "real glimpse" of the candidates and opening the
possibility that something important or new can happen in a live,
unscripted encounter between two opponents (Benoit, McKinney, & Stephenson,
2002).
Although debate exposure at all levels is much less universal than in the
days of the famous Kennedy-Nixon debate, when limited channel choice
guaranteed high viewership, debates are still seen by more voters than any
other single campaign event (Carlin & McKinney, 1994). Consequently,
researchers have given a great deal of attention to this civic forum and
have studied debates, primarily at the presidential level, from a variety
of theoretical viewpoints (Benoit, McKinney, & Holbert, 2001; Katz &
Feldman, 1962; Kendall, 1997; Meadow & Beeck, 1978; Roberts & Williams,
2003; Sears & Chaffee, 1979; Trent & Friedenberg, 1995).
Much of the early debate research was approached from a direct effects
perspective, examining the effects of debates on candidate images and on
vote choice (Katz & Feldman, 1962). Other researchers have focused on
what voters learn about campaign issues and the significant reinforcement
effects on voters of participating in and observing rituals that heighten
democratic traditions (Drew & Weaver, 1991; McKinney & Lamoureux, 1999).
Research has even found that high scores on free-recall questions may
persist as long as two weeks after a debate (Hullet & Louden, 1998),
suggesting that debates may have long-lasting effects on viewers. McKinney,
Kaid, and Robertson (2001) note three reasons debates are useful resources
for voters in that debates are: useful for those who have not followed the
campaign closely, reinforce existing attitudes for those familiar with the
campaign, and increase voter interest in the ongoing campaign.
Since the seminal agenda-setting study from McCombs and Shaw (1972), the
hypothesis that the media doesn't tell you what to think, but what to think
about (Cohen, 1963), has been tested in numerous studies, across a wide
variety of media channels (McCombs & Shaw, 1993). Indeed, 25 years later
McCombs and Shaw (1993) sought to identify contingent conditions that
either enhance or limit agenda setting. Although not addressed in that
specific article, one such circumstance in which to test agenda-setting can
be the political stage of debates.
The impact of issue importance as a result of exposure to debates has been
researched along various levels of office, including presidential debates
(Jackson-Breek & Meadow, 1979) congressional debates (Hullett & Louden,
1998), and presently in regards to gubernatorial debates. A historical
reflection of the agenda-setting hypothesis in regards to political debates
finds less support for agenda-setting than other, more traditional
agenda-setting studies.
For example, Jackson-Breek and Meadow (1979) looked for evidence of the
agenda setting hypothesis in the 1960 and 1976 presidential debates. The
researchers found a disconnect between the goals of the debate for
journalists, candidates, and the public. Interestingly, the study found
evidence that the public's agenda is not truly addressed during the
debate. They note that journalists ask questions that don't represent the
public agenda and candidate answers only bring up new issues themselves and
barely answer the journalist questions.
With this model, the researchers asked the obvious questions of "for whom
are the debates held" (Jackson-Breek & Meadow, 1979, p. 179) if no one's
agendas are properly pushed forward. Thus, the authors cite a "triple
agenda" for the journalists, candidates, and public at work. Research
shows that debates can be used to reinforce issue importance and knowledge
in that hearing someone argue a point in the direction of one's already
held strong belief further polarizes that stance and legitimizes a specific
issue's placement on the agenda (Judd & Kulik, 1980).
One additional reason for continued interest in debates has been research
findings that suggest that debates may be more likely to engage voters in
the political process in a positive way. In particular, researchers have
found that exposure to debates can reduce levels of political cynicism
(Kaid, 2003; Kaid, McKinney, & Tedesco, 2000). Certainly debates offer
opportunities for viewers to be more engaged in the political system,
particularly in situations where debate groups have been formed to enhance
participation and where the so-called "town hall format" has been used to
involve voters directly (Kaid, et al., 2000; McKinney, Spiker, & Kaid, 1998).
Medium Theory and Channel Effects
The medium through which a viewer experiences a debate may affect the
viewers' perceptions of the debate and of candidate performance. The
possibility that differing media channels can produce differing effects is,
of course, grounded in medium theory.
The most famous theorist of medium effects was Marshall McLuhan (1964) who
argued that the medium through which a person experienced content could
affect that person's interpretation of the content. In fact, one of
McLuhan's most famous examples of "hot and cold" media effects was his
interpretation of a study of the first 1960 Kennedy-Nixon debate that
suggested that those who saw the debate on television considered Kennedy
the winner, while those who listened on the radio judged Nixon's
performance as superior (Katz & Fledman, 1962; McLuhan, 1964; Rubin, 1967;
Vancil & Pendell, 1987). More recent formulations of medium theory have
been developed to explain how new communication technologies create
important environmental contexts that embody channels with effects beyond
the conveyance of content (Meyrowitz, 1985; Meyrowitz, 1994).
Meyrowitz's view of a micro and macro level of medium theory suggests
effects and aspects of channel variance that go beyond the technological
determinism that characterized McLuhan's hot and cold media views (McLuhan,
1964; Schwartz, 1973). Thus, medium theory focuses on the elements of each
medium that provide meaning in ways that differentiate each medium from
each other medium, including interpersonal encounters.
A vast body of research in education and learning has focused on the
question of whether differences in medium/channel affect the ability to
process information. For instance, the dual-coding hypothesis lends
importance to the use of visual and audio messages together (as in
television), instead of just text, since findings validate that the
combination can lead to higher recall (Lewis, 1997; Quealy & Langan-Fox,
1998; Walma van der Molen & van der Voort, 1998, 2000), perhaps because
"…more senses and stimulation are involved" (Lewis, 1997, p. 23). Work by
Lang, Borse, Wise, and David (2002), employing the limited-capacity
information-processing model rooted in cognitive psychology, has suggested
that responses to cognitive messages depend on the structural features of
the media message (e.g., cuts, sequence, etc.) as well as on the content.
Current research also provides evidence that the use of graphics and
animation enhances computer-aided learning (ChanLin, 1998; Rieber, Boyce, &
Assad, 1990), again putting an emphasis on the visual elements involved in
television and video. Specifically in regard to political media content,
Graber (1987) maintains that the visually-driven nature of television is a
key difference that can affect learning, as compared to other media
channels, since television requires such a low level of user
involvement. Hellweg, Pfau, and Brydon (1992) argue that it is this
emphasis on the visual image that influences an audiences' evaluation of
candidates most, as compared to other media channels.
New technologies have led to the need to expand medium theory and the
study of channel variables to include computer-assisted and Internet
channels. Interactivity is one of the most important distinguishing
factors of computer-mediated and Internet-based channels. The ability of
the viewers to move through materials at their own pace, to stop and start,
to go back and review materials, as well as to benefit from the integration
of text, audio, and visual materials are key aspects of the advantage of
such an environment for instructional materials. Early research on
computer-mediated communication (CMC), established that learning through
CMC is enhanced by interactivity and learner control (Romiszowski & Mason,
1994). In addition, interactivity can be important in computer learning
because it increases the user's ability to control the information-seeking
process, an important factor in learning (Eveland & Dunwoody, 2001)
Through survey research, Kiousis (2001) found evidence that differing
media channels effected how respondents evaluated information, with
newspapers ranking highest, online news outlets ranking second place, and
television news ranking the lowest. However, in an experimental study that
sought to evaluate channel effects on learning between participants exposed
to information via traditional print and the Web, Eveland and Dunwoody
(2002) found contradictory results that indicated respondents learned more
from the traditional print format than Web users.
However, little of this research in medium or channel theory has been
applied to political messages in general or to debates in particular,
although Cohen (1976) and Andreoli and Worchel (1976) have found some
differences in candidate evaluations based on channel variables. Beyond
the famous 1960 Kennedy-Nixon debate example provided by McLuhan and which
has been contested by Vancil and Pendell (1987), there is very little
research specifically on channel variable effects for political
debates. However, in a study of the third 1992 presidential debate,
McKinnon, Tedesco, and Kaid (1993) did provide some replication of
McLuhan's findings by showing that Clinton and Bush were evaluated
differently by television and radio audiences. In a similar study of the
1996 debates, McKinnon and Tedesco (1999) found that Bob Dole achieved a
significant increase in image evaluations on radio but not on television.
Kraus (1996) suggests that one reason for these differences may be that
radio listeners concentrate more on the actual content of the debate and
dissect the words for a clear understanding of candidate stances.
Alternatively televised debates may offer a particular opportunity to see
the impact of the visual component. Morello (1988) argues that the visual
imagery of television has become the primary basis of debate dialogue, and
that the impact of the images in televised debates not only influence, but
even affect comprehension and evaluation of candidates far more than the
actual verbal discourse. Although not a multi-channel debate study, Yawn,
Ellsworth, Beatty, and Kahan (1998) found that the medium, or venue, did
have significant effects in how respondents evaluated candidates when they
compared respondents who attended a debate in person as compared to viewing
it on television.
The limited number of channel or medium studies of debates, using
comparison of traditional media such as television and radio, have been
extended recently by comparisons that include the Internet or Web-based
channels in the mix of channels whose differing effects have been
assessed. For instance, in a study of the 1999 Gore-Bradley presidential
primary debate, McKinney and Gaddie (2000) found that respondents in the
Internet group gained more issue knowledge from viewing the debate than
respondents who viewed the debate on television. In addition, Kaid (2003)
found that viewers who watched 2000 presidential debate excerpts on the
Internet, instead of on the traditional television medium, evaluated both
Bush and Gore more positively but were significantly more likely to vote
for Gore and were also more likely to express their desire to participate
in the political process by turning out to vote.
Research Hypotheses
The combination of prior research on direct effects on candidate images of
exposure to political debates, as well as the findings on the differences
in channel of exposure led to the following hypotheses:
H1 The channel (TV, radio, Internet video, or Internet text) of exposure
to the 2002 Florida gubernatorial debate will significantly affect the
evaluations of candidates Bush and McBride.
H2 The channel (TV, radio, Internet video, or Internet text) of exposure to
the 2002 Florida gubernatorial debate will significantly affect voter
levels of political trust/cynicism.
H3 The channel (TV, radio, Internet video, or Internet text) of exposure to
the 2002 Florida gubernatorial debate will significantly affect voter
information seeking and participation.
H4 The channel (TV, radio, Internet video, or Internet text) of exposure to
the 2002 Florida gubernatorial debate will significantly affect voter issue
agendas.
Method
This project involved an experimental study of a televised debate between
the two major party candidates for governor, Republican Jeb Bush and
Democrat Bill McBride, in the 2002 Florida gubernatorial campaign. The
experiments used a pretest/posttest design. Undergraduate students at the
University of Florida were recruited to watch, read, listen, or download
the debate, which took place on October 22, 2002, in Orlando, Florida, and
was broadcast on local NBC affiliates throughout the state of Florida, on
MSNBC, and on C-SPAN.
For the experiments, undergraduate students in a large introductory-level
lecture class in the communications' college at the University of Florida
were recruited to participate voluntarily in the study. These students
filled out a pretest questionnaire, were exposed to the debate across
various channels (TV, radio, Internet video, or Internet text) in a
classroom setting with other students, and completed a posttest
questionnaire. A total of 136 students participated in the
experiment. These students were randomly assigned to five test
groups. The first group (n = 48) saw the debate on television, and in
addition responded to the pretest/posttest written questionnaires. These
students registered their responses to the candidates on hand-held computer
dials, which allowed a moment-by-moment response to the candidates as the
debate proceeded. The second group (n = 16) saw the debate on television
without the hand-held dials. The third group (n = 29) listened to the
debate on the radio. The fourth group (n = 25) watched the debate as
"streaming video" over the Internet while seated at individual computer
terminals in a media laboratory on October 24, 2002. The final group (n =
18) read the debate in a media laboratory on October 24, 2002, as a
complete transcript on a Web page hosted by a regional newspaper; the
transcript took participants between 30 minutes to one hour to read
completely through. Both Internet groups were told that they were "free to
access other information on the Internet if desired," and thus they were
allowed to seek out additional information about the campaigns or anything
else they desired. This announcement was made in an effort to recreate a
natural setting with the same possibilities that someone would have if
using the Internet on their own.
The total sample of young potential voters in these experiments consisted
of 38 men (28%) and 98 women (72%) with an average age of 21. They
registered their party affiliations as 45% Democrat, 31% Republican, and
24% independent/other. Post-experiment comparisons confirmed the success
of the random assignments in that there were no significant differences in
gender, age, party affiliation, or pre-debate vote intention among the five
groups.
The measuring instruments consisted of pretest and posttest
questionnaires. The questions included demographic information on the
participants, questions about their media habits, and a series of questions
specifically about the debate and the candidates' performance in it. Both
the pretest and the posttest included a feeling thermometer used to measure
reactions to the two candidates, McBride and Bush, on a scale of
0-100. This scale is the traditional one used by the National Election
Studies at the University of Michigan to indicate feelings toward
candidates (Rosenstone, Kinder, & Miller, 1997). Participants also
reacted to both candidates on a semantic differential scale that consisted
of 12 bi-polar adjectives, each rated from 1-7.1 These scales have
achieved high reliability when used as a scale to measure candidate image
(Kaid, 1995). In this debate the Cronbach's alphas were similarly high at
+ .85/.87 for Bush in the pre/posttest and + .81/.84 for McBride in the
pre/posttest. Political cynicism was measured by a series of items that
asked participants to agree-disagree on a 5-point Likert scale with
statements about political efficacy and trust in government and
politicians. Three of these items were adapted from prior scales used in
National Election Studies, and the remainder were developed and used in
earlier studies of the 1996 and 2000 presidential campaigns to measure
levels of political cynicism.2
Results
Candidate Evaluation by Channel of Exposure
The first hypothesis suggested that there would be differences in the
effects of viewing depending upon the channel through which the debate was
viewed. A comparison of the two television groups (one that viewed using
the hand-held dials and one that viewed the television version without the
dials) indicated that there were no significant differences in demographic
composition or in candidate ratings. Thus, these two groups were combined
into a single television-viewing group. Initial comparisons of the overall
pretest/posttest ratings of both candidates indicated that viewing the
debate, regardless of channel, resulted in a significant positive change
from 53.3 to 58.3 for Bush, t (131) = 7.98, p < .01; and a significant
negative change from 53.0 to 51.3 for McBride, t (132) = 2.24, p < .04 (see
Table 1).
However, when examining the pretest/posttest comparisons according to
channel of exposure, it is clear that there is a difference among the two
candidates in how channel affected the evaluations. In each of the four
channel treatments, scores for Bush rose significantly. There is a clear
difference by channel in regard to McBride's evaluations. While absolute
scores for McBride fell slightly in all channels after watching the debate,
the only treatment in which there was a statistically significant decline
for McBride was in the radio group, t (28) = 3.92, p < .001. Thus, the
results offer partial support for the first hypothesis.
Channel Effects on Political Cynicism Levels
The second hypothesis concerned the effect of debate exposure on levels of
political cynicism. Political cynicism was measured in both the pretest
and posttest in order to determine the effects of debate exposure based on
channel. Table 2 shows that differences in the pretest mean (24.8) and
posttest mean (23.8) scores for the entire sample do, indeed, indicate a
significant decline in feelings of political cynicism after exposure to the
debate, t (132) = 2.95, < .01. Again, there is a difference among
channels in these findings. When examined individually by treatment group,
Table 2 indicates that only in the radio group is the decline from pretest
(24.9) to posttest (22.5) statistically significant, t (26) = 3.02, p <
.01. Thus, the second hypothesis was supported in that exposure to the
debate did significantly affect voter levels of political trust and
cynicism according to channel.
Channel Effects on Political Information and Participation
Levels of voter information seeking and participation were gauged across
the four channels to test the third hypothesis (see Table 3). Rated on a
scale of 1-5, these questions asked respondents to report how likely they
would be to engage in the particular political activity after exposure to
the debate. There were very few differences in these comparisons, offering
only limited support for this hypothesis. For instance, respondents
exposed to the Internet video version of the debate were significantly more
likely to want to read additional information about the campaign in the
newspaper than were those exposed to the debate via television, t (86) = -
3.44, p < .001. Similarly, those who saw the debate on Internet video
(3.54) were significantly more likely also to seek additional information
from the newspaper, compared to those who heard the debate on the radio
(2.34), t = (51) = - 2.45, p < .02. Additionally, those who saw the
debate on Internet video were significantly more likely to want to "talk
with friends about the candidates and issues" than were those who saw only
the Internet text (2.44), t (40) = 1.18, p < .02.
Channel Effects on Agenda Setting
The final hypothesis suggested that exposure via different channels would
affect the agenda-setting potential of the debate. Table 4 shows that the
channel through which viewers were exposed to the debate had little
influence on the agenda of issues they judged to be salient. Respondents
were asked in both the pretest and the posttest to list the issues they
felt were the most important ones facing the state of Florida. For the
entire sample, education was the most frequently listed issue before
exposure to the debate, followed in order by the economy, health care, the
environment, crime, the elderly, problems related to children, and
terrorism and international concerns. This listing of issues was highly
correlated with the ranking of issues after viewing the debate (r = .93, p
< .01. The same was true for the pretest/posttest ranking of issues in the
television, radio, and Internet video groups. That is, the salience of
issues for respondents remained essentially the same, despite exposure to
the debate.
However, there was one shift in issue agenda salience. This shift took
place between the pretest and posttest for the Internet text-viewing
group. While this group's pretest agenda of issues was not statistically
different from that of the other pretest groups, the posttest issue
rankings of this group were not correlated at a statistically significant
level with the pretest rankings. Table 4 shows that the biggest difference
in issue judgments after exposure in this group was in the rankings of the
economy, the elderly, and crime. The economy, which had been ranked third
in the pretest by this group, was totally missing from the agenda concerns
of the group in the posttest. Crime, which had been ranked as the fifth
most important issue in the pretest, rose to the second place concern in
the posttest, and problems of the elderly, which had not been mentioned at
all in the pretest, took fourth place on the posttest agenda of the
Internet text group.
Discussion
The first hypothesis was supported in that the channel of exposure to the
debate significantly affected the evaluations Bill McBride. This result is
similar to that found by researchers studying the comparison of television
and radio exposure for presidential candidates in earlier elections
(McKinnon & Tedesco, 1999; McKinnon, Tedesco, & Kaid, 1993). It also
offers further support for findings that have suggested that there may be
an interaction between sources and channels in political message reception,
such that some candidates fare better on one medium than another. This
interaction between source and channel has previously been substantiated
more clearly in the study of political advertising messages (Cohen, 1976).
Of particular note was the radio listener's reaction to the debates. Here,
McBride, who was trying to paint himself as a "downhome, Florida boy" by
using simple communication strategies ended up coming across in a less than
stellar performance that actually lost more than three times as many points
of approval than any other communication channel. Thus, McBride's
communication strategy did not play well on the radio.
At the same time that McBride was losing voters through the radio channel,
Bush was picking them up. While radio resulted in the highest change for
Bush, he did quite well across all of the media in eliciting positive
evaluation scores from the public. Just as the radio played an impressive
role in candidate evaluations, results in this group stood out on the
political cynicism measure as well. While the overall level of cynicism
across all of the groups dropped one point after viewing the debates, the
radio cell reports of cynicism dropped 2.4 points. This might be explained
by the fact that hearing the debate on the radio allowed the listener to
ignore the contrived slick image of the candidate and concentrate on the
actual words, messages, and issues of the candidate.
In regards to information seeking and political participation across the
channels, Internet channels seemed to have a slight advantage in eliciting
active behaviors. Although all of these differences were not statistically
significant, from likelihood to contact a candidate to actively seeking out
campaign information, Internet video and Internet text participants rated
higher across the board in regards to information seeking and political
participation.
In a day where candidates have come to understand the importance of the
visual imagery of television in debates, it will become more and more
important to master all types of presentation formats. Candidates will
have to market themselves via Web sites and come across positively in
combinations of text and video on the Web. In preparing a performance for
one channel, the performance in another must not suffer as it did in
McBride's case. Candidates must understand how they are coming across in
all of the channels through which voters may be exposed to a candidate and
message.
Candidates must begin to understand what channels garner the most positive
change in regards to their own performances and begin to campaign actively
on such media. For some, that may mean engaging in online chats and other
mediated discussions, while for others it may mean hitting the radio talk
show circuit. Regardless, candidates and campaign managers should become
acutely aware of what works where and do everything possible to capitalize
on it.
As with most experimental studies, this research faced some
limitations. The sample size and composition result in limited
generalizablity. Although respondents were randomly assigned to the
various treatment groups, the sample itself was limited to young
people. There were also limitations to external validity caused by the
setting of the exposures. Respondents were watching on a screen in a
classroom with other students, not alone or with family/friends in their
living rooms on their own television screens. Forced exposure and the
pretest/posttest design created an atypical viewing environment.
As scholars move forward in understanding the effects of a mutlichannel
world and new technologies, future research should investigate not only why
certain people select specific channels, but how they use their primary or
preferred channel differently from other channels. Additionally,
researchers should begin to investigate the characteristics of users within
each channel so that messages can be better constructed for salience and
impact among them. Studies such as the one reported here should be
replicated in other state and national races and with other message
formats. Only with additional understanding of the characteristics of
individual channels and their effects can scholars advance the study of
both political effects and medium theory.
References
Andreoli, V., & Worchel, S. (1976). Effects of media, communicator and
message position on attitude change. Public Opinion Quarterly, 42, 59 – 70.
Benoit, W.L., McKinney, M.S., & Holbert, R. L. (2001). Beyond learning
and persona: Extending the scope of presidential debate
effects. Communication Monographs, 68(3), 259 – 273.
Benoit, W.L., McKinney, M.S., Stephenson, M.T. (2002). Effects of
watching primary debates in the 2000 U.S. presidential campaign. Journal
of Communication, 52, 316 – 331.
Carlin, D. P., & McKinney, M. S. (1994). The 1992 presidential debates in
focus. Westport, CT: Praeger.
ChanLin, L. (1998). Animation to teach students of different knowledge
levels. Journal of Instructional Psychology, 25(3), 166-175.
Cohen, A. (1976). Radio vs. TV: The effects of the medium. Journal of
Communication,
26, 29-35.
Cohen, B. (1963). The press and foreign policy. Princeton: Princeton
University Press.
Davis, P. (2002, November 8). What appeared to be a close contest in
Florida between Jeb Bush and Bill McBride turned out to be a 13-point win
for Bush. All Things Considered [Radio Program]. Washington, DC: National
Public Radio.
Drew, D., & Weaver, D. (1991). Voter learning in the 1988 presidential
election: Did the debates and the media matter? Journalism Quarterly, 68,
27-37.
Eveland, 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, 34-53.
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, 48-78.
Floyd, S., Trammell, K.D., & Williams, A.P. (2002) Mini rallies at
SuperWalMart: Measuring the effectiveness of Janet Reno's Red Truck
Tour. Paper presented at the Annual Meeting of the Florida Communication
Association, St. Petersburg, Fla.
Graber, D. A. (1987). Television news without pictures? Critical Studies in
Mass Communication, 4 (1), 74-79.
Hellweg, S., Pfau, M., & Brydon, S. (1992). Televised presidential
debates: Advocacy in
contemporary America. Westport, CT: Greenwood.
Hullett, C.R., & Louden, A.D. (1998). Audience recall of issues and image
in congressional debates. Argumentation & Advocacy, 34(4), 189-203.
Jackson-Breek, M., & Meadow, R. G. (1979). The triple agenda of
presidential debates. Public Opinion Quarterly, 173-180.
Judd, C., & Kulik, J. (1980). Schematic effects of social attitudes on
information processing and recall. Journal of Personality and Social
Psychology, 38, 569-578.
Kaid, L. L. (1995). Measuring candidate images with semantic
differentials. In K. L. Hacker (Ed.), Candidate Images in Presidential
Elections (pp. 131-134). Westport, CN: Praeger.
Kaid, L. L. (2003). Comparing Internet and traditional media: Effects on
voters. American Behavioral Scientist, 46, 677-691..
Kaid, L. L., McKinney, M. S., & Tedesco, J. C. (2000). Civic dialogue in
the 1996 Presidential campaign: Candidate, media, and public voices.
Cresskill, NJ: Hampton Press.
Kaid, L.L. (2002). Political advertising and information
seeking: Comparing exposure via traditional and Internet
channels. Journal of Advertising, 31(1), 27-35.
Katz, E., & Feldman, J. (1962). The debates in light of research: A survey
of Surveys. In S. Kraus (Ed.), The great debates (pp.
173-223). Bloomington, IN: Indiana University Press.
Kendall, K. E. (1997). Presidential debates through media
eyes. American Behavioral Scientist, 40(8), 1193-2002.
Kiousis, S. (2001). Public trust or mistrust? Perceptions of media
credibility in the information age. Mass Communication & Society, 4, 381-404.
Kraus, S. (1996). Winners of the first 1960 televised presidential debate
between Kennedy and Nixon. Journal of Communication, 46(4), 78-96.
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, 215-145.
Lewis, C.S. (1997). Interactive multimedia brings new possibilities to
adult learning. Adult Learning, 8(5-6), 23-25.
McCombs, M. E, & Shaw, D. L. (1972): The agenda-setting function of mass
media. Public Opinion Quarterly, 36 (2), 176-187.
McCombs, M. E., & Shaw, D. L. (1993). The evolution of agenda-setting
research: Twenty-five years in the marketplace of ideas. Journal of
Communication, 43(2), 58 – 67.
McKinney, M.S., & Gaddie, K.C. (2000, November). The medium and the
message: TV vs. Internet viewing of a presidential primary debate. Paper
presented at the National Communication Association Convention, Seattle, WA.
McKinney, M.S., Kaid, L.L., & Robertson, T.A. (2001). The front-runner,
contenders, and also-rans: Effects of watching a 2000 Republican primary
debate. American Behavioral Scientist, 44(12), 2232-2251.
McKinney, M. S., & Lamoureux, E. R. (1999). Citizen response to the 1996
presidentialdebates: Focusing on the focus groups. In L. L. Kaid & D. G.
Bystrom (Eds.), The electronic election: Perspectives on the 1996
campaign communication (163-177). Mahwah, N.J.: Lawrence Erlbaum.
McKinney, M. S., Spiker, J. A., & Kaid, L. L. (1998). DebateWatch '96 and
citizen
engagement: Measuring political alienation. In T. J. Johnson, C. E. Hays, &
S. P. Hays (Eds.), Engaging the public: How government and the media can
reinvigorate American democracy (185-193). Rowman and
Littlefield: Lanham, MD.
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 (191-206). Mahwah, N.J.: 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, 106-118.
McLuhan, M. (1964). Understanding media. New York: New York American Library.
Meadow, R., & Jackson-Beeck, M. (1978). Issue evolution: A new
perspective on presidential debates. Journal of Communication, 28, 84 – 92.
Meyrowitz, J. (1985). No sense of place: The impact of electronic media on
social behavior. New York: Oxford University Press.
Meyrowitz, J. (1994). Medium theory. In D. Crowley & D. Mitchell (Eds.),
Communication theory today (pp. 50-77). Polity Press: Cambridge.
Morello, J. T. (1988). Argumentation and visual structuring in the 1984
Mondale-Reagan debates. Western Journal of Speech Communication, (52), 277-290.
Quealy, J., & Langan-Fox, J. (1998). Attributes of delivery media in
computer-assisted
instruction. Ergonomics, 41 (3), 257-279.
Rieber, L. P., Boyce, M., & Assad, C. (1990). The effects of computer
animation on adult learning and retrieval tasks. Journal of Computer-Based
Instruction, 17(2), 46-52.
Romiszowski, A. J., & Mason, R. (1994). Computer-mediated communication,
483-455. In D. M. Moore & F. M. Dwyer (Eds.), Visual literacy: A spectrum
of visual learning. Englewood Cliffs, N.J.: Educational Technology
Publications.
Roberts, M.S., & Williams, A.P. (2003-in press). Lockbox and fuzzy
math: Associations of viewers' debate recall and voter behavior in the
2000 presidential Campaign. In L. L. Kaid, J. C. Tedesco, D. Bystrom, & M.
S. McKinney (Eds.). The millennium election: Communication in the 2000
campaigns. Lanham, MD: Rowman and Littlefield Publishers.
Rosenstone, S. J., Kinder, D. R., Miller, W. E., & the National Election
Studies (1997).
American national election study,1996: Pre- and post-election survey
Computer file]. Ann Arbor, MI: University of Michigan, Center for Political
Studies [producer], 1997. Ann Arbor, MI: Inter-university Consortium for
Political and Social Research [distributor].
Rubin, B. (1967). Political television. Belmont: Wadsworth.
Schwartz, T. (1973). The responsive chord. Garden City, NY: Doubleday.
Sears, D.O., & Chaffee, S.H. (1979). Uses and effects of the 1976
debates: An overview of
empirical studies. In S. Kraus (Ed.), The great debates: Carter
vs.Ford,1976 (223-261). Bloomington, IN: Indiana University.
Trammell, K. D., Kaid, L. L., Williams, A. P., & Landreville, K. (2003—in
press). Under the international spotlight: Marketing politics through
debates. Business Research Yearbook.
Trent, J.S., & Friedenberg, R.V. (1995). Political campaign
communication: Principles and practices (3rd Ed.). Connecticut: Praeger
Publishers.
Vancil, D. L., & Pendell, S.D. (1987). The myth of viewer-listener
disagreement in the first Kennedy-Nixon debates. Central States Speech
Journal, 38, 16-27.
Walma van der Molen, J. H., & van der Voort, T. H. A. (1997). Children's
recall of television and print news: A media comparison study. Journal of
Educational Psychology, 89 (1), 82-91.
Walma van der Molen, J. H., & van der Voort, T.H. A. (2000). The impact
of television, print, and audio on children's recall of the news. Human
Communication Research, 26 (1), 3-26.
Walsh, E. (2002, November 7). Democrats gain, but both sides claim
success. The Washington Post, p. A41.
Weaver, D., & Drew, D. (1995). Voter learning in the 1992 presidential
election: Did the "nontraditional" media and debates matter? Journalism &
Mass Communication Quarterly, 72(1), 7-17.
Yawn, M., Ellsworth, K., Beatty B, & Kahan, K. F. (1998). How a
presidential primary debated changed attitudes of audience members.
Political Behavior, 20, 155 – 181.
Notes
1. The 12 bipolar adjective pairs used were: qualified-unqualified,
sophisticated-unsophisticated, honest-dishonest, believable-unbelievable,
successful-unsuccessful, attractive-unattractive, friendly-unfriendly,
sincere-insincere, calm-excitable, aggressive-unaggressive, strong-weak,
active-inactive.
2. These eight items were developed to use as a scale for measuring
cynicism in the electorate. Several items were adapted from the National
Election Studies (Rosenstone et al., 1997) which have used several similar
measures for many years. The specific items, which were answered on a scale
of 1 (strongly disagree) to 5 (strongly agree) were: (1) Whether I vote or
not has no influence on what politicians do, (2) One never really knows
what politicians think, (3) People like me don't have much say in what the
government does, (4) Politics and government are too complicated to
understand, (5) Politicians will not always do the right thing, (6)
Politicians forget their promises after the election is over, (7)
Politicians are more interested in power than in what people think, and (8)
One can always trust what politicians say. Although these scale items have
achieved acceptable reliability measures in prior studies (Kaid, McKinney,
& Tedesco, 2000), they did not perform as well in this study, achieving a
Cronbach's alpha of only .59 in the pretest and .66 in the posttest.
Table 1
Evaluations of Candidates Before and After Debate Exposure by Channel
Bush McBride
Pre-Debate Post-Debate Pre-Debate Post-Debate
All Groups Combined
(N = 136) 53.3 58.3a 53.0 51.3a
TV (n = 64) 53.7 57.8a 52.0 53.0
Radio (n = 29) 52.8 59.9a 53.7 47.3a
Internet Video (n = 25) 54.4 60.5a 53.5 51.1
Internet Text (n = 18) 50.8 54.4a 54.4 52.0
at-test indicates difference between pretest and posttest is significant at
p < .05.
Table 2
Political Cynicism Levels Before and After Debate Viewing by Channel
Pre-Debate Post-Debate
All Groups Combined (N = 136) 24.8 23.8a
TV Only (n = 64) 25.0 24.4
Radio (n = 29) 24.9 22.5a
Internet Video (n = 25) 23.9 23.2 Internet Text (n =
18) 25.6 24.6
---------------------------------------------------------------------------------
at-test indicates difference between pretest and posttest is significant at
p < .05.
Table 3
Differences in Information Seeking and Political Participation
TV Radio Internet Video Internet Text
Watch more ads about
candidates on TV 3.34 3.34 4.04 3.39
Watch news for more
information 2.70 2.79 3.25 2.83
Volunteer for campaign 5.16 5.76 5.71 5.78
Participate in on-line chat 5.75 5.66 5.79 6.11
Talk with friends about
candidates and issuesf 2.95 2.83 3.63f 2.44f
Read newspapersb,d 2.31b 2.34e 3.54b 2.67e
Contact candidate 5.48 5.52 5.71 5.67
Use Internet for
general campaign 3.83 3.72 4.42 3.94
information
Vote in next election 2.11 1.93 2.29 2.39
Contribute money to
candidate 5.80 6.24 5.92 6.39
Use Internet for more
info on political issues 4.03 3.93 4.25 3.94
Go to candidate Web
site 4.11 4.10 4.58 4.61
a The t-test shows that the difference between TV and Radio is signif. at p
< .05.
b The t-test shows that the difference between TV and Internet Video is
signif. at p < .05.
c The t-test shows that the difference between TV and Internet Text is
signif. at p < .05.
d The t-test shows that the difference between Radio and Internet Video is
signif. at p < .05.
e The t-test shows that the difference between Radio and Internet Text is
signif. at p < .05.
f The t-test shows that the difference between Internet Video and
Internet Text is signif.
at p < .05.
0
Comparing
Campaign Communication Channels
Table 4
Issue Agendas Before and After Debate Exposure by Channel
Ranking
Issue Total TV Radio Internet Video Internet Text
Pre Post Pre Post Pre Post Pre Post Pre Post
Education 1 1 1 1 1 1 1 1 1 1
Economy 2 3 2 2 2 3 2 2 3 -
Health Care 3 2 3 3 3 2 2 3 2 2
Environment 4 5 4 5 4 5 5 5 3 4
Crime 5 4 5 5 5 4 4 4 5 2
Elderly/S.S. 6 7 7 7 5 6 6 7 - 4
Children 7 6 8 6 - 6 7 6 5 6
International/
Terrorism 8 8 7 7 7 8 8 - - 6
Spearman's Rho .93a .92a .90a .96a .71
----------------------------------
aSpearman's Rho correlation is significant at p < .05
|
