influence of message framing, involvement, and … · traditional anti-smoking psa research rather...
TRANSCRIPT
INFLUENCE OF MESSAGE FRAMING, INVOLVEMENT, AND NICOTINE DEPENDENCE ON EFFECTIVENESS OF ANTI-SMOKING PUBLIC SERVICE
ANNOUNCEMENTS
By
WAN SEOP JUNG
A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ADVERTISING
UNIVERSITY OF FLORIDA
2008
1
© 2008 Wan Seop Jung
2
To my wife and parents
3
ACKNOWLEDGMENTS
This work is an accomplishment of all who helped me directly and indirectly. It is
especially dedicated to my beloved wife, Soo, who has encouraged and supported me. It is also
dedicated to my parents, Dong Kun Jung and Bok Sook Seo; without their helping hands, it
would have been impossible to complete this thesis.
I extend my sincere gratitude to my advisor and chair, Dr. Jorge Villegas, who has guided,
encouraged, and supported me with the warmest heart and has given me tremendous guidance
and support. I also want to express my thanks to my wonderful committee members, Dr. Chang-
Hoan Cho and Dr. Hyojin Kim, for their invaluable advice and encouragement. I appreciate all
the time and efforts they put into helping me with my thesis, and challenging me to learn more.
Although I first assumed that graduate schools are not fun, my adorable friends at UF
proved my assumption to be wrong. I express my thanks to my classmates. I will never forget
those wonderful moments spent with them. I especially appreciate Korean gators, Jay, Hyung
Seok, Jun Seong, Jun, Sang Won, Chang Dea, Mihyun, Yeuseung, Hyunji, Minji, Hyunmin,
Seonmi, Sooyeon and all other sisters and brothers, for helping me to go through all of those
hard tasks.
4
TABLE OF CONTENTS page
ACKNOWLEDGMENTS ...............................................................................................................4
LIST OF TABLES...........................................................................................................................7
LIST OF FIGURES .........................................................................................................................8
LIST OF ABBREVIATIONS..........................................................................................................9
ABSTRACT...................................................................................................................................10
CHAPTER
1 INTRODUCTION ..................................................................................................................11
2 LITERATURE REVIEW .......................................................................................................15
Message Framing....................................................................................................................15
Involvement ............................................................................................................................17
Connection of Involvement with Message Framing...............................................................19
Nicotine Dependence..............................................................................................................21
Connection of Nicotine Dependence Level with Message Framing ......................................22
3 METHODOLOGY .................................................................................................................24
4 RESULTS...............................................................................................................................29
Hypothesis 1 ...........................................................................................................................29
Manipulation Check................................................................................................................29
Involvement ............................................................................................................................30
Test of Hypothesis 1 ...............................................................................................................30
Hypothesis 2 ...........................................................................................................................31
Manipulation Check................................................................................................................32
Nicotine Dependence..............................................................................................................32
Test of Hypothesis 2 ...............................................................................................................33
5 DISCUSSION AND CONCLUSION ....................................................................................38
Discussion...............................................................................................................................38
Implications ............................................................................................................................39
Limitations and Future Research ............................................................................................40 APPENDIX
A INFORMED CONSENT FORM............................................................................................44
5
B QUESTIONNAIRE ................................................................................................................45
LIST OF REFERENCES...............................................................................................................49
BIOGRAPHICAL SKETCH .........................................................................................................56
6
LIST OF TABLES
Table page 4-1 Manipulation Check of Positively Framed Message for Item 1 in H1 Testing ....................35
4-2 Manipulation Check of Negatively Framed Message for Item 2 in H1 Testing...................35
4-3 Descriptive Statistics of H1 ..................................................................................................35
4-4 Model Statistics of H1 ..........................................................................................................35
4-5 Regression of H1...................................................................................................................35
4-6 Follow-up Analyses for Interaction in H1 ............................................................................36
4-7 Manipulation Check of Positively Framed Message for Item 1 in H2 Testing ....................36
4-8 Manipulation Check of Negatively Framed Message for Item 2 in H2 Testing...................36
4-9 Descriptive Statistics of H2 ..................................................................................................36
4-10 Model Statistics of H2 ..........................................................................................................36
4-11 Regression of H2...................................................................................................................37
4-12 Follow-up Analyses for Interaction in H2 ............................................................................37
7
LIST OF FIGURES
Figure page 3-1. Positively Framed PSA...........................................................................................................42
3-2. Negatively Framed PSA .........................................................................................................43
8
LIST OF ABBREVIATIONS
CDC Centers for Disease Control
FCC Federal Communications Commission
PSA Public Service Announcement
USDHHS U.S. Department of Health and Human Service
9
Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Advertising
INFLUENCE OF MESSAGE FRAMING, INVOLVEMENT, AND NICOTINE
DEPENDENCE ON EFFECTIVENESS OF ANTI-SMOKING PUBLIC SERVICE ANNOUNCEMENTS
By
Wan Seop Jung
August 2008
Chair: Jorge Villegas Major: Advertising
Studies examining message framing effects have yielded conflicting findings. Some
studies find that positively framed messages are more persuasive than negatively framed
messages. Other studies report reverse outcomes. This study tried to find the reasons for the
conflicting results in message framing research, and found that the effects of message framing
was varied depending on the involvement level and smoking addiction level. The study suggests
that subjects who are more involved with smoking issue prefer negative messages, while
subjects who are less involved with smoking issue prefer positive messages. The second finding
is that the more dependent on cigarette smoking subjects are, the more favorable attitude toward
negatively framed PSAs they tend to have. The contrary is true when anti-smoking PSAs are
framed positively. Other findings and implications are discussed.
10
CHAPTER 1 INTRODUCTION
Cigarette smoking has great societal and clinical importance (American Cancer Society,
2001). According to the American Cancer Society (2008), smoking was responsible for
approximately 559,312 U.S. premature deaths in 2005. Moreover, a number of non-smokers
suffered from second-hand smoking. A sizable percentage of youth (20%-30%) begin smoking
as early as middle school (Centers for Disease Control [CDC], 1998). A large number of
smokers initiate cigarette smoking during youth-88% by age 18 (U.S. Department of Health and
Human Service [USDHHS], 1994), with approximately fifty percent becoming addicted (CDC,
1991). Smoking behavior in middle school has a tendency to last for a long time due to the
smoking addiction (Oei, 1987). Anti-smoking PSAs are useful due to its preventive effects
(USDHHS, 1994). In addition to the health-related damages, other economical and
environmental damages are caused by smoking. The following are examples of the damages:
productivity decrease due to cigarette breaks, dirtiness from cigarette butts, and budgetary
allocations for anti-smoking. Therefore, much research of anti-smoking Public Service
Announcements (PSAs) tried to abate the prevalence of smoking.
Schneider et al. (1981) estimated the effects of the Fairness Doctrine, which was a
Federal Communications Commission (FCC) regulation regarding broadcasting license
acquisition. Tobacco companies were subjected to restriction on inserting their advertisements in
radio or TV because of the Fairness Doctrine. After the Fairness Doctrine, cigarette consumption
decreased from 1967 to 1970. Researchers have experienced that the restriction of tobacco
advertisements had an influence on decrease of cigarette consumption. However, in 1984, the
Supreme Court recommended the FCC to abolish the Fairness Doctrine because it harmed the
public interest and violated the First Amendment. Then, there had been an emerging interest in
11
attempting to find other devices to abate the prevalence of smoking instead of restriction of
tobacco advertising. Therefore, researchers began to study anti-smoking PSAs.
A substantial number of studies have supported the efficacy of anti-smoking PSAs
empirically and practically. Goldman and Glantz (1998) found that anti-smoking PSAs had an
influence on decline in cigarette consumption in California, Massachusetts, and Arizona.
Popham and Potter (1993) did a research to test whether state- or government-sponsored PSAs
were effective by using the California Department of Health Service sponsored anti-smoking
PSAs, conducted from 1990 to 1991. They concluded that the government-sponsored PSAs
helped smokers to quit significantly.
In recent years, researchers tried to test anti-smoking PSAs using different target groups
(e.g., youth, adolescent, and adult) instead of general population. For instance, Seigel and Biener
(2000) reported that anti-smoking PSAs played a role in decrease of cigarette consumption
among youth in Massachusetts. Other studies found that anti-smoking PSAs were persuasive to
reduce tobacco consumption both for adults and for youth in California (Hu, Sung, & Keeler,
1995; Pierce et al., 1998). Bauer et al. (2000) suggested that anti-smoking PSAs have reduced
and prevented youth cigarette smoking. A result showed that anti-smoking PSAs could have an
effect both on adolescents and on young adults in Florida (Sly, Trapido, & Ray, 2002).
There has been an attempt to understand the persuasion processes that occur when
smokers are exposed to anti-smoking PSAs. For example, a Florida baseline research revealed
that anti-smoking PSAs were effective to change attitude and beliefs toward cigarette smoking
and smoking behavior among youth, and helped reduce the prevalence of smoking among youth
(Sly, Heald, & Ray, 2001). The study was the evidence that anti-smoking PSAs had an influence
on not only changing attitude and belief toward smoking behavior but smoking cessation.
12
A large amount of studies have supported the positive effect of anti-smoking PSAs in
terms of decrease of cigarette consumption or smoking cessation. Many researchers, however,
suggested different results. For instance, O’Keefe (1971) reported that PSAs produced an effect
on smoking behavior in a limited way. He found that the effectiveness of PSAs depended on
recipients’ inclination to give up smoking. In other words, anti-smoking PSAs were effective
only for the smokers who were determined to give up smoking. Wallack and Corbett (1987)
tested how a variety of PSAs such as anti-drug, anti-drinking, and anti-smoking diminished the
rate of smoking, drinking, and drug-taking. The study showed that the contribution of PSAs was
minor in the context of lowering of the rates of smoking, drinking, and drug-taking because
addictive behaviors were unwilling to be changed. Wallack (1981) reported the minor effects of
PSAs on smoking cessation as well.
Traditional anti-smoking PSA research rather related to the efficacy of anti-smoking
PSA in terms of whether persuasion occurs. Recently, researchers have expanded research topics
of anti-smoking PSAs to understand how and why the effect of PSAs varied; further, they tried
to optimize the effectiveness of PSAs. For example, Goldman and Glantz (1998) have examined
a variety of message themes of anti-smoking PSAs such as industry manipulation, addiction,
health issue and second-hand smoke to find the relative effectiveness of each message theme,
and revealed that the industry deception theme was the most effective. A number of studies have
tested the relative effectiveness of ad appeal types among fear, humor, dirtiness (e.g., bad breath
and odor), sports, and sociability because those were frequently used ad appeals in anti-smoking
PSAs (Blum, 1980; Millstein. 1986; Hale & Dillard, 1995; Witte, 1995). The studies found that
fear and humor were the most effective appeals in their appeal type research, and other appeal
types failed to reduce the prevalence of smoking.
13
In recent years there have been numerous attempts to apply message framing effects
(effects of negatively framing message versus effects of positively framed message) to anti-
smoking PSAs. This is because a substantial amount of health-related studies have tested the
effectiveness of message framing, and yielded inconsistent results. Some studies reported
positively framed messages to be more effective than negatively framed messages. Reverse
results have been obtained in other studies. Anti-smoking researchers also tried to test the
relative effectiveness between differently framed messages how the effects of message framing
differ from other health-related research. Anti-smoking studies that have studied message
framing effects have produced conflicting finding as well. The purpose of this study is to find the
reason of the conflicting results in message framing research, and to explore how message
recipients react to two types of anti-smoking messages. Specifically, it will be tested how the
involvement and nicotine dependence level moderate the relative effectiveness of two different
messages. The next chapters will cover previous literatures about message framing, involvement,
and nicotine dependence and methodology used to test hypotheses. The findings of this study
will be helpful to create effective PSAs in terms of lowering smoking rate or smoking cessation.
14
CHAPTER 2 LITERATURE REVIEW
In this study, it will be investigated how to create more effective anti-smoking PSAs by
identifying the role of involvement and nicotine dependence in message framing effects.
Message Framing
Message framing effects can be conceptualized as the perspective that individuals
process and respond differently to dissimilar but objectively equivalent messages (Kühberger,
1998). The analogy of gambling being labeled as “80% of probability of losing money” or “20%
of probability of earning money” is an example of message framing. As stated by Kühberger
(1998), same message tone, manner, and value should be applied to different messages. That is,
different messages should contain a factually equal value. Message framing can be
operationalized either by stressing on benefits gained, namely positively framed message, or by
stressing on benefit lost, namely negatively framed message. Inspired by the message framing
effects, a substantial stream of literature has investigated the effect of message framing on
individual’ decisions (Puto, 1987). Puto’s study revealed that people tend to be risk averse when
they process a positively framed message, but they tend to be risk seeking when they receive a
negatively framed message.
There has been little research addressing how and why people response differently to
different messages (Levin, Schneider & Gaeth, 1998; Meyerowitz & Chaiken, 1987;
Maheswaran & Meyers-Levy, 1990; Stuart & Blanton, 2002; Umphrey, 2003). Moreover, those
studies failed to yield consistent results. For instances, in Levin and Gaeth’s research (1988;
Levin, 1987), people have more favorable attitude toward positively framed messages than
negatively framed ones when they are exposed to messages, stressing either percentage of lean in
total meat or percentage of fat (e.g. containing 75% of lean vs. containing 25% of fat). However,
15
Meyerowitz and Chaiken (1987) found opposing results about message framing effect. In the
Meyerowitz and Chaiken’s study (1987), a negatively framed breast self-examination (BSE)
message was more effective than positively framed one when women were exposed to both
negatively and positively framed BSE messages. They have observed that the negatively framed
message attracts more attention than the positively framed one. Since people are more responsive
and emotional towards negative messages, the researchers have supported the effect of negative
framing (Pratto and John, 1991).
Recent message framing studies tried to find reasons for the conflicting findings. The
studies found that the effectiveness of message framing would differ depending on various
factors. For instance, Ganzach and Karsahi (1995) reported that a study setting can moderate the
effectiveness of the message framing on buying behavior in terms of credit card uses.
Specifically, when the experiment was conducted in the laboratory, a positive message was more
persuasive in terms of increase of card use, whereas a negative one was more effective when the
experiment was conducted in the natural market environment. They found that individuals’
perceived importance level varied in situations, i.e., laboratory setting versus natural market
environment setting. That is because message recipients regarded the issue as more important so
they are motivated to process the ad information in the natural market environment than
laboratory setting.
Smith (1996) revealed that a consumer education level has a moderating effect on the
message framing effects. The study assumed that the education level is highly related to
information processing ability which determines the dominant information processing route
(central vs. peripheral). The central route to persuasion refers to “the persuasion processes
involved with elaboration is relatively high,” whereas the peripheral route to persuasion refers to
16
“the persuasion processes involved when elaboration is relatively low” (O’Keefe, 2002, p. 139).
Elaboration is meant “engaging in issue-relevant thinking” (O’Keefe, 2002, p. 138). Since the
education level plays a key part in cognitive ability to process information, he concluded that
positively framed messages are more effective for the more-educated persons than for the less-
educated. The contrary is true when messages are framed negatively.
Zhang and Buda (1999) reported that there is the moderating effect of one’s need for
cognition (NFC) on the processing of framed advertising messages. Since NFC is an indicator
for the motivation of information processing, it plays an important moderating role in message
framing effects. When individuals’ NFC level is high, they devote more attention to advertising
messages. On the other hand, when individuals’ NFC level is low, they tend to avoid paying
attention to advertising messages. As a result, positively framed messages were more effective
when one’s NFC was high. The contrary was true when one’s NFC was low.
A number of studies suggested that people’s involvement level can influence how they
react to and process information (Greenwald & Leavitt, 1984; Kardes, 1988). This study begins
by exploring role of involvement and nicotine dependence, and relevant theory.
Involvement
Since Krugman (1965) presented the construct of involvement as personal relevance,
this topic has played a vital role in the understanding of persuasion. There is general agreement
that involvement plays a key role in explaining consumer behavior and information processing
procedure. However, the effect of involvement is debatable due to different ideas for
conceptualization and operationalization. The following two issues are barriers for obtaining
general agreement about involvement: type of involvement and intensity of involvement. Those
two are in the midst of involvement issue. The type of involvement is the first issue for the
disagreement. Some researchers proposed two types of involvement to explain the construct
17
better, namely, enduring involvement and situational involvement (Houston & Rothschild, 1978;
Rothschild, 1979). The categorization of involvement helped researchers begin to propose the
reasons for the disagreement. The second issue for the disagreement is the intensity of
involvement, namely, high involvement and low involvement (Gainer, 1993). Involvement is
also debatable due to the measuring the involvement construct (Celsi and Olson, 1988). While
researchers have defined the construct of involvement differently (Gainer, 1993), the general
agreement of involvement is the person’s perceived relevance of a product or situation on the
basis of personal goals, values, and interests (Zaichkowsky, 1985).
Enduring involvement can be defined as “enduring personally relevant knowledge
structures” or ongoing concern (Martin & Marchall, 1999, p. 208). The structures are stored in
the relatively long-term memory and obtained from prior experiences. Situational involvement
can be defined as “situation-specific, involving the cues and stimuli that consumers perceive as
personally relevant” (Martin & Marchall, 1999, p. 208). Situational involvement occurs only in
specific situations. The analogy of situations being labeled as “being interested in cars” or
“purchasing a car” is an example of two involvement types. The level of enduring involvement is
concluded by how well the issue is pertained to personally relevant value, whereas the level of
situational involvement is concluded by how well the situation arouses personal goals and values.
The level of situational involvement is dramatically dropped when the personal goals and values
are fulfilled. Since situational involvement rather relates to characteristics of the decision
situation, it is inapposite to the purpose of this study which tries to explain how enduring or long-
term involvement moderates the effectiveness of the message framing. Therefore, this study
employed enduring involvement as a moderator of message framing effects.
18
There has been a substantial amount of research about message framing and involvement,
respectively. However, the research about how involvement interacts with message framing has
considerably been neglected. Since both involvement and framing have been considered as
important factors to predict advertising effectiveness, this disregard is amazing. In this study, it
will be investigated how message framing effects will be moderated by intensity of involvement.
Connection of Involvement with Message Framing
Petty and Cacioppo’s elaboration likelihood model (ELM) (1986) is helpful to
understand how involvement affects message framing effects. The model is composed of a
central route and a peripheral route. When individuals are more involved with an issue or a
product, they use the central route to process messages, whereas they use the peripheral route to
process messages when they are less involved with an issue or a product. The model also
suggests that those who are more involved with an issue or product are more motivated to
process relevant messages thoroughly, whereas those who are less involved with an issue or a
product are less motivated to process messages. Motivation in the model is defined as inner
arousal for goal achievement. The level of message processing motivation has an effect on how
much individuals pay more attention to and devote to process information (Celsi & Olson, 1988).
According to the model, the level of message processing motivation determines whether
message claims predominate in persuasion because individuals employ either a central route or a
peripheral route in persuasion process. That is, the motivation level determines the use of one of
processing routes. Under conditions of low message processing motivation, individuals are more
likely to form an attitude by relying on peripheral cues, such as message tone, wording, and
message attractiveness instead of scrutinizing message claims thoroughly and integrating
relevant thoughts and beliefs to form an overall attitude. Hence, the positively framed PSA “You
are saving $1,000 every year by being a non-smoker” may be interpreted as “Non-smokers cash
19
in without any efforts”. Similarly, the negatively framed PSA “You are burning $1,000 every
year by smoking cigarettes” may be interpreted as “Smoking costs a lot.” If so, positively framed
PSAs should be more effective under low involvement conditions.
Under conditions of high message processing motivation, individuals are more inclined
to process all information in the ad to arrive at a decision. Thus, they process messages in detail
with prior knowledge to judge the validity of messages (Maheswaran & Meyers-Levy, 1990),
and integrate relevant thoughts and beliefs to form an overall attitude (Petty, Cacioppo, &
Schumann, 1983). Wright (1981) suggested that negative messages received more considerable
weight and influence than positive messages when individuals integrate message-relevant
information. That is because “overweighting may only occur when an audience member is
sufficiently concerned over the message content to bother generating reactions and integrating
those into an overall impression, and to worry about making errors in this” (Wright, 1981, p.
279-280). A variety of studies have supported the overweighting to high involvement conditions
(Kanouse, 1984; Weinberger, Allen, & Dillon, 1981). The implication of those studies is that
messages should be processed thoroughly and negative messages should be more persuasive
under high involvement conditions. Based on the previous theoretical background, following
hypotheses can be predicted:
H1: The relationship between message framing effects and attitude toward anti-smoking PSAs is moderated by involvement.
H1.1: For those who are highly involved in smoking, an anti-smoking PSA is more
persuasive when it is framed negatively. H1.2: For those who are less involved in smoking, an anti-smoking PSA is more persuasive when it is framed positively.
20
Nicotine Dependence
In the first hypothesis, it is suggested that involvement moderates the effectiveness of
message framing by using general subjects including both smokers and non-smokers. This is
because the message framing effects depend on the motivation of message recipients to process
information. The second hypothesis is inspired by the Petty and Cacioppo’s model (1983) as well.
In their model, motivation influences the way to process information by using either peripheral
route or central one. In the second hypothesis, it is suggested that the nicotine dependence level
moderates the effectiveness of message framing by using smoker samples instead of general
subjects. A number of anti-smoking researchers revealed the relationship between nicotine
dependence and motivation. Reflecting the relationship among nicotine dependence, motivation
and the way of information processes, the second hypothesis examines how the cigarette
addiction level affect attitudes toward anti-smoking PSAs differently depending on the message
framing.
Dependence is conceptualized as compulsive physiological need for a habit-forming
substance. The Diagnostic and Statistical Manual of Mental disorders (APA, 1994) provides
several criteria for dependence which are (1) tolerance; (2) withdrawal; (3) using larger amounts
or longer than intended; (4) unsuccessful efforts to cut down; (5) much time spent to obtain
substance; (6) negative social and/or occupational consequences; (7) persistent physical and/or
psychological problems. The manual recommends that an individual must meet at least three
criteria to be diagnosed with dependence. Nicotine dependence refers to the physical
vulnerability of our body to the chemical nicotine, which is possibly addicting when carried by
various cigarette products (MayoClinic.com, 2006).
21
Connection of Nicotine Dependence Level with Message Framing
Bauer (1960) presented the construct of perceived risk. And it was introduced to the
marketing literature. Perceived risk is conceptualized as the perspective that “the amount that
would be lost if the consequences of an act were not favorable, and the individual’s subjective
feeling of certainty that the consequences will be unfavorable” (Cunningham, 1967, p.37).
Similarly, Stone and Winter (1987) considered risk as an expectation of loss. In connection with
this study, perceived risk regarding smoking can be re-conceptualized as the perspective how
well individuals understand the risk of smoking.
The second hypothesis focuses on the nicotine dependence level as a moderator of
message framing effects by using a sample of smokers. In Wong and McMurray (2002) study,
the nicotine dependence level was related to one’s perceived health risk level. In addition,
Rindfleisch and Crockett (1999) tested the relationship between nicotine dependence and
perceived risk by using a college student sample. Specifically, the less dependent on cigarette
smoking subjects are, the lower level of perceived risk they tend to have. Conversely the more
dependent on cigarette smoking subjects are, the higher level of perceived risk they tend to have.
A number studies proposed that a perceived health risk level be strongly related to one’s
motivation to process health-related information (Block & Keller, 1995; Rothman & Salovey,
1997; Umphrey, 2003). In addition, Bock et al. (2000) revealed the relationship between risk
perception and motivation. They found that subjects who had higher risk perceptions were more
motivated to process smoking cessation information and consult doctors to discuss smoking
issues than those who had lower risk perceptions. The sum of the relationship among nicotine
dependence, perceived risk and motivation is that highly addicted subjects are more motivated to
process information than less addicted subjects, while less addicted subjects are less motivated to
process information than highly addicted subjects.
22
As for the relationship between motivation and message framing, the causality that
individuals’ motivation to process information affects the effectiveness of message framing was
supported (Kanouse, 1984; Maheswaran & Meyers-Levy, 1990; Weinberger, Allen, & Dillon,
1981). The more motivated individuals are, the more favorable attitude toward negatively framed
PSAs formed, while the less motivated individuals are, the more favorable attitude toward
positively framed PSAs. Therefore, nicotine dependence has an impact on perceived risk,
perceived risk relates to motivation. In addition, the motivation would moderate message
framing effects. Finally, such relationships affect attitude toward anti-smoking PSAs.
H2: The relationship between message framing effects and attitude toward anti-smoking PSAs is moderated by nicotine dependence.
H2.1: For smokers who are highly dependent on smoking, an anti-smoking PSA is more
persuasive when it is framed negatively. H2.2: For smokers who are less dependent on smoking, an anti-smoking PSA is more
persuasive when it is framed positively.
23
CHAPTER 3 METHODOLOGY
The purpose of this study is to investigate how involvement and nicotine dependence
moderate the message framing effects. The experiment was conducted to test two hypotheses
from March 3rd to March 11th. at the same period. In the H1, the study was conducted to test the
moderating effects of involvement on message framing with general samples, including both
smokers and non-smokers. In the H2, the study was conducted to test the moderating effects of
nicotine dependence on message framing with only smoker samples among collected data.
Study design. To test H1, two independent variables were involvement and message
framing with attitudes toward anti-smoking PSAs as a dependent variable. The involvement level
as one of independent variables was continuous (from 0: low to 7: high), and message framing
was categorical variable manipulated (positive versus negative). The dependent variable,
attitudes toward anti-smoking PSAs, was measured from 1 (unfavorable) to 7 (favorable). To test
H2, two independent variables were nicotine dependence and message framing with attitudes
toward anti-smoking PSAs as a dependent variable. The nicotine dependence level as one of
independent variables was continuous (from 0: low to 10: high), and message framing was
categorical variable manipulated (positive versus negative). The dependent variable, attitudes
toward anti-smoking PSAs, was measured from 1 (unfavorable) to 7 (favorable).
Participants. In the H1 testing, the total number of participants was one hundred and
eighty eight. They were recruited from undergraduate-level classes and a street on the campus in
the area of a southeastern American university. Due to relatively small number of participants
who currently smoke, more subjects were additionally recruited from a street on the campus. The
courses where students were recruited did not include smoking topics that might have sensitized
24
subjects to the purpose of the study. Female participants made up 51.1 % of the sample, and
18.1% were smokers. The dominant percentage of ethnicity was Caucasians (54.9%), and
followed by Hispanics (20.7%), Asian-Americans (19.0%), and African-Americans (3.8%). In
the H2 testing, a sample of smokers among collect data was used. Out of original participant,
forty six were smokers. Female participants made up 42.2% of the sample, and all experiment
participants were smokers. The dominant percentage of ethnicity was Caucasians (55.6%), and
followed by Hispanics (15.6%), Asian-Americans (26.7%), and African-Americans (2.2%).
Stimuli. Two versions of a print anti-smoking public service announcements were
created to test both hypotheses. As shown Figure 1 and 2, visuals and layout were equal across
experimental groups. Two advertisements contained either a positively framed message or a
negatively framed message. Both advertisements presented the argument that stopping or
continuing with cigarette consumption can lead to a financial gain or loss equivalent to two
iPhones which is one of the most appealing and familiar products among college students. Thus,
experiment participants can recognize the value of financial gain or loss easily that is caused by
stopping or continuing with cigarette consumption. The manipulation of message framing
occurred in the body copy. For the negatively framed message, the following message was used:
“You are burning $1,000 every year by smoking cigarettes. You can buy 2 iPhones with $1,000.”
For the positively framed message, the following message was used: “You are saving $1,000
every year by being a non-smoker. You can buy 2 iPhones with $1,000.”
Procedure. Sessions were conducted in small classroom settings and on a street on the
campus. Participants were instructed that the purpose of the study was to understand consumers’
ideas about smoking message. After the informed consent was secured, participants were
randomly assigned into one of the two experimental conditions. In the H1 testing, ninety one
25
participants (48.4%) were assigned into Group 1 and shown the negatively framed PSA. Ninety
seven participants (51.6%) were assigned into Group 2 and shown the positively framed PSA. In
the H2 testing, twenty four participants (52.2%) were assigned into Group 1 and shown the
negatively framed PSA. Twenty two participants (47.7%) were assigned into Group 2 and shown
the positively framed PSA. After reading the anti-smoking PSA, they completed the scales
designed to check manipulation, involvement, and measure of dependent variable. For the
duration of each session, the experimenter ensured that participants completed the instrument
independently and did not turn back to the pages they had already read. Upon completion of the
instrument, subjects were thanked and debriefed. Each session took about ten minutes.
Independent Variables
Message framing (positive vs. negative). Message framing was employed as an
independent variable to test both hypotheses. The positively-framed PSA illustrated the positive
effect of quitting smoking by stating “You are saving $1,000 every year by being a non-smoker.
You can buy 2 iPhones with $1,000,” while the negatively framed PSA emphasized the adverse
effects of smoking by stating “You are burning $1,000 every year by smoking cigarettes. You can
buy 2 iPhones with $1,000.” The participants reported on a two-item, seven-point (1 = strongly
disagree, 7 = strongly agree) Likert scale how much they agreed with the following two
statements: “The ad is focusing on the positive effect of smoking cessation,” and “The ad is
focusing on the negative effect of smoking.”
Involvement. Involvement was used as a measured variable in the H1. Participants were
asked to answer smoking involvement items. Five seven-point semantic differential items were
used to measure involvement level (1 = important, irrelevant, valuable, useful, and interesting, 7
26
= unimportant, relevant, invaluable, useless, and not interesting). The scales were adapted from
Zaichkowsky (1985). The scale was internally consistent (α = .89).
Nicotine dependence. The nicotine dependence level was measured using the
Fagerström Nicotine Dependence Scale (Heatherton, 1991) in the H2. Smokers were asked to
answer the following questions:
1. How soon after you wake up do you smoke your first cigarette? a. Within 5 minutes (3 points) b. Within 6-30 minutes (2 points) c. Within 31-60 minutes (1 point) d. After 60 minutes (0 points)
2. Do you find it difficult to refrain from smoking in places where it is forbidden (e.g., in church, at the library, in cinema, etc)? a. Yes (1 point) b. No (0 points)
3. Which cigarette would you hate most to give up? a. The first one in the morning (1 point) b. Any other (0 points)
4. How many cigarettes per day do you smoke? a. 10 or less (0 points) b. 10-20 (1 point) c. 21-30 (2 points) d. 31 or more (3 points)
5. Do you smoke more during the first hours after waking than during the rest of the day? a. Yes (1 point) b. No (0 points)
6. Do you smoke even when you are ill enough to be in bed most of the day?” a. Yes (1 point) b. No (0 points)
All points were added up on the basis of subjects’ responses and the assigned points. For
example, if someone smokes his or her first cigarette after waking up within 5 minutes, finds it
difficult to refrain from smoking-free places, hates to give up the first morning cigarette, smokes
27
31 or more cigarette per day, smokes more cigarette during the first hour after waking, and
smokes even when he or she is ill enough to be in bed most of the day, the nicotine dependence
level indicates 10. The nicotine dependence level was measured being continuous. The range of
the nicotine dependence level is from 0 (low) to 10 (high). Heatherton et al. (1991) suggested
that a score of 5 or more indicates a significant smoking dependence, while a score of 4 or less
shows a low to moderate dependence.
Dependent variable.
Attitude toward Ad. One dependent variable was measured to test both hypotheses:
Attitudes toward Ad. The dependent variable was measured using a four-item, seven-point
semantic differential scale (1 = unpleasant, unlikable, not irritating, and not interesting, 7 =
pleasant, likable, irritating, and interesting) (Olney, Holbrook, and Batra, 1991). The third item
for measuring involvement level (i.e., not irritating vs. irritating) was reverse coded. The scale
was internally consistent (α = .85), and the responses were averaged to represent the construct for
the H1 (M = 4.32, SD = 1.34). For the H2, the scale was internally consistent (α = .93), and the
responses were averaged to represent the construct (M = 4.26, SD = 1.63).
28
CHAPTER 4 RESULTS
Hypothesis 1
Multiple regressions were conducted to test H1, with two independent variables,
involvement and message framing, and one dependent variable, attitude toward ad. A sample of
smokers and non-smokers was used to test how involvement moderates message framing effects.
Manipulation Check
Before testing H1, a manipulation check was conducted. As expected, the message that
“You are saving $1,000 every year by being a non-smoker, You can buy 2 iPhones with $1,000”
was perceived as the positive anti-smoking message, whereas the message that “You are burning
$1,000 every year by smoking cigarettes. You can buy 2 iPhones with $1,000” was perceived as
the negative message. Participants reported on a two-item, seven point (1 = strongly agree, 7 =
strongly disagree) Likert scale how much they agreed with the following two statements: “The
ad is focusing on the positive effect of smoking cessation,” and “The ad is focusing on the
negative effect of smoking.” As shown Table 1, difference between observed value for positively
framed message and test value (= 4) was significant on item one, t = -4.96, p < .001,
MPositiveMessage = 3.06, SD = 1.86. As shown Table 2, difference between observed value for
negatively framed message and test value (=4) was significant on item two, t = -6.07, p < .001,
MNegativeMessage = 2.88, SD = 1.76. Subjects felt that the anti-smoking PSA was more negative
when it was negatively framed rather than positively framed, and they felt that the anti-smoking
PSA was more positive when it was positively framed rather than negatively framed. Therefore,
these manipulation check measures suggest that the intended factors were manipulated
successfully.
29
Involvement
In the H1 testing, the study was conducted to investigate how involvement moderates the
message framing effects. And the message framing was manipulated. Involvement was measured
as a continuous variable on the basis upon Zaichkowsky (1985). The scale measuring
involvement level was internally consistent (α = .89).
Test of Hypothesis 1
Influence of Involvement and Message Framing
This study investigated how involvement moderates the effectiveness of message
framing. Specifically, Subjects who are more involved with smoking issue would have more
favorable attitude toward negatively framed PSAs, while subjects who are less involved with
smoking issue would have more favorable attitude toward positively framed PSA.
A multiple regression was conducted to test H1, with attitude toward the ad as a
dependent variable, with the involvement level and message framing as independent variables.
The descriptive statistics are presented in Table 3. The mean of involvement was 2.67 (SD =
1.62), .52 (SD = .50) for message framing, and 4.31 (SD = 1.33) for attitude toward ad,
respectively.
The model with two independent variables, the involvement level and the message
framing, and the joint effect of these two variables was statistically significant, F (3, 184) = 7.75,
p < .001. Message framing was employed as a dummy variable where a negatively framed PSA
was 0, and a positively framed PSA was 1. The results are presented in Table 4 and 5. Initially, a
multiple regression was conducted with only two independent variables, involvement and
message framing. Its F-value was 1.86. Adding the interaction, then, contributed to 9.3% of the
model with only two independent variables. F-value change was 19.18. The effect of the
interaction between the addiction level and message framing was b = -.50, t = -4.38, p < .001.
30
The interaction term was again explored by re-computing the regression analyses
separately for groups of subjects exposed to negative message or positive message in terms of
message framing. Results from follow-up analyses revealed that involvement was positively
associated with attitude toward anti-smoking PSA among subjects who were exposed to the
negative message (F (1, 89) = 2.34, β = .16, t = 1.53, p > .05), whereas involvement was
negatively associated with attitude toward anti-smoking PSA among subjects who were exposed
to positive message (F (1, 95) = 24.40, β = -.45, t = -4.94, p < .001). Results of the interaction
tests are presented in Table 6.
Thus, the more involved in smoking, the more favorable attitude toward negatively framed
PSAs individuals tend to have, while the less involved in smoking, the more favorable attitude
toward positively framed PSAs individuals tend to have. These results support overall
assumption of H1. However, when regression analyses were conducted separately by selecting
cases (e.g., subjects who were exposed to the negatively framed PSA vs. subjects who were
exposed to the positively framed PSA), H1 was partially supported. That is, H1.1 was supported
whereas H1.2 was not supported. The main effect of the involvement level was b = .13, t = 1.65,
p > .05. The main effect of message framing was b = 1.32, t = 3.71, p < .001. Since this result is
irrelevant to this study, no interpretations are provided for main effects.
Hypothesis 2
Multiple regressions were conducted to test H1, with two independent variables, nicotine
dependence and message framing, and one dependent variable, attitude toward ad. A sample of
smokers among collected data was used to test how nicotine dependence moderates message
framing effects.
31
Manipulation Check
Before testing H2, manipulation check was conducted. As expected, the message that “You
are saving $1,000 every year by being a non-smoker, You can buy 2 iPhones with $1,000” was
perceived as the positive anti-smoking message, whereas the message that “You are burning
$1,000 every year by smoking cigarettes. You can buy 2 iPhones with $1,000” was perceived as
the negative message. Participants reported on a two-item, seven point (1 = strongly agree, 7 =
strongly disagree) Likert scale how much they agreed with the following two statements: “The
ad is focusing on the positive effect of smoking cessation,” and “The ad is focusing on the
negative effect of smoking.” As shown Table 7, difference between observed value for positively
framed message and test value (= 4) was significant on item one, t = -5.89, p < .001,
MPositiveMessage = 2.05, SD = 1.56. As shown Table 8, difference between observed value for
negatively framed message and test value (=4) was significant on item two, t = -3.09, p < .01,
MNegativeMessage = 2.92, SD = 1.72. Subjects felt that the anti-smoking PSA was more negative
when it was negatively framed rather than positively framed, and they felt that the anti-smoking
PSA was more positive when it was positively framed rather than negatively framed. Therefore,
these manipulation check measures suggest that the intended factors were manipulated
successfully.
Nicotine Dependence
In the H2 testing, the study was conducted to investigate how nicotine dependence
moderates the message framing effects. And the message framing was manipulated. Nicotine
dependence was measured as a continuous variable on the basis upon Heatherton et al. (1991).
32
Test of Hypothesis 2
Influence of Nicotine dependence and Message Framing
This study investigated how addiction level affects attitude toward anti-smoking PSA
when it is framed either positively or negatively. Specifically, Subjects who are more addicted to
smoking would have more favorable attitude toward negatively framed PSAs, while subjects
who are less addicted to smoking would have more favorable attitude toward positively framed
PSA.
A multiple regression was conducted to test H2, with attitude toward the ad as the
dependent variable, and with the addiction level and message framing as independent variables.
The descriptive statistics were presented in Table 9. The mean of smoking addiction was 3.11
(SD = 2.27), .48 (SD = .51) for message framing, and 4.26 (SD = 1.63) for attitude toward ad,
respectively.
The model with two independent variables, the addiction level and the message framing,
and the joint effect of these two variables was statistically significant, F (3, 42) = 6.72, p < .001.
Message framing was employed as a dummy variable where a negatively framed PSA was 0, and
a positively framed PSA was 1. The results are presented in Table 10 and 11. Initially, a multiple
regression was conducted with only two independent variables, nicotine dependence and
message framing. Its F-value was 7.46. Adding the interaction, then, contributed to 6.7% of the
model with only two independent variables. F-value change was 4.15. The effect of the
interaction between the nicotine dependence level and message framing was b = -.38, t = -2.04, p
< .05.
Interaction was again explored by re-computing the regression analyses separately for
groups of subjects exposed to negative message or positive message in terms of message framing.
Results from follow-up analyses revealed that nicotine dependence was positively associated
33
with attitude toward anti-smoking PSA among subjects who were exposed to the negative
message (F (1, 22) = .02, β = .03, t = .13, p > .05), whereas nicotine dependence was negatively
associated with attitude toward anti-smoking PSA among subjects who were exposed to positive
message (F (1, 20) =9.63, β = -.57, t = -3.10, p < .01). The results of interaction tests are
presented in Table 12.
Thus, the more dependent on cigarette smoking, the more favorable attitude toward
negatively framed PSAs individuals tend to have, while the less dependent on cigarette smoking,
the more favorable attitude toward positively framed PSAs individuals tend to have. These
results support overall assumption of H1. However, when regression analyses were conducted
separately by selecting cases (e.g., subjects who were exposed to the negatively framed PSA vs.
subjects who were exposed to the positively framed PSA), H1 was partially supported. That is,
H1.1 was supported whereas H1.2 was not supported. The main effect of the addiction level was
b = .02, t = .14, p > .05. The main effect of message framing was b = -.16, t = -.23, p > .05. Thus,
both main effects were not statistically significant.
34
Table 4-1. Manipulation Check of Positively Framed Message for Item 1 in H1 Testing N Mean SD t df Sig. Mean Difference Positively Framed Message (Item 1) 97 3.06 1.86 -4.96 96 .000 -.94
Test value = 4 Table 4-2. Manipulation Check of Negatively Framed Message for Item 2 in H1 Testing N Mean SD t df Sig. Mean Difference Negatively Framed Message (Item 2) 91 2.88 1.76 -6.07 90 .000 -1.12
Test value = 4 Table 4-3. Descriptive Statistics of H1 M SD
Attitude toward Ad 4.31 1.33Involvement 2.67 1.62Message Framing .52 .50 Table 4-4. Model Statistics of H1 Model Source SS df MS F Sig. R R Square R Adj
Involvement Regression 6.55 2 3.27Framing Residual 325.93 185 1.76Model 1 Total 332.48 187
1.86 .159 .14 .02 .01
Involvement Regression 37.32 3 12.44Framing Residual 295.17 184 1.60Model 2 Interaction Total 332.48 187
7.75 .000 .34 .11 .10
Table 4-5. Regression of H1 Source B SE β t Sig.
(Constant) 3.96 .25 15.68 .000Involvement* .13 .08 .16 1.65 .100Framing** 1.32 .36 .50 3.71 .000Interaction -.50 .11 -.66 -4.38 .000*Involvement: 1 (low) to 7 (high) **Message Framing: 0 (negatively framed message), 1 (positively framed message)
35
Table 4-6. Follow-up Analyses for Interaction in H1 Message Framing IV B SE β t Sig. Negative Involvement .133 .087 .160 1.529 .130 R = .160, R2 = .026, R2
Adj = .015
Positive Involvement -.367 .074 -.452 -4.939 .000 R = .452, R2 = .204, R2
Adj = .196 Table 4-7. Manipulation Check of Positively Framed Message for Item 1 in H2 Testing N Mean SD t df Sig. Mean Difference Positively Framed Message (Item 1) 22 2.05 1.56 -5.89 21 .000 -1.96
Test value = 4 Table 4-8. Manipulation Check of Negatively Framed Message for Item 2 in H2 Testing N Mean SD t df Sig. Mean Difference Negatively Framed Message (Item 1) 24 2.92 1.72 -3.09 23 .005 -1.08
Test value = 4 Table 4-9. Descriptive Statistics of H2 M SD
Attitude toward Ad 4.26 1.63 Addiction Level 3.11 2.27 Message Framing 0.48 0.51 Table 4-10. Model Statistics of H2 Model Source SS df MS F Sig. R R Square R Adj
Addiction Regression 30.90 2 15.45 Framing Residual 89.03 43 2.07 Model 1 Total 119.94 45
7.46 0.002 0.51 0.26 0.22
Addiction Regression 38.91 3 12.97 Framing Residual 81.02 42 1.93 Model 2 Interaction Total 119.94 45
6.72 0.001 0.57 0.32 0.28
36
Table 4-11. Regression of H2 Source B SE β t Sig.
(Constant) 4.85 0.52 9.36 0.000 Addiction* 0.02 0.14 0.03 0.14 0.892 Framing** -0.16 0.71 -0.05 -0.23 0.820 Interaction -0.38 0.19 -0.55 -2.04 0.048 *Nicotine Addiction: 1 (low) to 10 (high) **Message Framing: 0 (negatively framed message), 1 (positively framed message)
Table 4-12. Follow-up Analyses for Interaction in H2 Message Framing IV B SE β t Sig. Negative Nicotine Dependence .019 .146 .028 .133 .895 R = .028, R2 = .001, R2
Adj = -.045
Positive Nicotine Dependence -.358 .115 -.570 -3.104 .006 R = .570, R2 = .325, R2
Adj = .291
37
CHAPTER 5 DISCUSSION AND CONCLUSION
Discussion
This study was designed to resolve the conflicting findings of message framing effects
and identify the variables that have an impact on the effectiveness of message framing. Overall,
the results are compatible with this study’s grand assumption that involvement level and cigarette
addiction level moderate message framing effects.
In the H1, the contention of this study is that involvement level moderates the
effectiveness of either positively or negatively framed anti-smoking PSAs. This contention is
based on Petty and Cacioppo’s model and previous research about overweighting of negative and
positive messages (Kanouse, 1984; Petty & Cacioppo, 1986; Weinberger, Allen, & Dillon, 1981;
Wright, 1981). Consistent with the H1, the involvement level indicates favorability of either a
positively framed PSA or a negatively framed one. These results are consistent with previous
research dealing with involvement and message framing (Maheswaran & Meyers-Levy, 1990)
that proposed the moderating role of involvement.
However, after conducting separate analyses by selecting cases whether subjects were
exposed to either negatively or positively framed message, H1 was partially supported.
Specifically, in the H1.1, the following was predicted: Under high involvement conditions, an
anti-smoking PSA is more persuasive when it is framed negatively. Although an intended
direction was found, it was not statistically significant. Therefore, H1.1 was not supported. In the
H1.2, the following was predicted: Under low involvement conditions, an anti-smoking PSA is
more persuasive when it is framed negatively. Intended significant results were found. Thus,
H1.2 was supported.
38
In the H2, the contention of this study is that smoker’s nicotine dependence level
moderates the effectiveness of either positively or negatively framed anti-smoking PSAs. This
contention is based on the relationship among nicotine dependence, perceived risk, motivation,
and message framing. The causalities are that nicotine dependence has an impact on perceived
risk (Rindfleisch & Crockett, 1999; Wong & McMurray, 2002), perceived risk relates to
motivation (Block & Keller, 1995; Rothman & Salovey, 1997; Bock et al., 2000; Umphrey,
2003). In addition, the motivation would moderate message framing effects (Kanouse, 1984;
Weinberger, Allen, and Dillon, 1981). Consistent with the H2, the nicotine dependence level
indicates favorability of either a positively framed PSA or a negatively framed one.
However, after conducting separate analyses by selecting cases whether subjects were
exposed to either negatively or positively framed message, H2 was partially supported.
Specifically, in the H2.1, the following was predicted: Under high nicotine dependence
conditions, an anti-smoking PSA is more persuasive when it is framed negatively. Although an
intended direction was found, it was not statistically significant. Therefore, H2.1 was not
supported. In the H2.2, the following was predicted: Under low nicotine dependence conditions,
an anti-smoking PSA is more persuasive when it is framed negatively. Intended significant
results were found. Thus, H2.2 was supported.
Implications
This study improves our understanding how individuals process and respond differently to
dissimilar but objectively equivalent messages. In the practice, anti-smoking clinic consultants
might be advised to use either negatively or positively framed message depending on clients’
nicotine dependence level. For example, the consultants can measure their clients’ nicotine
dependence level. When they consult with clients with low nicotine dependence, positive
message will be more effective in terms of lowering cigarette consumption or smoking cessation.
39
In other words, negative message will be more effective when they consult with clients with high
nicotine dependence. Anti-smoking PSA executioners might be advised to employ either
negatively or positively framed message depending on message recipients’ involvement level.
Moreover, it is also helpful to choose appropriate place and time for implementing anti-smoking
PSAs. For instance, when PSA executioners place anti-smoking PSAs in magazines, the message
framing will depend on magazine types. More specifically, when a positively framed PSA is
placed in a teen magazine, we can predict better effects because teens’ involvement level might
be low. Therefore, anti-smoking consultants and PSA executioners can create either negative or
positive message by categorizing target groups into either smoker or non-smoker group.
Limitations and Future Research
There are a few limitations in the study. This project is based the use of student
participants as a sample. Although the use of student participants is more acceptable in
experimental studies than in survey, homogeneity of the sample may have produced results that
are not typically found in the population. Findings of this project need to be replicated using
more a heterogeneous, adult sample to achieve external validity.
This study tested anti-smoking PSAs’ impact on attitude toward messages, not intention
to quit smoking or attitude toward smoking behavior. Attitude toward ad has been posited to be a
mediator of advertising effectiveness which includes brand attitudes and future intentions
(Mitchell & Olson, 1981; Shimp, 1981). This study measured one dependent variable: attitude
toward ad. This study found significant interaction effects by measuring attitude toward ad as a
dependent variable in both experiments. However, it might have different results if different
dependent variables such as attitude toward smoking behavior or intention to quit were employed.
This is because the majority of smokers do not have an intention to give up smoking in the near
future (Hennrikus, Jeffery & Lando, 1995; John, Meyer, Rumpf, & Hapke, 2003). Therefore,
40
future research should employ other dependent variables, and test whether attitude toward ad can
predict future intentions in anti-smoking PSA research.
In the message framing studies, different messages should contain a factually equal
value. The two versions of stimuli used in this study should be equal across experimental groups
except message framing (e.g., saving money vs. burning money). Except for the framing,
causalities of messages are different. For example, in the positively framed message, the message
posits that non-smoking can save $1,000 every year, whereas, in the negatively framed message,
the message posits that smoking can burn $1,000 every year. The different causalities in the
messages may have produced unintended results. Future message framing studies should create
same messages across experimental groups.
In addition to the message framing valence, involvement scales employed in this study
may have produced unexpected results. The Zaichkowsky’s scale (1985) measures an issue
involvement. The stimuli in this study focused on cessation of smoking. Even though this study
intended to measure an issue involvement, participants might think of smoking cessation issue
rather than smoking issue. Future studies should use an appropriate scale to measure involvement
level.
41
Figure 3-1. Positively Framed PSA
42
Figure 3-2. Negatively Framed PSA
43
APPENDIX A INFORMED CONSENT FORM
Protocol Title: College Students’ Ideas About Smoking Please read this consent statement carefully before you decide to participate in this study Purpose of the research study: The purpose of this research is to study college students’ ideas about smoking. The participants will be asked questions about their beliefs and personal experiences about smoking. What you will be asked to do in this study: You will be invited to a laboratory. After the investigator’s instruction, print your name and student ID on the attached name sheet, sign the consent form, and complete the questionnaire adequately apart from others. When your instructor requests the survey, detach the name sheet and this consent form from the main survey. Then submit the consent form, name sheet, and the envelope separately. Time required: About 15 minutes Risks and Benefits: We do not anticipate there will be any risks or direct benefits to you as a consequence of your decision to complete the survey. Confidentiality: Your name or student ID will not appear on the questionnaire itself. Your identity will be kept confidential to the extent provided by law. Your instructor will need your student ID to give you extra credit, but will not receive your name before the final exam. Your responses are completely confidential and no reference will be made in any oral or written report that would link you individually to the study. Further, the principal investigator works independently of your instructor. Voluntary participation: Your participation in this study is completely voluntary. Right to withdraw from the study: You have the full right to withdraw from the study at anytime without consequence. Whom to contact if you have questions about the study: Wan Seop Jung, Master’s student, College of Communications, University of Florida, address: 2018 Weimer Hall, University of Florida, Gainesville, FL, 32611-8400. phone: 352.870.2077 email: [email protected]. Jorge Villegas, Assistant Professor, College of Communications, University of Florida, address: 2084 Weimer Hall, University of Florida, Gainesville, FL, 32611-8400. phone: 352.392.5059. email: [email protected] Whom to contact about your right as a research participant in this study: UFIRB, Box 112250, University of Florida, Gainesville, FL 32611-2250; ph 392-0433 Agreement: I have read the procedure described above. I voluntarily agree to participate in the procedure and I have received a copy of this description. Participant: __________________________________ Date: ______________________ Principal Investigator: __________________________ Date: ______________________
44
APPENDIX B QUESTIONNAIRE
Instructions: Please review the following advertisement and then answer the questions that correspond with a particular advertisement. Directions: The following questionnaire asks you to indicate your opinion about the anti-smoking ad. The scales in some of the questions are meant to measure your opinions on each of the ad. There is no right or wrong answers.
45
Please read each set of adjectives carefully, and decide where your opinion would be the most accurately reflected on the continuum. Then, mark on the square that most closely reflects your opinion. You may refuse to answer any questions. Remember, there are no right or wrong answers. 1) Do you agree that the ad is focusing on positive effect of smoking cessation?
1 2 3 4 5 6 7 Strongly agree
Strongly disagree
2) Do you agree that the ad is focusing on negative effect of smoking?
1 2 3 4 5 6 7 Strongly agree
Strongly disagree
Please check the scale that best reflects your agreement with each of the following items. 3) In general, smoking is.
1 2 3 4 5 6 7 Important Unimportant
1 2 3 4 5 6 7 Irrelevant Relevant
1 2 3 4 5 6 7 Valuable Invaluable
1 2 3 4 5 6 7 Useful Useless
1 2 3 4 5 6 7 Interesting
Not interesting
46
4) Do you find the advertisement is (EVALUATION OF THE ADVERTISEMENT): 1 2 3 4 5 6 7 Unpleasant Pleasant
1 2 3 4 5 6 7 Unlikable Likable
1 2 3 4 5 6 7 Not irritating Irritating
1 2 3 4 5 6 7 Not
interesting Interesting
5) Please check your smoking behavior on the square. (If you are not a smoker, go to the next page) 1. How soon after you wake up do you smoke your first cigarette?
0-5 min 6-30 min 31-60 min After 60 min 2. Do you find it difficult to refrain from smoking in places where it is forbidden (e.g., church library, cinema)?
Yes No 3. Which cigarette would you be the most unwilling to give up?
First in the morning Any of the others 4. How many cigarettes per day do you smoke?
10 or less 11 to 20 21 to 30 31 or more 5. Do you smoke more frequently during the first hours after waking than during the rest of the day?
Yes No 6. Do you smoke if you are so ill that you are in bed most of the day?
Yes No
47
Demographics 1) What is your gender? Female ____ Male ____ 2) How old are you? _______ years old 3) What is your ethnic background?
White, not Hispanic ____ Hispanic, of any race ____ African American, not Hispanic ____ Asian or Pacific Islander ____ Native American ____ Other ____
4) Please check your smoking experience. Never smoked _____ I used to smoke, do not smoke currently _____ I smoke currently _____
Thank you for your participation. Debriefing Statement The messages in the ads were fictional. Although those ads were created for this study, the messages in the ads were based on facts.
48
LIST OF REFERENCES
American Cancer Society. (2008, February 20). Cancer facts and figures 1996: Tobacco use. Retrived March 13, 2008, from http://www.cancer.org/statistics/96cff/tobacco.html
American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders,
(3rd ed.), Washington, D.C. Badura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological
Review, 84(2), 191-215. Bauer, R. A. (1960). Consumer behavior as risk taking. In Hancock R. (Ed.), Dynamic
Marketing for a Changing World, American Marketing Assoication, p389-398, Chicago, IL. Bauer, U. E., Johnson, T. M., Hopkins, R. S., & Brooks, R. G. (2000). Changes in youth
cigarette use and intentions following implementation of a tobacco control program. Journal of the American Medical Association, 284(6), 723-728.
Botvin, G. J., & Eng, A. (1980). A comprehensive school-based smoking prevention program.
Journal of School Health, 50, 209–213. Braun, K. A., Gaeth, G. J., & Levin, I. P. (1997). Framing effects with differential impact: the
role of attribute salience. Advances in Consumer Research, 27, 405-411. Beaudoin, C. E. (2002). Exploring antismoking ads: appeals, themes, and consequences. Journal
of Health Communication, 7, 123-137. Blum, A. (1980). Medicine vs. Madison avenue: fighting smoke with smoke. Journal of the
American Medical Association, 243(8), 739-740. Buto, C. P. (1987). The framing of buying decisions. Journal of Consumer Research, 14(3), 301-
315. Block, L. G., & Keller, P. A. (1995). When to accentuate the negative: the effects of perceived
efficacy and message framing on intention to perform a health-related behavior, Journal of Marketing Research, 32, 192-203.
Bock, B. C., Becker, B., Niaura, R., & Partridge, R. (2000). Smoking among emergency chest
pain patients: motivation to quit, risk perception and physician intervention. Nicotine & Tobacco Research, 2, 93-96.
Cacioppo, J. T., & Petty R. E. (1979). Effects of message repetition and position on cognitive
response, recall, and persuasion. Journal of Personality and Social Psychology, 37(1), 97-109.
49
Centers for Disease Control (CDC). (1991). Selected tobacco-use behaviors and dietary patterns among high school students-United States, 1991. Morbidity and Mortality Weekly Report, 41(24), 417-421.
Chandy, R. K., Tellis, G. J., Macinnis, D. J., & Thaivanich, P. (2001). What to say when:
advertising appeals in evolving markets. Journal of Marketing Research, 38(4), 399-414. Cunningham, S. M. (1967). The major dimensions of perceived risk. In Cox, D.F.(Ed.), Risk
taking and information handling in consumer behavior (pp. 82-108). Boston, MA: Harvard University Press.
Edwards, K. (1990). The interplay of affect and cognition in attitude formation and change.
Journal of Personality and Social Psychology, 59(2), 202-216. Gainer, B. (1993). An empirical investigation of the role of involvement with a gendered product.
Psychology & Marketing, 10(4), 265-283. Ganzach, Y., & Karsahi, N. (1995). Message framing and buying behavior: A field experiment.
Journal of Business Research, 32, 11-17. Goldman, L. K., & Glantz, S. A. (1998). Evaluation of antismoking advertising campaigns.
Journal of the American Medical Association, 279(10), 772-777. Grandpre, J., Alvaro, E. M., Burgoon, M., Miller, C. H., & Hall, J. R. (2003). Adolescent
reactance and anti-smoking campaigns: a theoretical approach. Health Communication, 15(3), 349-366.
Greenwald, A., & Leavitt, C. (1984). Audience involvement in advertising: Four levels. Journal
of Consumer Research, 11(June), 581-592. Hale, J. L., & Dillard, J. P. (1995). Fear appeals in health promotion campaigns: Too much, too
little, or just right? Designing health messages. Thousand Oaks, CA: Sage Publications. Hamilton, J. L. (1972). The demand for cigarettes: advertising, the health scare, and the cigarette
advertising ban. Review of Economics & Statistics, 54(4), 401-411. Hennrikus, D. J., Jeffery, R. W., & Lando, H. A. (1995). The smoking cessation process:
longitudinal observation in a working population. Preventive Medicine, 24, 235-244. Heatherton, T. F., Kozlowski, L. T., Frecker, R. C., & Fagerström, K. O. (1991). The Fagerström
Test for Nicotine Dependence: A revision of the Fagerström Tolerance Questionnaire. British Journal of Addiction, 86, 1119-1127.
Houston, M. J. & Rothschild, M. L. (1978). Conceptual and methodological perspectives on
involvement. In S. C. Jain (Ed.), 1978 Educator’s Proceedings (pp. 184-187). Chicago: American Marketing Assoication.
50
Hu, T., Sung, J., & Keeler, T. E. (1995). Reducing cigarette consumption in California: tobacco
taxes vs. an anti-smoking media campaign. American Journal of Public Health, 85(9), 1218-1222.
John, U., Meyer, C., Rumpf, H. J., & Hapke, U. (2003). Relation among stage of change,
demographic characteristics, smoking history, and nicotine dependence in an adult German population, Preventive Medicine, 37, 368-374.
Kanouse, D. E. (1984). Explaining negativity biases in evaluation and choice behavior: Theory
and research. Advances in Consumer Research, 11(1), 703-708. Kardes, F. R. (1988). Spontaneous inference processes in advertising: The effects of conclusion
omission and involvement persuasion. Journal of Consumer Research, 15(September), 225-233.
Kühberger, A. (1998). The influence of framing on risky decisions: a meta-analysis.
Organizational Behavior and Human Decision Processes, 75(1), 23-55. Lau, R. R. (1985). Two explanations for negativity effects in political behavior. American
Journal of Political Science, 29(1), 119-138. Levin, I. P. (1987) Associative effects of information framing. Bulletin of the Psychonomic
Society, 25, 85-86. Levin, I. P., Schneider, S. L., & Gaeth, G. J. (1998). All framed are not created equal: a critical
review of differences in valence framing effects. Organizational Behavior and Human Decision Processes, 76, 149-188.
MacKenzie, S. B., Lutz, R. J., & Belch, G. (1986). The role of attitude toward ad as a mediator
of advertising effectiveness: a test of comparing explanations. Journal of Marketing Research. 23(May), 130-143.
MacInnis, D. J., Moorman, C., & Jaworski, B. J. (1991). Enhancing and measuring consumers'
motivation, opportunity, and ability to process brand information from ads. Journal of Marketing, 55(4), 32-53.
Maheswaran, D., & Meyers-Levy, J. (1990). The influence of message framing and issue
involvement. Journal of Marketing Research, 27(3), 361-367. Manfredi, C., Lacey, L., Warnercke, R., & Buis, M. (1992). Smoking-related behavior, beliefs,
and social environment of young black women in subsidized public housing in Chicago. American Journal of Public Health, 82(2), 267-272.
MayoClinic. (2006, October 26). Nicotine dependence. Retrieved March 20, 2008, from
http://www.mayoclinic.com/health/nicotine-dependence/DS00307.
51
Meyerowitz, B. E., & Chaiken, S. (1987). The effect of message framing on breast self-
examination attitudes, intentions, and behavior. Journal of Personality and Social Psychology, 52 (March), 500-510.
Mitchell, A. A., & Olson, J. C. (1981). Are product attribute beliefs the only mediator of
advertising effects on brand attitude? Journal of Marketing Research, 18(August), 318-332. Oei, T. P. (1987). Smoking prevention program for children: A review. Journal of Drug
Education, 17, 11–42. O’Keefe, D. J. (2002). Persuasion: Theory & Research, (2nd ed.). 137-167, Thousand Oaks, CA: Sage Publications O'Keefe, T. (1971). The anti-smoking commercials: a study of television’s impact on behavior.
Public Opinion Quarterly, 35(2), 240-248. Olney, T. J., Holbrook, M. B., & Batra, R. (1991). Consumer responses to advertising: the effects
of ad content, emotions, and attitude toward the ad on viewing time. Journal of Consumer Research, 17(March), 440-453.
Olson, J. M., & Zanna, M. P. (1993). Attitudes and attitude change. Annual Review of
Psychology, 44, 117-154. Pechmann, C., Zhao, G.., Goldberg, M. E., & Reibling, E. T. (2003). What to convey in
antismoking advertisements for adolescents: the use of protection motivation theory to identify effective message themes. Journal of Marketing, 67(2), 1-18.
Pechmann, C., & Reibling, E. T. (2000). Planning an effective anti-smoking mass media
campaign targeting adolescents. Public Health Management Practice, 6(3), 80-94. Pechman, C. & Stewart, D. W. (1988). Advertising repetition: a critical review of wearin and
wearout. Journal of Current Issues & Research in Advertising, 11(2), 285-329. Petty, R. E., & Cacioppo, J. T. (1986). Communication and Persuasion: Central and Peripheral
Routes to Attitude Change. New York, New York: Springer-Verlag Petty, R. E., Cacioppo, J. T., & Schumann, D. (1983). Central and peripheral routes to
advertising effectiveness: the moderating role of involvement. Journal of Consumer Research, 10(September), 135-146.
Pierce, J. P., Gilpin, E. A., & Emery, S. L. (1998). Has the California tobacco control program
reduced smoking? Journal of the American Medical Association, 280(10), 893-899. Pratto, F., & John, O. P. (1991) Automatic vigilance: The attention-grabbing power of negative
social information, Journal of Personality and Social Psychology, 61(3), 380-391.
52
Popham, W. J., & Potter, L. D. (1993). Do anti-smoking media campaigns help smokers quit?
Public Health Reports, 108(4), 510-513. Ratneshwar, S., & Chaiken, S. (1991). Comprehension's role in persuasion: the case of its
moderating effect on the persuasive impact of source cues. Journal of Consumer Research, 18(1), 52-62.
Rindfleisch, A, & Crockett, D. X. (1999). Cigarette smoking and perceived risk: a
multidimensional investigation. Journal of Public Policy & Marketing, 18(2), 159-171. Rothman, A. J., & Salovey, P. (1997). Shaping perception to motivate healthy behavior: the role
of message framing. Psychological Bulletin, 121(1), 3-19. Rothschild, M. L. (1979). Advertising strategies for high and low involvement situations. In J.
Maloney, & B. Silverman (Eds.), Attitude Research Plays for High Stakes (pp. 74-93). Chicago: American Marketing Association.
Schneider, L., Klein, B., & Murphy, K. M. (1981). Governmental regulation of cigarette health
information. Journal of Law & Economics, 24(3), 575-612. Shimp, T. A. (1981). Attitude toward ad as a mediator of consumer brand choice. Journal of
Advertising, 10(2), 9-15. Shiv, B., Edell, J. A., & Payne, J. W. (1997). Factors affecting the impact of negatively and
positively framed ad messages. Journal of Consumer Research, 24, 285-294. Siegel, M. (2002). Antismoking advertising: figuring out what works. Journal of Health
Communication, 7, 157-162. Siegel, M., & Biener, L. (2000). The impact of an antismoking media campaign on progression
to established smoking: results of a longitudinal youth study. American Journal of Public Health, 90(3), 380-386.
Sly, D. F., Heald, G. R., & Ray, S. (2001). The Florida "truth" anti-tobacco media evaluation:
design, first year results, and implications for planning future state media evaluations. Tobacco Control, 10(1), 9-15.
Sly, D. F., Trapido, E., & Ray, S. (2002). Evidence of the dose effects of an antitobacco
counteradvertising campaign. Preventive Medicine, 35(5), 511-518. Smith, G. E. (1996). Framing in advertising and the moderating impact of consumer education.
Journal of Advertising Research, 36(5), 49-64.
53
Stone, R. N., & Winter, F. W. (1987). Risk: is it still uncertainty times consequences? In Belk, R.W. et al. (Ed.), Proceedings of the American Marketing Association (pp. 261-265). Chicago, IL: Winter Educators Conference.
Stuart, A. E., & Blanton, H. (2003). The effects of message framing on behavioral prevalence
assumptions. European Journal of Social Psychology, 33, 93-102. Umphrey, L. R. (2003). The effects of message framing and message processing on testicular
self-examination attitudes and perceived susceptibility. Communication Research Reports, 20(2), 97-105.
U.S. Public Health Services Centers for Disease Control and Prevention, Office on Smoking and
Health. (1994, July). Preventing tobacco use among young people: A report of the Surgeon General. Washington, DC: U.S. Government Printing Office.
Wakefield, M., Flay, B., Nichter, M., & Giovino, G. (2002). Effects of anti-smoking advertising
on youth smoking: a review. Journal of Health Communication, 8, 229-247. Wallack, L. M. (1983). Mass media campaigns in a hostile environment: Advertising as anti-
health education. Journal of Alcohol and Drug Education, 28(2), 51-63. Wallack, L. M., & Corbett, K. (1987). Alcohol, tobacco and marijuana use among youth: An
overview of epidemiological, program and policy trends. Health Education Quarterly, 14(2), 223-249.
Warner, K. E. (1977). The Effects of the anti-smoking campaign on cigarette consumption.
American Journal of Public Health, 67(7), 645-650. Weinberger, M. G., Allen, C. T., & Dillon, W. R. (1981). Negative information: Perspectives and
research directions. Advances in Consumer Research, 8(1), 398-404. Witte, K. (1995). Fishing for success: using the persuasive health message framework to
generate effective campaign messages, Designing health messages. 145-166, Thousand Oaks, CA: Sage Publications
Wong, C. O., & McMurray, N. E. (2002) Framing communication: Communicating the
antismoking message effectively to all smokers. Journal of Community Psychology, 30(4), 433-447.
Wright, P. (1974). The harassed decision maker: Time pressures, distractions, and the use of
evidence. Journal of Applied Psychology, 59(5), 555-561. Zhang, Y., & Buda, R. (1999). Moderating effects of need for cognition on responses to
positively versus negatively framed advertising message. Journal of Advertising, 28(2), p1-15.
54
Zhang, Y., & Neelankavil, J. P. (1997). The influence of culture on advertising effectiveness in China and the USA. European Journal of Marketing, 31(1/2), 134-149.
55
BIOGRAPHICAL SKETCH
Wan Seop Jung was born in Busan, Korea. He earned his Bachelor of Advertising and
Public Relations from Chung-Ang University in Seoul, Korea. He came to the University of
Florida in August 2006 to pursue a Master of Advertising. He is received his Master of
Advertising degree in August of 2008. He continues with his doctoral studies at the University of
Florida.
56