ArticlePDF Available

Gans, R. (2014). The politics of HPV vaccination advocacy: Effects of source expertise on effectiveness of a pro-vaccine message. Proceedings of the New York State Communication Association: Vol. 2013, Article 3. http://docs.rwu.edu/nyscaproceedings/vol2013/iss2013/3/The Politics of HPV Vaccination Advocacy: Effects of Source Expertise on Effectiveness of a Pro-Vaccine Message

Authors:

Abstract and Figures

Persistent public resistance to an apparently safe, effective and life-saving public health practice such as HPV vaccination illustrates a significant issue in the communication of behavioral recommendations based on evidence-based scientific data and consensus views of scientific and medical experts. This study examines the influence of source expertise on pro-HPV-vaccine advocacy messaging effectiveness among audiences of differing political ideologies. The findings support prior research indicating greater resistance to HPV vaccination among political conservatives. Subjects who self-identified politically as Centrists and Conservatives were significantly less likely to think deeply about a pro-HPV advocacy message delivered by an expert spokesperson than were politically self-identified Progressives. Conservatives who viewed a pro-HPV vaccination message delivered by a non-expert spokesperson had significantly more positive attitudes toward HPV vaccination than Conservatives who received no advocacy message (the control condition). By contrast, attitudes of Conservatives who viewed a pro-HPV vaccination message delivered by an expert spokesperson were not significantly different from those who received no advocacy message. The findings suggest an over-reliance on expert spokespeople for delivering science-based behavioral recommendations.
Content may be subject to copyright.
Proceedings of the New York State Communication Association
Volume 2013 Proceedings of the 71st New York State
Communication Association Article 3
2014
e Politics of HPV Vaccination Advocacy: Eects
of Source Expertise on Eectiveness of a Pro-
Vaccine Message
Roger Gans
University at Albany, rgans@albany.edu
Follow this and additional works at: hp://docs.rwu.edu/nyscaproceedings
Part of the Community Health and Preventive Medicine Commons,Health Communication
Commons,Mass Communication Commons,Public Relations and Advertising Commons, and the
Social Inuence and Political Communication Commons
is Conference Paper is brought to you for free and open access by the Journals at DOCS@RWU. It has been accepted for inclusion in Proceedings of
the New York State Communication Association by an authorized administrator of DOCS@RWU. For more information, please contact
mwu@rwu.edu.
Recommended Citation
Gans, Roger (2014) "e Politics of HPV Vaccination Advocacy: Eects of Source Expertise on Eectiveness of a Pro-Vaccine
Message," Proceedings of the New York State Communication Association: Vol. 2013, Article 3.
Available at: hp://docs.rwu.edu/nyscaproceedings/vol2013/iss2013/3
The Politics of HPV Vaccination Advocacy: Effects of
Source Expertise on Effectiveness of a Pro-Vaccine
Message
RogerGans
UniversityatAlbany,SUNY
__________________________________________________________________
Persistent public resistance to an apparently safe, effective and life-saving public
health practice such as HPV vaccination illustrates a significant issue in the
communication of behavioral recommendations based on evidence-based scientific
data and consensus views of scientific and medical experts. This study examines
the influence of source expertise on pro-HPV-vaccine advocacy messaging
effectiveness among audiences of differing political ideologies. The findings support
prior research indicating greater resistance to HPV vaccination among political
conservatives. Subjects who self-identified politically as Centrists and
Conservatives were significantly less likely to think deeply about a pro-HPV
advocacy message delivered by an expert spokesperson than were politically
self-identified Progressives. Conservatives who viewed a pro-HPV vaccination
message delivered by a non-expert spokesperson had significantly more positive
attitudes toward HPV vaccination than Conservatives who received no advocacy
message (the control condition). By contrast, attitudes of Conservatives who
viewed a pro-HPV vaccination message delivered by an expert spokesperson were
not significantly different from those who received no advocacy message. The
findings suggest an over-reliance on expert spokespeople for delivering
science-based behavioral recommendations.
__________________________________________________________________
The debate over public support for the human papillomavirus (HPV) vaccine crosses the
boundaries of science, health and politics. Should it be mandatory for all school girls
approaching puberty? For all school children of both genders? Should it be voluntary?
Should it be a publicly supported program at all? Instead of being based on objective
evaluation of empirical evidence and expert recommendations, people’s answers to these
questionsseemtoalignalongideologicallypartisanlines(Kahanetal.,2010).
The logical, rational, utilitarian case for universal vaccination against HPV seems highly
persuasive. Approved by the FDA in 2006 for use with females and in 2010 for use with
males, the HPV vaccine was expected to reduce deaths due to cervical cancer by 70
percent (Kaufman, 2006), prevent up to 14,000 cases of a wide range of cancers annually,
and save thousands of lives (Centers for Disease Control, 2013a). Acceptance and
utilization of the vaccine has been far from universal, however. Nationwide, only about a
third of adolescent girls are getting the recommended vaccinations, and only about one
percent of adolescent boys (Centers for Disease Control, 2013b; Pruitt and Schootman,
2010). In fact, according to the National Immunization Survey of Teens, 2008–2010, the
percentage of parents actively choosing to refuse HPV vaccination for their children is
increasing(Dardenetal.,2013).
1
Gans: HPV Vaccination Advocacy and Source Expertise
Published by DOCS@RWU, 2014
With the persistence of low vaccination rates throughout the United States largely
attributed to misinformation and active resistance rather than lack of availability (Pruitt
and Schootman, 2010), it is clear that finding more effectively persuasive means of
communicating the benefits of HPV vaccination would be desirable. This study examines
the influence of source expertise on proHPVvaccine advocacy messaging effectiveness
amongaudiencesofdifferingpoliticalideologies.
Literature Review
Communication of Risk
In a complex world filled with constantly changing and newly emerging dangers, the
effective communication of risk—and of gaining compliance with and adoption of
appropriate behavioral responses to risk—is vital to the safety and wellbeing of any
society. Various reviews of research and best practices regarding risk communications
suggest that theorybased communication practices are more likely to be effective than
those based solely on intuition, but that even the most thoroughly researched practices
should be evaluated (Fischhoff, Brewer & Downs, 2011; Glanz and Bishop, 2004).
Despite this evidence, however, many riskrelated public health communication initiatives
are based on intuition and “common sense” rather than on tested theories, and not
systematically evaluated, which hinders the improvement or elimination of programs that
are ineffective and prevents the maximization of benefits from those that are effective
(Miche and Abraham, 2004; Wilson, 2011). A significant number of interventions even
produce what some researchers call “the Boomerang Effect” (Fishbein et al., 2002; Miller
et al., 2006; Sylvia, 2006), acting to encourage and increase the very behaviors they are
intendedtodiscourage.
Audiences Predisposed Against Science
Complicating the task of communicating researchbased information and behavioral
recommendations is an often uninformed, uninterested and surprisingly hostile audience.
The general public is far from an enthusiastic or actively engaged audience for scientific
information, judging by general reading and viewing preferences (Alliance for Audited
Media, 2013; Gorman, 2011; Kondolojy, 2012a; Kondolojy, 2012b). A large segment of the
population actively resists and argues against science and the recommendations of
scientists (Mooney, 2005, 2012; Newport, 2012; Public Religion Research Institute, 2011).
Political ideology seems to play a role in attitude toward scientists and scientific
information, with people who are politically more conservative less likely to trust scientists
asasourceofinformationaboutpublicpolicyissues(Hamilton,2010).
Scientific data such as morbidity and mortality statistics and peerreviewed reports about
longterm health effects accruing from various causes can be intimidating for many
people, even those for whom it may have significant personal relevance (Kahan et al.,
2010; Larsson, 2006; Ramanadhan and Viswanath, 2006). Even people who do not
actively avoid scientific and healthrelated information tend to process it in a
lessthanfully attentive, rational manner: they use mental shortcuts that can lead to errors
in judgment and resulting behaviors that are not of optimal selfinterest (Kahan et al.,
2010; Gilovitch et al., 2002; Kahneman, 2011). Kahan et al. (2010) suggest that
2
Proceedings of the New York State Communication Association, Vol. 2013 [2014], Art. 3
http://docs.rwu.edu/nyscaproceedings/vol2013/iss2013/3
personality types associated with partisan ideologies and valuesstructures play a role in
whichkindsofmentalshortcutsareused.
Kahan et al. (2011) suggest that individuals’ perceptions of risk and the validity of the
facts on which they may be based are influenced by those individuals’ cultural
predispositions and “shared moral evaluations” (p. 148). In a study of perception of risks
associated with HPV vaccination, Kahan et al. (2010) found that subjects with cultural
predispositions associated with political and social conservatism perceived much higher
risk from HPV vaccination than subjects whose cultural cognitions aligned more with
liberal and progressive political and social values. Using surveyadministered cultural
values scales, Kahan et al. (2010) rated their study participants on their placement on two
distinct personality continua: “individualistic vs. communitarian,” and “hierarchical vs.
egalitarian” (p. 507). The first two of each of these pairs are strongly associated with
conservative political and social views, while the latter two are strongly associated with
progressive and liberal views. Subjects scoring high on the individualistic and hierarchical
scales were more likely to be risk averse and distrusting of science than those scoring
high in the egalitarian and communitarian scales, suggesting that those scoring high on the
individualistic and hierarchical scales were more prone to biased assimilation—the
“tendency of individuals selectively to credit and dismiss information in a manner that
confirmstheirpriorbeliefs”(p.504).
Source Credibility and Persuasion
The prevailing view of most persuasion practitioners and communication researchers
seems to be that highly credible sources of persuasive messages are more effective than
less credible sources, and that expertise is a general cognate for credibility in most
situations (Sternthal et al., 1978). This is a fairly wellresearched view (see Pornpitakpan,
2004), but it may also be an example of the kind of intuitive assumption warned against by
Fischhoff, Brewer & Downs (2011) and Wilson (2011). While many studies find expertise
and trustworthiness to be significantly influential in obtaining behavioral compliance, others
find different degrees of ascendancy for these qualities, and others have identified
conditions in which source credibility—expertise, in particular—can be a liability
(Pornpitakpan,2004;Sternthaletal.,1978).
Tormala, Briñol & Petty (2007) suggest that source credibility—which they describe as a
combination of expertise and trustworthiness—can play different roles under different
conditions of elaboration. Under conditions of low elaboration, source expertise acts as a
heuristic cue, with the greater the expertise the greater the agreement with the expert
viewpoint. Under conditions of moderate elaboration, source expertise influences the
amount of thought by subjects about the issue under examination, with high credibility
sourcesinspiringmoreissuerelevantthinking.
Source Expertise and Resistance to Persuasion: Psychological Reactance
A number of researchers have cited Brehm’s Theory of Psychological Reactance in
describing conditions in which advocacy by an authoritative source high in expertise and
credibility can stimulate resistance rather than compliance (e.g., Burgoon et al., 2002;
Miller et al., 2006). The theory proposes that psychological reactance occurs in response
to perceived threats to freedom, and can lead to responses such as “simply ignoring the
3
Gans: HPV Vaccination Advocacy and Source Expertise
Published by DOCS@RWU, 2014
persuasive attempt, derogating the source, and even producing even more of the undesired
behaviors as a means of demonstrating choice or restoring attitudinal freedom” (Burgoon
etal.,2002,p.215).
The theory of psychological reactance provides an explanation for the persistent
resistance of subjects with firmly held beliefs when faced with attempts by highly credible
expert authority figures to present persuasive arguments that threaten those beliefs. This
hardening of resistance as a reaction to perceived threats to freedom and individual
autonomy is consistent with personality types that lean toward political and social
conservatism, which have also been shown to be more likely to oppose vaccination against
HPV(Kahan,2010).
Even the mere anticipation of a potentially persuasive counterattitudinal argument can
lead to greater resistance. Subjects who are forewarned of a forthcoming argument
against a current belief or position seem to engage in “anticipatory argumentation” (Petty
and Cacioppo, 1977, p. 645) and thereby become even more resistant to persuasion (Petty
andCacioppo,1977,1979).
Source Credibility, Attitude Certainty and Counter-Attitudinal Advocacy
In a series of studies, Tormala and colleagues (Tormala & Petty, 2002, 2004; Tormala,
Clarkson & Petty, 2006) demonstrated that while resisting a strong counterattitudinal
argument—such as one from a credible source—generally leads to greater attitude
certainty, resisting a weak argument—such as from a source low in credibility—can lead
to reduced attitude certainty. They speculate that while defending one’s beliefs from a
powerful argument can lead to greater confidence in the rightness of those beliefs,
successfully defending against a weak argument can leave one wondering whether one’s
beliefswouldhavestooduptoastrongerargument(TormalaandPetty,2004).
Karmarkar and Tormala (2010) present evidence supporting the notion that when low
expertise sources express certainty about an issue, they “violate expectancies, stimulate
involvement, and promote persuasion” (p. 1033). In other words, the unexpected
advocacy of a nonexpert spokesperson is likely to gain attention and stimulate elaboration
about a persuasive argument in situations in which the words of an expert spokesperson
mightbelostamidthegeneralcacophonyofmessagescompetingforattention.
Source Expertise as a Heuristic Cue and Moderating Factor in
Elaboration Likelihood
Much of the recent research on source effects and attitude change has been conducted
under conditions of induced elaboration, with subjects instructed to think about and record
their thoughts and in some cases their counterattitudinal arguments (e.g., Clark, Evans &
Wegener, 2011; Clark et al., 2012; Lemansky and Lee, 2012; Tormala, Briñol and Petty,
2007; Tormala and Petty, 2002, 2004). This focus on conditions of high elaboration is
based at least in part on the assumption that source credibility functions as a fairly
simplistic heuristic cue under conditions of low elaboration without a significant influence
on critical message consideration. There may also be an underlying assumption that
conditions of high elaboration are more conducive or more necessary for significant
4
Proceedings of the New York State Communication Association, Vol. 2013 [2014], Art. 3
http://docs.rwu.edu/nyscaproceedings/vol2013/iss2013/3
attitude change, consistent with a broad understanding of the Elaboration Likelihood
Model(CacioppoandPetty,1984).
Another approach to examining these issues might be to consider whether and how
source credibility influences a subject’s level of message elaboration, as well as the
resultinginfluenceonattitudechange.
Research Questions
Applied to the universe of attitudes and predispositions toward HPV vaccination, the past
sixplus years of “expert” sourced advocacy messages have led to (or at least coincided
with)increasedresistancetothevaccine(Dardenetal.,2013).
Prior research has shown that this resistance tends to correlate strongly along lines of
political and social conservatism (Kahan et al., 2010). Other research has shown that
scientific expertise and scientific information in general seem to generate resistance,
which can take the form of actively thoughtful counterargument, active avoidance of any
thought on the matter, and a continuum of varyingly passivetoactive resistance between
those two positions (Hamilton, 2010, 2011; Mooney, 2005, 2012; Newport, 2012; Public
ReligionResearchInstitute,2011;RamanadhanandViswanath,2006).
Examination of these prior findings raises the question of whether people who are
predisposed to oppose vaccination of preadolescent children against HPV infection would
be likely to react differently to a provaccination advocacy message from an authoritative
expert source—one whose mere appearance might represent forewarning of strong
counterattitudinal arguments (Petty and Cacioppo, 1977, 1979)—than from an obviously
nonauthoritative, nonexpert source. Faced with an authoritative expert, would such
antivaccination partisans tend to begin marshalling their counter arguments even before
considering the proadvocacy message, or reject the arguments of the “expert” instantly
and heuristically, without significant elaboration on the provaccination advocacy
message?Underthesekindsofconditions,persuasionwouldbevirtuallyimpossible.
Conversely, when faced with an obviously nonauthoritative, nonexpert spokesperson,
would partisans be more likely to listen to—and rationally process—the arguments of the
nonexpert, a condition in which at least some possibility of persuasion exists? And
following this hypothetical thread to the possibility of practical application for maximum
public benefit, would the use of an obviously nonexpert spokesperson to present the case
for HPV vaccination be a more effective strategy than the use of an expert
spokesperson?
For the purposes of operationalization of the current study, these issues can be
summarizedinthefollowingresearchquestions:
R1. Will spokesperson expertise have a significant effect on subjects’ degree of
elaborationonandattitudetowardaproHPVvaccineadvocacymessage?
5
Gans: HPV Vaccination Advocacy and Source Expertise
Published by DOCS@RWU, 2014
R2. Will political ideology function as a significant moderating factor to spokesperson
expertise in regard to message elaboration and/or attitude toward the object of the
messageadvocacy?
Method
Overview
The current study was constructed as an experiment to test how people’s attitudes
regarding HPV vaccination are influenced by the perceived expertise of the person
delivering a provaccination advocacy message. Using an online survey, the study
measured subjects attitude toward HPV vaccination in three randomly assigned
conditions: (1) Control: after a basic informational overview of the HPV vaccine; (2)
Expert: after the same informational overview plus a provaccine advocacy statement
delivered by an expert spokesperson; and (3) NonExpert: after the same informational
overview plus an identical provaccine advocacy statement delivered by a nonexpert
spokesperson. Subjects in the Expert and NonExpert conditions were also asked to
complete a series of questions designed to measure their degree of elaboration on the
advocacymessagedeliveredbythespokespersonintheircondition.
A pretest was administered across all conditions to collect demographic data, which
included a political ideology scale that asked subjects to rate themselves politically on a
scale of (1) to (5) in which (1) = Strongly Progressive, (2) = Moderately Progressive, (3)
=Centrist,(4)=ModeratelyConservative,and(5)=StronglyConservative.
These three variables (spokesperson expertise, message elaboration, and political
ideology)comprisedtheprimaryeffectsexaminedinthisstudy.
Sample: Participants
The experiment’s subjects consisted of a demographically diverse sample of 474 adults
recruited online through two separate online platforms, the Amazon Mechanical Turk
workforce marketplace, and LinkedIn members of a number of interest groups relevant to
healthcareandmarketingcommunications.
The overall sample of 474 adults ranged in age from 18 to 82, and included 215 (45.4%)
males and 259 (54.6%) females. The average age was 35.1 years, and the average
education level was nearly two years of college. Politically, the sample included 88
(18.6%) who identified themselves as Republicans, 195 (41.1%) who selfidentified as
Independents, 181 (38.2%) who selfidentified as Democrats, and 10 (2.1%) who
identifiedthemselvesas“Other”.
Manipulation: Expert and Non-Expert Message Sources
Based on information from the Centers for Disease Control and a proHPV vaccination
message developed for a prior study (Kahan et al., 2010), a 281word advocacy message
was developed using language that could reasonably be attributed to either an expert or a
6
Proceedings of the New York State Communication Association, Vol. 2013 [2014], Art. 3
http://docs.rwu.edu/nyscaproceedings/vol2013/iss2013/3
nonexpert spokesperson as the message source. This advocacy message was repeated
wordforwordasaconstantinbothmessagesourceconditions(SeeAppendixA).
Participants were randomly assigned to one of three advocacy message conditions.
Roughly onethird received a proHPV vaccination advocacy message attributed to an
authoritativelooking spokesperson identified as a physician; another third received an
identical message attributed to a female middleschool student. The identities were
entirely fictitious, with spokesperson photos obtained from a professional stockphoto
service. (See Figure 1, below.) Participants assigned to the Control condition received no
advocacy message and were instructed to skip the spokesperson evaluation (see
ManipulationCheck,below)andmessageelaborationquestions.
Figure 1. Images and Identifiers of Expert and Non-Expert Spokespeople
Manipulation Check
As a manipulation check, the subjects in the two experimental conditions were asked to
rate the spokesperson delivering the advocacy message using a sevenpoint scale on three
different attributes: honesty/trustworthiness, likability, and authoritative expertise. There
was no significant difference in honesty/trustworthiness between the Expert
Spokesperson and NonExpert Spokesperson, but the intended difference in perception of
expertise was confirmed (Expert: M = 5.54, SD = 1.31; NonExpert: M = 3.70, SD =
1.66; t(312) = 10.92, p < .001 (twotailed)). The nonexpert was rated a bit higher in
likability than the expert (Expert: M = 5.43, SD = 1.22; NonExpert: M = 5.73, SD = 1.16;
t(310)=2.23,p<.05(twotailed)).
Measures
PRETEST
The survey pretest collected information about standard demographic variables such as
age, gender, education, political and religious affiliation. Most significant for the purposes
of the study was the variable of political ideology, for which subjects were asked to rate
themselves on a scale of 1 to 5 along a continuum from Strongly Progressive to Strongly
Conservative. For statistical analysis, responses were recoded to combine the Strongly
Progressive and Moderately Progressive responses as a single level of a threelevel
variable, with Centrist responses as a second level, and the Moderately Conservative and
StronglyConservativeresponsesasathirdlevel.
7
Gans: HPV Vaccination Advocacy and Source Expertise
Published by DOCS@RWU, 2014
As a check on the predictive validity of these measures of political ideology, both the
5point scale (M = 2.74, SD = 1.17) and the 3point scale (M = 1.82, SD = 0.87) were
compared with scores of attitude toward HPV vaccination (M = 25.95, SD = 11.29) for
subjectsinthecontrolcondition,andweakbutsignificantrelationshipswerefound:
Political Ideology (3level variable)/Attitude toward HPV Vaccination: Pearson
correlation=.215,p<.01,2tailed(F=3.96,df=2,157;p<.05;etasquared=0.49)
Political Ideology (5level variable)/Attitude toward HPV Vaccination: Pearson
correlation=.257,p<.01,2tailed(F=3.06,df=4,157;p<.05;etasquared=0.74)
Figure 2. Means of Attitude toward HPV Vaccination by Political Ideology (Control)
The pattern of attitude toward HPV vaccination by political ideology in this control
condition is illustrated in Figure 2, above. Conservatives were more opposed to HPV
vaccination than Centrists, and Centrists more opposed than Progressives. As indicated by
a oneway analysis of variance, Political Ideology had a significant effect on attitude
toward HPV vaccination in this control condition (F = 14.91, df 2, 157; p<.001),
accounting for approximately 16% of the variance as indicated by an eta squared value of
.160. Post hoc Scheffé tests indicated that while Centrists were not significantly different
from Progressives (F = 1.74, df 2, 157; p = .22), they were significantly different from
Conservatives (F = 2.92, df 2, 157; p = .016), and Conservatives were significantly
different from Progressives (F = 5.46, df 2, 157; p<.001). This analysis of variance will be
examinedingreaterdetailintheresultssection,below.
8
Proceedings of the New York State Communication Association, Vol. 2013 [2014], Art. 3
http://docs.rwu.edu/nyscaproceedings/vol2013/iss2013/3
POSTMANIPULATIONMEASUREMENTS
Message Elaboration Scale: In addition to the manipulation check on spokesperson  
expertise noted above, subjects in the Expert and NonExpert conditions were presented
with a series of seven questions to measure their degree of attention and thought given to
the actual advocacy message. The sum of the responses to these message elaboration
questions was used to create a Message Elaboration Scale (Cronbach’s α = .73, M =
34.04,SD=6.55).
DEPENDENTVARIABLE
Attitude toward HPV vaccination: Subjects in all three conditions of the study (Control,    
Expert, NonExpert) answered a series of questions designed to measure their attitudes
and behavioral intentions regarding HPV vaccination. A series of seven questions
addressed subjects’ perceptions about HPV vaccination in terms of likedislike, good
ideabad idea, riskysafe, and effectiveineffective, using a sevenpoint Likerttype scale.
Questions in which proHPV attitudes called for lower scores were reverse coded. The
results of these questions were combined into an overall attitude scale (Cronbach’s α =
.93;M=20.67,SD=10.48).
Results
Attitude toward HPV Vaccination as influenced by Spokesperson Expertise across
three levels of Political Ideology: Split File Analysis
Because of the significant differences between subjects in the three different levels of the
political ideology variable, these levels were examined as different populations, and a
oneway analysis of variance was used to analyze the effects of the three levels of the
spokesperson variable on attitude toward HPV vaccination for subjects in each of the
three political ideology levels separately. Examination of Levene’s Test of Equality of
Error Variances and plots of residuals indicated that assumptions of homogeneity,
normality and linearity were met for all three conditions. Means and standard deviations of
HPV vaccination attitude scores by source expertise and political ideology are presented
inTable1,below.AsummaryoftheresultsoftheanalysiscanbefoundinTable2.
Table 1. Means and Standard Deviations for Attitude toward HPV Vaccination by Source
Expertise and Political Ideology
Spokesperson
Expertise
Political
Ideology
n
mean
Standard
Deviation
Control
Progressive
77
29.90
9.66
Centrist
35
26.20
9.01
Conservative
48
19.44
12.36
Expert
Progressive
80
33.06
7.65
Centrist
47
26.45
10.40
Conservative
29
23.97
10.65
NonExpert
Progressive
80
31.80
9.04
Centrist
35
28.29
9.76
Conservative
43
27.00
10.61
9
Gans: HPV Vaccination Advocacy and Source Expertise
Published by DOCS@RWU, 2014
Table 2. Analysis of Variance: Attitude toward HPV Vaccination by Source Expertise and
Political Ideology
Source
SS
df
F
Progressive
Spokesperson
Expertise
397.85
2
2.56
Error
18170.66
234
Total
18568.51
236
Centrist
Spokesperson
Expertise
93.95
2
0.18
Error
10974.36
114
Total
11068.31
116
Conservative
Spokesperson
Expertise
1317.21
2
5.11**
Error
15084.78
117
Total
16401.99
119
**p < .01
As can be seen in Table 2, above, spokesperson expertise had a significant effect on
attitude toward HPV vaccination with subjects who identify themselves as Conservative
in political ideology (F = 5.11, df = 2, 117; p<.01), accounting for 8.0% of the variance in
attitude among Conservatives (eta squared = .080). Among Conservatives, post hoc
Scheffé tests indicated a significant difference between attitudes of subjects from the
control group and attitudes of subjects exposed the provaccination message delivered by
thenonexpertspokesperson(F=3.17,df=2,117;p<.01).
Figure 3. Means of Attitude toward HPV Vaccination among Conservatives
10
Proceedings of the New York State Communication Association, Vol. 2013 [2014], Art. 3
http://docs.rwu.edu/nyscaproceedings/vol2013/iss2013/3
As shown in Figure 3, above, Conservatives who were exposed to the provaccination
advocacy message from the nonexpert spokesperson expressed significantly more
positive attitudes toward HPV vaccination than those in the control group. Interestingly,
while exposure to the same message from the expert spokesperson also produced more
positive attitudes than the control condition, this exposure was not statistically significant
(F=1.70,df=2,117;p=.242).
While the difference between the expert and nonexpert conditions does not achieve
statistical significance (p = .268 as measured by a Helmert contrast), comparison of the
attitudes of subjects exposed to each spokesperson condition with attitudes of subjects in
the control condition suggests a fairly strong case for the nonexpert as the more effective
spokesperson when delivering a proHPV vaccination message to a conservative
audience. This finding provides positive support for the moderating role of political
ideology (RQ2), and directional support for the comparative persuasive effectiveness of
theNonExpertspokesperson(RQ1).
Analysis of Message Elaboration by Spokesperson Expertise and Political Ideology
To examine the effects of spokesperson expertise and political ideology on degree of
elaboration about the spokespersondelivered advocacy message, a oneway analysis of
variance was used to analyze the data using a splitsample approach. Message elaboration
in each of the three levels of the political ideology variable was examined separately for
the expert spokesperson condition and for the nonexpert spokesperson condition. Tests of
homogeneity of variance and plots of residuals indicated that assumptions of homogeneity,
normality and linearity were met for both conditions (Levene’s Test of Equality of Error
Variances: Expert, F = 1.15, df = 2, 153; p>.05; NonExpert, F = 1.69, df = 2, 155; p>.05).
Means and standard deviations for message elaboration by political ideology and source
expertisecanbefoundinTable3,below.ResultsofthisanalysiscanbefoundinTable4.
Table 3. Means and Standard Deviations for Message Elaboration by Source Expertise
and Political Ideology
Spokesperson
Expertise
PoliticalIdeology
n
mean
Standard
Deviation
Expert
Progressive
80
35.30
7.38
Centrist
47
31.66
5.69
Conservative
29
33.48
7.19
NonExpert
Progressive
80
34.31
5.59
Centrist
35
33.69
7.44
Conservative
43
34.47
5.57
Table 4. Analysis of Variance: Elaboration by Source Expertise and Political Ideology
11
Gans: HPV Vaccination Advocacy and Source Expertise
Published by DOCS@RWU, 2014
Source
SS
df
MS
F
ExpertSpokesperson
PoliticalIdeology
397.58
2
198.79
4.20*
Error
7238.60
153
47.31
Total
7636.17
155
NonExpert
Spokesperson
PoliticalIdeology
13.26
2
6.63
0.18
Error
5747.43
155
37.08
Total
5760.68
157
*p < .05
As can be seen in Table 4, above, political ideology had a significant effect on message
elaboration in the expert spokesperson condition (F = 4.20, df = 2, 153; p<.05; eta squared
= .052), but not in the nonexpert condition (F = 0.18, df = 2, 155; p>.15; eta squared =
.002). A Difference contrast indicated a significant difference between Centrists and
Progressives in the Expert condition (F = 2.88, df = 2, 153; p<.01), with Centrists scoring
significantly lower on the elaboration scale than Progressives. A Helmert contrast also
indicated a significant difference between Progressives and all others (F = 2.44, df = 2,
153; p<.05), with Centrists and Conservatives, taken together, scoring significantly lower
thanProgressivesontheelaborationscale.
These findings suggest that subjects likely to be less supportive of HPV vaccination tend
to think less deeply when they encounter a proHPV advocacy message delivered by an
expert, and they support the notion that a nonexpert spokesperson can be a strong
alternative to an expert in some circumstances. It can be inferred from these findings that
when confronted with an expert spokesperson, selfdescribed Centrists are less likely to
pay attention to or think deeply about the spokesperson’s message than either
Progressives or Conservatives, with a significant gap between Centrists’ level of
elaboration and that of Progressives. Additionally, these findings indicate that, while
people of differing political ideologies react differently in terms of thought processing to
messages delivered by an expert spokesperson, there seem to be no significant
differences between their levels of attention and elaboration when the spokesperson is not
an expert. While the differences are not significantly large, Centrists and
Conservatives—groups that tend to oppose HPV vaccination—seem to pay more
attention and think more deeply when confronted with a proHPV vaccination advocacy
messagefromanonexpertsourcethanfromanexpertsource.
On balance, if the goal is to encourage the greatest number of HPVvaccination opposers
to at least consider changing their position, these findings suggest that choosing a
nonexpert spokesperson would be at least as effective as choosing an expert
spokesperson, and particularly where Centrists are concerned, probably significantly more
effective.
Another inference from these results applies to the frequent reliance on expert
spokespeople for HPV vaccination advocacy and, by extension, perhaps for
communicating sciencebased behavioral recommendations in general. Since those in
12
Proceedings of the New York State Communication Association, Vol. 2013 [2014], Art. 3
http://docs.rwu.edu/nyscaproceedings/vol2013/iss2013/3
favor of HPV vaccination seem to think more deeply about a provaccination message
when it is delivered by an expert than when it is delivered by a nonexpert, it may be
natural for them to assume that people of all vaccinationpredispositions react that way.
This may be just the sort of “natural” assumption about risk communication that should be
evaluated before being put into widespread, unexamined practice to prevent valuable
resources from being invested in efforts that are ineffective or do more harm than good
(Fischhoff,Brewer&Downs,2011;Fishbeinetal.,2002;Wilson,2011).
Discussion
The lack of universal acceptance for—and, in fact, the increasing public resistance to—an
apparently safe, effective and lifesaving public health practice such as HPV vaccination
illustrates a significant issue in the communication of behavioral recommendations based
onevidencebasedscientificdataandconsensusviewsofscientificandmedicalexperts.
A review of the related literature examining this arena suggests that at least some of the
efforts to communicate such behavioral recommendations may have suffered from
misguided intuitive assumptions and/or lack of sound theoretical foundations (Betsch et al.,
2012; Fischhoff, Brewer & Downs, 2011; Fishbein et al., 2002; Glanz and Bishop, 2004;
Miche and Abraham, 2004; Wilson, 2011). It also suggests that the recipients of such
behavioral recommendations do not present a conveniently monolithic, homogenous and
predictably rational or sympathetic audience (Hamilton, 2010, 2011; Mooney, 2005, 2012;
Newport, 2012; Public Religion Research Institute, 2011; Ramanadhan and Viswanath,
2006).
While there is value in motivating those already predisposed to comply with
provaccination advocacy messages to take positive action, the current research is more
concerned with the apparently growing numbers of parents and policymakers who are
resisting those messages. The current study provides empirical support for the role of
political ideology as an influential factor in attitude toward HPV vaccination, with
Centrists and Conservatives significantly less in favor of HPV vaccination than
Progressives. This suggests that efforts to address issues of growing resistance to the
HPV vaccination might be more productive if they focus on identifying strategies that are
more effectively influential with Centrists and Conservatives than with a more generalized
population that includes Progressives. The significant differences in attitude toward HPV
vaccination found between Centrists and Conservatives also suggest it might be more
effective to develop strategies targeting each group separately rather than lumping them
togetherasasinglegroup.
In the current study, significant effects for the influence of spokesperson expertise on
message elaboration were found when comparing Progressives to Centrists as well as to
the combined population of Centrists and Conservatives in the sample. Centrists as well as
Centrists and Conservatives taken together were significantly less likely than Progressives
to think deeply about the proHPV vaccine advocacy message when it was delivered by a
seriouslooking authoritative expert. Whether the seriouslooking expert suppressed their
message elaboration or the young girl stimulated it is impossible to deduce from this
research design. However, it is clear that for encouraging attention and thought about a
provaccine message among those most likely to harbor predispositions to resist the
13
Gans: HPV Vaccination Advocacy and Source Expertise
Published by DOCS@RWU, 2014
behavioral recommendations it carries (by virtue of their political ideology, at least), these
findings support the choice of the nonexpert young girl rather than the seriouslooking
authoritativeexpertasthespokespersontodeliverthemessage.
Since it can be assumed from the literature (Kahan et al., 2010, for instance) and the
current findings that those who are the most active public advocates for HPV vaccination
are likely to be Progressives, it is reasonable to suggest that they have been choosing their
messaging strategies based on their own sensibilities and intuitions—perhaps leading to an
unexamined overreliance on the scientifically based exhortations of experts who may, in
fact, be undermining their own cause (as suggested by Burgoon et al., 2002; Fishbein et
al.,2002;Milleretal.,2006;andWilson,2011).
In general, the current study suggests that for delivering proHPV vaccination advocacy
messages to Centrists and Conservatives, a nonexpert spokesperson may well be more
effectivelypersuasivethananexpert.
Study Limitations and Implications for Future Research
In order to provide a baseline reference for attitude toward HPV vaccination, the control
condition featured no spokesperson or advocacy message and therefore there was also no
measurement of message elaboration for subjects in the control condition. Without a
baseline elaboration measurement, it is impossible to know whether the conditions of
spokesperson expertise in this study stimulated or suppressed message elaboration. To
address this issue in a future study, a second control condition could be added in which
subjects were exposed to the advocacy message without any source attribution, or with
the message presented as a newspaper article. Subjects in this second control condition
could then be asked to answer the same message elaboration measurement questions as
subjects in the spokesperson conditions. Another approach to this issue might include
pretests using Cacioppo, Petty & Kao’s (1984) 18item “Need for Cognition” scale,
whichwouldhelpinformanalysisoftheotherelaborationmeasures.
The current study examined political ideology as a moderating factor. Additional
moderating factors that seem worthy of examination regarding attitudes toward HPV
vaccination in particular but that also may be relevant to a range of other politically
charged issues include gender, religious observance, family composition and role,
education,geographyandattitudestowardotherpoliticalandsocialissues.
In addition to supporting the broader and more frequent use of nonexpert spokespeople to
deliver politically charged healthandscience related advocacy messages, this study
suggests a need for further study of the mediating factor of message elaboration in the
persuasive effectiveness of such messages as well as the broader use of heuristic cues in
those messages. Since the current study was intended to examine effects of source
expertise on advocacy message effectiveness, the message was constructed so it could
be presented by either an expert or a nonexpert spokesperson without appearing to be an
unnatural expression from either source. The findings provided support for the persuasive
effectiveness of the advocacy message as compared to the control condition, regardless
of the expertise of the spokesperson. Future studies might address the relative persuasive
power of different messaging strategies and their possible interactions with different
14
Proceedings of the New York State Communication Association, Vol. 2013 [2014], Art. 3
http://docs.rwu.edu/nyscaproceedings/vol2013/iss2013/3
target audiences. As suggested above, a persuasive appeal strong on heuristic cues might
be more effective than one full of scientific evidence, particularly for a
lowinformationseeking audience. Instead of featuring headlines carrying an implication
of hard cognitive work ahead, such as “Centers for Disease Control Provide Statistical
Evidence of Effectiveness of HPV Vaccination,” low information seekers might be better
persuaded by messages requiring less intense attention, such as, “Don’t be the last one on
yourblocktogetyourchildvaccinated.”
References
AllianceforAuditedMedia(2013)Top25U.S.consumermagazinesforthesecondhalf
of2012.AllianceforAuditedMedia,TheNewAuditBureauofCirculations
[ONLINE]Retrievedfrom
http://www.auditedmedia.com/news/blog/top25usconsumermagazinesbytotal
paidandverifiedcirculation.aspx
Betsch,C.,Brewer,N.T.,Brocard,P.,Davies,P.,Gaissmaier,W.,Haase,N.,Leask,J.,
Renkewitz,F.,Renner,B.,Reyna,V.F.,Rossmann,C.,Sachse,K.,Schachinger,
A.,Siegrist,M.,&Stryk,M.(2012).OpportunitiesandchallengesofWeb2.0for
vaccinationdecisions,Vaccine30,3727–3733.
Burgoon,M.,Alvaro,E.,Grandpre,J.,&Voloudakis,M.(2002).Revisitingthetheoryof
psychologicalreactance:Communicatingthreatstoattitudinalfreedom.InJ.
Dillard&M.Pfau(Eds.),Thepersuasionhandbook:Theoryandpractice
(pp.213–232).ThousandOaks,CA:Sage.
Cacioppo,J.T.,&Petty,R.E.(1984).Theelaborationlikelihoodmodelofpersuasion.
AdvancesinConsumerResearch,11(1),673–675.
Cacioppo,J.T.,Petty,R.E.,&Kao,C.F.(1984).Theefficientassessmentofneedfor
cognition.JournalofPersonalityAssessment,48,306–307.
CentersforDiseaseControlandPrevention(2013a).HPVvaccineforpreteensand
teens.Retrievedfromhttp://www.cdc.gov/vaccines/who/teens/vaccines/hpv.html
CentersforDiseaseControlandPrevention(2013b)Teenvaccinationcoverage:2011
NationalImmunizationSurvey(NIS)–Teen.Retrievedfrom
http://www.cdc.gov/vaccines/who/teens/vaccinationcoverage.html
Clark,J.K.,Evans,A.T.,&Wegener,D.T.(2011).Perceptionsofsourceefficacyand
persuasion:Multiplemechanismsforsourceeffectsonattitudes.European
JournalofSocialPsychology,41,596–607.
Clark,J.K.,Wegener,D.T.,Habashi,M.M.,&Evans,A.T.(2012).Sourceexpertise
andpersuasion:Theeffectsofperceivedoppositionorsupportonmessage
scrutiny.PersonalityandSocialPsychologyBulletin,38(1),90–100.
Darden,P.M.,Thompson,D.M.,Roberts,J.R.,Hale,J.J.,Pope,C.,Naifeh,M.,&
Jacobson,R.M.(2013).Reasonsfornotvaccinatingadolescents:National
15
Gans: HPV Vaccination Advocacy and Source Expertise
Published by DOCS@RWU, 2014
ImmunizationSurveyofTeens:2008–2010.Pediatrics,131,645–651.Retrieved
from
http://pediatrics.aappublications.org/content/early/2013/03/12/peds.20122384
Fischhoff,B.,Brewer,N.T.,&Downs,J.S.(Eds).(2011).Communicatingrisksand
benefits:Anevidencebaseduser’sguide.FDA.Retrievedfrom
http://www.fda.gov/AboutFDA/ReportsManualsForms/Reports/ucm268078.htm
Fishbein,M.,HallJamieson,K.,Zimmer,E.,vonHaeften,I.,&Nabi,R.(2002).Avoiding
theboomerang:Testingtherelativeeffectivenessofantidrugpublicservice
announcementsbeforeanationalcampaign.AmericanJournalofPublic
Health,92(2),238–245.
Glanz,K.&Bishop,D.B.(2010).Theroleofbehavioralsciencetheoryindevelopment
andimplementationofpublichealthinterventions.AnnualReviewofPublic
Health,31,399–418.
Gorman,B.(2011).Fridaycable:“Smackdown”topsnight;HurricaneIreneboosts
WeatherChannel.ZAP2it.com:TVbythenumbers.Retrievedfrom
http://tvbythenumbers.zap2it.com/2011/08/29/fridaycablesmackdowntopsnight
hurricaneireneboostsweatherchannelhaventhundercatsmore/101630/
Hamilton,L.C.(2011).Climatechange:Partisanship,understanding,andpublicopinion.
CarseyInstituteIssueBrief,26.Retrievedfrom
http://carseyinstitute.unh.edu/publication/climatechangepartisanshipunderstandin
gandpublicopinion
Hamilton,L.C.(2010).Doscientistsagreeaboutclimatechange?Publicperceptions
fromaNewHampshiresurvey.CarseyInstituteIssueBrief,22.Retrievedfrom
http://carseyinstitute.unh.edu/sites/carseyinstitute.unh.edu/files/publications/IB_Ha
milton_Climate_Survey.pdf
Kahan,D.M.,JenkinsSmith,H.,&Braman,D.(2011).Culturalcognitionofscientific
consensus.JournalofRiskResearch,14(2),147–174.
Kahan,D.,Braman,D.,Cohen,G.,Gastil,J.,&Slovic,P.(2010).WhofearstheHPV
vaccine,whodoesn’t,andwhy?Anexperimentalstudyofthemechanismsof
culturalcognition.Law&HumanBehavior,34(6),501–516.
Karmarkar,U.R.,&Tormala,Z.L.(2010).Believeme,IhavenoideawhatI’mtalking
about:Theeffectsofsourcecertaintyonconsumerinvolvementand
persuasion.JournalofConsumerResearch,36(6),1033–1049.
Kaufman,M.(2006,June9).FDAapprovesvaccinethatshouldpreventmostcervical
cancers.WashingtonPost,p.A1.
Kondolojy,A.(2012a).Tuesdaycableratings:“TeenMom”winsnight.ZAP2it.com:TV
bythenumbers.Retrievedfrom
http://tvbythenumbers.zap2it.com/2012/08/15/tuesdaycableratingsteenmomwin
16
Proceedings of the New York State Communication Association, Vol. 2013 [2014], Art. 3
http://docs.rwu.edu/nyscaproceedings/vol2013/iss2013/3
snightrizzoliislesstoragewarshardcorepawncountingcarshowjawschange
dtheworldmore/144966/
Kondolojy,A.(2012b).Thursdaycableratings:“Suits”winsnight.ZAP2it.com:TVby
thenumbers.Retrievedfrom
http://tvbythenumbers.zap2it.com/2012/08/17/thursdaycableratingssuitsburnnot
icegreatwhitehighwaysullivansonangermanagementmore/145216/
Lemanski,J.L.,&Lee,HS.(2012).Attitudecertaintyandresistancetopersuasion:
Investigatingtheimpactofsourcetrustworthinessinadvertising.International
JournalofBusiness&SocialScience,3(1),66–75.
Miche,S.,&Abraham,C.(2004).Interventionstochangehealthbehaviors:
Evidencebasedorevidenceinspired?PsychologyandHealth,19(1),29–49.
Miller,C.H.,Burgoon,M.,Grandpre,J.R.,&Alvaro,E.M.(2006).Identifyingprincipal
riskfactorsfortheinitiationofadolescentsmokingbehaviors:Thesignificanceof
psychologicalreactance.HealthCommunication,19(3),241–252.
Mooney,C.(2012).Therepublicanbrain:Thescienceofwhytheydeny
science—andreality.Hoboken,NJ:JohnWileyandSons.
Mooney,C.(2005).Therepublicanwaronscience.NewYork:BasicBooks.
Newport,F.(2012,June1).InU.S.,46%holdcreationistviewofhumanorigins:Highly
religiousAmericansmostlikelytobelieveincreationism.GallupPolitics.
Retrievedfrom
http://www.gallup.com/poll/155003/holdcreationistviewhumanorigins.aspx
Petty,R.E.,&Briñol,P.(2008).Persuasion:Fromsingletomultipletometacognitive
processes.PerspectivesonPsychologicalScience,3(2),137–147.
Petty,R.E.,&Cacioppo,J.T.(1977).Forewarning,cognitiveresponding,andresistance
topersuasion.JournalofPersonalityandSocialPsychology,35(9),645–655.
Petty,R.E.,&Cacioppo,J.T.(1979).Effectsofforewarningofpersuasiveintentand
involvementoncognitiveresponses.PersonalityandSocialPsychology
Bulletin,5(2),173–176.
Petty,R.E.,&Cacioppo,J.T.(1984).Sourcefactorsandtheelaborationlikelihood
modelofpersuasion.AdvancesinConsumerResearch,11(1),668–672.
Pornpitakpan,C.(2004).Thepersuasivenessofsourcecredibility:Acriticalreviewof
fivedecades’evidence.JournalofAppliedSocialPsychology,34(2),
243–281.
17
Gans: HPV Vaccination Advocacy and Source Expertise
Published by DOCS@RWU, 2014
Pruitt,S.L.,&Schootman,M.(2010).Geographicdisparity,areapoverty,andhuman
papillomavirusvaccination.AmericanJournalofPreventiveMedicine,38(5),
525–533.
PublicReligionResearchInstitute.(2011,September22).Survey:Climatechangeand
evolutioninthe2012elections.MonthlyReligionNewsSurveys.Retrievedfrom
http://publicreligion.org/research/2011/09/climatechangeevolution2012/
Ramanadhan,S.,&Viswanath,K.(2006).Healthandtheinformationnonseeker:A
profile.HealthCommunication,20(2),131–139.
Sternthal,B.,Phillips,L.W.,&Dholakia,R.(1978).Thepersuasiveeffectofsource
credibility:Asituationalanalysis.PublicOpinionQuarterly,42(3),285–296.
Silvia,P.J.(2006).Reactanceandthedynamicsofdisagreement:Multiplepathsfrom
threatenedfreedomtoresistancetopersuasion.EuropeanJournalofSocial
Psychology,36(5),673–685.
Taber,C.S.,&Lodge,M.(2006).Motivatedskepticismintheevaluationofpolitical
beliefs.AmericanJournalofPoliticalScience,50(3),755–769.
Tormala,Z.L.,Briñol,P.,&Petty,R.E.(2007).Multiplerolesforsourcecredibility
underhighelaboration:It’sallinthetiming.SocialCognition,25(4),536–552.
Tormala,Z.L.,Clarkson,J.J.,&Petty,R.E.(2006).Resistingpersuasionbytheskinof
one’steeth:Thehiddensuccessofresistedpersuasivemessages.Journalof
Personality&SocialPsychology,91(3),423–435.
Tormala,Z.L.,&Petty,R.E.(2004).Sourcecredibilityandattitudecertainty:A
metacognitiveanalysisofresistancetopersuasion.JournalofConsumer
Psychology,14(4),427–442.
Tormala,Z.L.,&Petty,R.E.(2002).Whatdoesn’tkillmemakesmestronger:The
effectsofresistingpersuasiononattitudecertainty.JournalofPersonality&
SocialPsychology,83(6),1298–1313.
Wilson,E.J.(1987).Sourceeffectsinpersuasionexperiments:Ametaanalysis.
AdvancesinConsumerResearch,14(1),577–577.
Wilson,T.D.(2011).Redirect:Thesurprisingnewscienceofpsychologicalchange.
NewYork:Little,BrownandCompany.
 
18
Proceedings of the New York State Communication Association, Vol. 2013 [2014], Art. 3
http://docs.rwu.edu/nyscaproceedings/vol2013/iss2013/3
APPENDIXA:AdvocacyMessagesandSpokespersonVariable
Intheexperimentalconditions,thecontentoftheadvocacymessagewasconstantexcept
fortheappearanceandidentificationofthespokesperson.
“Expert”
 
19
Gans: HPV Vaccination Advocacy and Source Expertise
Published by DOCS@RWU, 2014
“NonExpert”
20
Proceedings of the New York State Communication Association, Vol. 2013 [2014], Art. 3
http://docs.rwu.edu/nyscaproceedings/vol2013/iss2013/3
ResearchGate has not been able to resolve any citations for this publication.
Article
Full-text available
The Elaboration Likelihood Model of persuasion (ELM) is discussed as it relates to source factors in persuasion. The ELM proposes that under low elaboration likelihood, source factors serve as simple acceptance or rejection cues under moderate elaboration likelihood source factors guide the extent of thinking; and under high elaboration likelihood source factors are unimportant as cues or general motivators of thought but (if relevant) serve as persuasive arguments or help in interpreting arguments. Several experiments are described which provide empirical support for these propositions.
Article
Full-text available
This critical review assesses whether evaluation studies can answer three key questions about behaviour change interventions: 'Do they work? How well do they work? How do they work?' Reviews of intervention evaluations are examined, particularly those addressing decreasing unprotected sexual intercourse and smoking. Selection of outcome measures and calculation of effect sizes are discussed. The article also considers the extent to which evaluation reports specify (i) discrete intervention techniques and (ii) psychological mechanisms that account for observed behavioural change. It is concluded that intervention descriptions are often not specific about the techniques employed and that there is no clear correspondence between theoretical inspiration and adoption of particular change techniques. The review calls for experimental testing of specific theory-based techniques, separately and in combination.
Article
Full-text available
Past research suggests that under high elaboration conditions, source credibility can play more than one role in persuasion. In particular, source credibility can affect the valence of people's thoughts generated in response to persuasive messages or it can affect the confidence with which people hold those thoughts. In the present research, two experiments explore the conditions under which these conceptually distinct effects occur. It. is demonstrated that the effect of source credibility on thought confidence is dominant when source information follows, rather than precedes, a persuasive message. When source information precedes a message, it affects the valence of issue-relevant thinking.
Article
Full-text available
In 2010, Carsey Institute researchers began including three new questions about climate change on a series of regional surveys. They asked how much people understand about the issue of global warming or climate change; whether they think that most scientists agree that climate change is happening now as a result of human activities; and what they believe personally about the topic. Survey results show that while large majorities agree that climate change is happening now, they split on whether this is attributed mainly to human or natural causes. Brief author Lawrence Hamilton concludes that most people gather information about climate change not directly from scientists but indirectly—through news media, political activists, acquaintances, and other non-science sources. Their understanding reflects not simply scientific knowledge, but rather the adoption of views promoted by political or opinion leaders they follow. While public beliefs about physical reality remain strikingly politicized, leading science organizations agree that human activities are now changing the Earth’s climate. Interestingly, the strong scientific agreement on this point contrasts with the partisan disagreement seen on all of the Carsey surveys.
Article
Full-text available
This report, a collaboration of the Carsey Institute, the UNH Survey Center, and the UNH Office of Sustainability, is the first of a new initiative that will track public perceptions about climate change as they change over time. Questions related to climate change were asked as part of New Hampshire's Granite State Poll, which surveyed 512 New Hampshire residents in April 2010.
Article
College undergraduates were either warned or not warned of the persuasive intent of a communication which was either of direct or only indirect personal relevance to them. Consistent with the hypothesis based on reactance theory (Brehm, 1966), the inhibiting effect of the forewarning (i.e., reduced persuasion, increased counterargumentation, and reduced favorable thoughts) was greater under high than low involvement conditions.
Article
We propose a model of motivated skepticism that helps explain when and why citizens are biased-information processors. Two experimental studies explore how citizens evaluate arguments about affirmative action and gun control, finding strong evidence of a prior attitude effect such that attitudinally congruent arguments are evaluated as stronger than attitudinally incongruent arguments. When reading pro and con arguments, participants (Ps) counterargue the contrary arguments and uncritically accept supporting arguments, evidence of a disconfirmation bias. We also find a confirmation bias—the seeking out of confirmatory evidence—when Ps are free to self-select the source of the arguments they read. Both the confirmation and disconfirmation biases lead to attitude polarization—the strengthening of t2 over t1 attitudes—especially among those with the strongest priors and highest levels of political sophistication. We conclude with a discussion of the normative implications of these findings for rational behavior in a democracy.