|
This paper was presented at the Association for Education in Journalism and
Mass Communication in San Antonio, Texas August 2005.
If you have questions about this paper, please contact the author
directly. If you have questions about the archives, email
rakyat [ at ] eparker.org. For an explanation of the subject line,
send email to
[log in to unmask] with just the four words, "get help info aejmc," in the
body (drop the "").
(Jan 2006)
Thank you.
Elliott Parker
====================================================================
The Effects of Self-Efficacy Statements in Humorous Anti-Alcohol
Abuse Messages Targeting College Students:
Who is in Charge?
By
Moon J. Lee, Assistant Professor, Washington State University
Myiah Hutchens Hively, MA Student, Washington State University
Edward R. Murrow School of Communication
Washington State University
PO Box 642520
Pullman, WA 99164
(509)335-4225
[log in to unmask]
Running head: Effects of self-efficacy in messages targeting college students
Abstract
This study examined the effects of self-efficacy statements in
humorous, positively reinforced anti-alcohol abuse messages. The
experiment was a post test only design with 124 college students.
Results indicate that highly rebellious individuals who watched ads
with a self-efficacy statement (i.e. You are in control of the
situation) indicated lower alcohol expectancies, higher risk
perceptions, and higher intentions to change their drinking behavior
than those in the non-self-efficacy condition. Implications of the
results and suggestions for future research are discussed.
Problems resulting from college students' excessive drinking (also
known as binge drinking problems) are not new, but continually cause
serious concerns for many parents, educators, health campaign
practitioners and government officials (Lee & Bichard, 2002; Lee &
Chen, 2004). Binge drinking, which has been defined as consuming more
than five drinks in one sitting for a male and more than four drinks
in one sitting for females (Wechsler, Dowdall, Maener, Gledhill-Hoyt,
& Lee, 1998) has been the subject of multiple research efforts
(Skuttle, 1999; Dorsey, Miller & Scherer, 1999; Oei & Burrow, 2000,
Hasking & Oei, 2002; Blume, Schmaling & Marlatt, 2003).
A study conducted by the Washington State Department of Health in
2002 indicated that 18 to 24 year olds are the most likely age group
to abuse alcohol. For example, it was found that 39 percent of 18 to
24 year olds indicated that they had at least one binge drinking
incident in the past month. In fact, it is estimated that more than
40 percent of college students are binge drinkers (Wechsler, et al., 1998).
Previous studies have investigated how to tailor messages to at-risk
individuals (known as rebellious risk-takers) (Lee & Ferguson, 2002;
Lee et al., 2002; Lee, 2003; Lee et al., 2004). However, although
several types of messages were identified as promising for targeting
at-risk individuals, the effort to identify effective messages to
achieve the desired results – i.e. an increased intention to modify
their drinking behaviors– has been less fruitful than desired.
Lee and Ferguson (2002) argued that message designers should look at
individuals' rebellious tendencies when designing messages because
individuals who are rebellious seem to react to messages differently
from individuals with low rebellious tendencies. The major
characteristic of rebellious individuals is that they tend to take
risks to oppose social norms and enjoy being labeled as a rebel
(Ferguson, Valenti & Melwani, 1991; Lee et al., 2002). Since these
individuals take risks for the notoriety of being a risk-taker, as
opposed to the perceived benefits of the action (Lee et al., 2002),
it is logical to take this tendency into consideration when designing
effective messages. In fact, Lee et al. (2002) found that rebellious
individuals respond better to humorous messages than messages that
utilized fear appeals in anti-tobacco advertisements. A follow up
study (Lee, 2003) again found the similar results in anti-alcohol
abuse messages. However, rebellious individuals' intentions to change
their drinking behaviors did not seem to be influenced by the
humorous messages.
One of the important and well-researched concepts in the health
communication field is self-efficacy (i.e., Annis & Davis, 1988;
Skuttle, 1999). Self-efficacy is defined as one's belief about
his/her ability to change or control his/her own behavior (Bandura,
1994). Research has shown that higher self-efficacy leads to fewer
binge drinking episodes (Blume, Schmaling & Marlatt, 2003; Oei &
Morawska, 2004) and lower alcohol consumption (Oei & Burrow, 2000;
Skuttle, 1999). Furthermore, increasing self-efficacy in alcohol
treatment facilities has shown to be successful in reducing drinking
problems (Hasking & Oei, 2002). However little research has been
conducted on how to design media messages that appeal to individuals'
self-efficacy and how adding self-efficacy statements in anti-alcohol
abuse messages influences at-risk individuals' reactions.
Therefore, the purpose of this study is to investigate the effects of
self-efficacy statements in anti-alcohol abuse messages particularly
when targeting rebellious individuals. An experiment was conducted
with 124 college students in a large northwestern university in which
half of the participants viewed humorous anti-alcohol abuse ads that
had not been manipulated (will be referred to as the
non-self-efficacy condition), and the other half of the participants
viewed humorous anti-alcohol abuse ads that had a textual
self-efficacy statement (e.g. You are in charge) inserted into each
ad (the self-efficacy condition).
Drinking and Risk Taking Tendencies.
In order to understand how to tailor messages to individuals at risk,
message designers need to understand how those individuals make
choices. Researchers have concluded that individuals' risk-taking
tendencies help to create a clearer picture of how they make
decisions (Ferguson et al., 1991; Lee et al., 2002; Moore & Gullone,
1996; Zuckerman, 1979).
Previous research has indicated that high sensation seeking
individuals are more likely to use drugs, and use them more often,
than peers who have lower levels of sensation seeking, especially
sensation seekers who fall under the categories of Disinhibition and
Boredom Susceptibility (Zuckerman, 1979; Donohew, Palmgreen & Lorch,
1994; Lee et al., 2002). A concept similar to sensation seeking, but
differed for its emphasis on behavioral characteristics is an
individual's risk-taking tendency, a behavioral tendency to take
different types of risk. Studies have shown that young adults have
higher risk-taking tendencies than those in younger or older age
groups (Ferguson et al., 1991; Lee et al., 2002). One of the most
relevant types to individuals' alcohol abuse is their rebellious
risk-taking tendency. Rebellious individuals are known to seek
stimulation through partying and social drinking. Rebellious
individuals tend to take risks in order to oppose social norms and
enjoy being labeled a rebel (Ferguson et al., 1991; Lee et al.,
2002). The notoriety of being a risk-taker in addition to the
perceived benefits of the action seems to motivate this type of
risk-taker (Lee et al., 2002). Therefore, this type of risk-takers
are hard to persuade even though they are a prime target for
anti-alcohol abuse messages.
Effective Message Design.
Researchers suggested that effective health campaign efforts should
target specific audiences in terms of their specific characteristics
(Austin & Meili, 1994; Lee et al., 2002). A handful of research has
been conducted as to how to better tailor media messages to their
intended audiences, specifically rebellious individuals because they
respond to messages differently and tend to put themselves at risk
through their behaviors (Lee et al., 2002; Lee, 2003; Lee et al.,
2004). One approach that has promising results was using humorous
appeals as opposed to fear appeals.
Humor Appeals. Traditionally, humor appeals are more frequently used
in promotional messages such as alcohol advertisements rather than in
health campaign messages. Humor is found to be effective in drawing
attention (Madden & Weinberg, 1984; Monahan, 1994; Weinberger &
Gulas, 1992), generating brand recognition, and enhancing liking of a
product (Weinberger et al., 1992).
Although there is some evidence that humor enhances persuasion (Batra
& Ray, 1986; Lee et al, 2002; Lee, 2003; Weinberger et al., 1992),
the effects of humor appeals do not seem to last long since humor
tends to elicit heuristic information processing rather than
thoughtful and elaborated information processing (Monahan, 1994).
Perhaps, this heuristic information processing of humor appeal
reduces the threatening feelings presented in the messages (Monahan,
1994). Humor appeals, therefore, can provide another means to reduce
a target audience's defensive mechanisms and enhance their
susceptibility of a recommended action in health communication
messages (Monahan, 1994; Lee et al., 2002). However, the major
challenge is how to design messages that evoke a positive emotion
that eventually leads to intended attitudinal, intentional, or
behavioral changes (Lee et al., 2002).
Recent research determined that individuals with rebellious
risk-taking tendencies react negatively to fear appeals, which have
been commonly used in anti-substance abuse messages (Lee et al.,
2002). However, when humor was used to relay the message, the
negative correlation between individuals' rebellious tendencies and
their intentions to change their behaviors was not present. Lee et
al., (2002) speculated that messages which use fear draw the
attention of rebellious individuals but will lead to defensive
reactions, which hinders further acceptance of a recommended action.
Therefore, message designers need to tailor their messages to target
this particular public in a manner that diminishes their defensive
reactions by looking at not only how to gain their attention, but
also how to minimize possible defensive reactions to the given messages.
One study found that heavy drinkers who watched the humorous
anti-alcohol abuse ads reported higher levels of intention to change
their drinking behaviors than those who watched the fear ads as well
as those who were in the control group (Lee, 2003). A follow-up study
examined how college drinkers would react to humorous anti-alcohol
abuse messages with different types of reinforcements; positive vs.
negative reinforcements (Lee et al., 2004). In the study, positive
reinforcement was classified as messages that contained positive
verbs such as do, can or be; and negative reinforcement was
classified as messages that contained negative verbs such as don't or
can't. It was found that moderate drinkers exhibited a higher level
of intention to change their drinking behavior in the negative
reinforcement condition than those in the positive reinforcement
condition. However, no significant condition effect was found among
heavy drinkers in terms of their intention to change their drinking
behaviors even though those who watched the ads with positive
reinforcements perceived higher levels of risk associated with
alcohol abuse than those who watched the ads with negative reinforcements.
Self-Efficacy Albert Bandura's Social Cognitive Theory, which
includes the concept of self-efficacy, has been widely applied in
health research as well as clinical practices. Bandura (1994) defines
self-efficacy as "people's beliefs about their capabilities to
produce designated levels of performance that exercise influence over
events that affect their lives" (p. 71).
According to Bandura (1994), self-efficacy can be built up through
"mastery experiences," vicarious experiences, and/or social
persuasion. Bandura (1994) stated that the best way to acquire
self-efficacy is through life-long experiences. However, it is also
possible for one to develop self-efficacy through vicarious
experiences or social persuasion, either by having social models for
an individual to identify with and learn from or by being helped to
believe that he/she is in control of his/her behavior.
People with high self-efficacy can rebound from failure more
easily than those with low self-efficacy. Self-efficacy influences
human functioning in cognitive, motivational, affective, and
selection process (Bandura, 1994). In cognitive processes, the higher
the sense of self-efficacy, the higher goals people set for
themselves. In motivational processes, self-efficacy determines not
only the goals people set for themselves, but also the amount of
effort, the length of perseverance when faced with difficulty, and
their resilience to failures. In affective processes, self-efficacy
helps exercise control over such negative emotions as anxiety,
depression and stress. In selection processes, perceived
self-efficacy determines the kind of life people choose to live and
the kind of occupation they choose to pursue.
Bandura (1995) later emphasized the health-promoting role of
self-efficacy. He pointed out that perceived self-efficacy was an
important determinant of people's beliefs in their ability to cope
with and their execution of direct control over health-impairing
habits such as cigarette smoking, alcohol and drug abuse, and
sexually transmitted diseases. "People's beliefs that they can
motivate themselves and regulate their own behavior play a crucial
role in whether they even consider changing detrimental health
habits" (p. 28).
Self-Efficacy and Drinking Behavior Many health researchers have used
Bandura's concept to help explain problematic drinking behaviors and
how to treat them (Oei & Morawsak 2004; Blume, Schmaling & Marlatt
2003; Hasking & Oei 2002; Oei & Burrow 2000; Dorsey, Miller & Scherer
1999; Skuttle 1999; Annis & Davis 1988).
Skuttle (1999) discovered that self-efficacy scores using Annis'
Situational Confidence Questionnaire (1984) were negatively
correlated with perceived benefits from drinking (alcohol expectancy)
and amounts of abuse. In other words, the lower self-efficacy scores
individuals had, the higher perceived benefits from drinking as well
as the higher amounts of abuse they exhibited.
Several researchers have looked specifically at college-aged students
and how their perceptions of self-efficacy relate to their drinking
behaviors (Dorsey, Miller & Scherer, 1999; Blume, Schmaling &
Marlatt, 2003). It was discovered that the students' self-efficacy
had a strong effect on their drinking behaviors. Students with higher
self-efficacy scores consistently had fewer binge drinking
experiences. The researchers caution, however, that all results are
based only on self-reported short-term effects, so more research
needs to be conducted in order to determine the long term effects of
self-efficacy.
Oei & Burrow (2000) further examined drinking refusal self-efficacy
to assure that drinking refusal self-efficacy was measuring
self-efficacy regarding drinking behaviors rather than other types of
substance abuses such as smoking or caffine consumption. Their
research once again indicated that self-efficacy, specifically
drinking refusal self-efficacy, was a critical factor in alcohol
consumption. In addition, it was speculated that alcohol consumption
would be better predicted by self-efficacy than alcohol expectancies.
A follow-up study (Oei & Morawsak, 2004) found that self-efficacy was
indeed a better predictor than alcohol expectancies for alcohol
consumption. Self-efficacy predicted both amount and frequency of
consumption while alcohol expectancies only predicted whether or not
they would drink at all.
Hasking & Oei (2002) examined the relationships between self-efficacy
and coping in both community and clinical samples to predict alcohol
consumption. The researchers used the Kharvari Alcohol Test (Khavari
& Farber, 1978) to measure frequency of drinking, the Drinking
Expectancy Profile, parts one and two (Young & Oei, 1996), to measure
self-efficacy and alcohol expectancies, and the Cope Scale (Carver,
Scheier & Weintraub, 1989) to measure different ways of coping. It
was found that self-efficacy was significantly correlated to
frequency and amount of alcohol consumed in the community sample.
While the clinical group had lower self-efficacy scores, it was not
as highly correlated to alcohol consumption as the community group.
The most significant factor for the clinical group was alcohol
expectancies, and the most significant for the community group was
refusal self-efficacy. Hasking et al. (2002) concluded that in
moderate drinkers, self-efficacy is the predominate factor when
deciding how much to drink.
Based on the literature on self-efficacy and its relationship to
alcohol consumption and treatment, as well as individuals' rebellious
risk-taking tendencies, it was suspected that encouraging an
individual's self-efficacy though anti-alcohol abuse media messages
would influence individuals with rebellious tendencies positively.
The purpose of this study was to investigate the effects of
self-efficacy statements in humorous, positively reinforced
anti-alcohol abuse messages based on individuals' rebellious
tendencies. It was suspected that rebellious individuals would
exhibit less defensive reactions to the messages with self-efficacy
statements and greater intentions to change their drinking behaviors
than to the messages without self-efficacy statements.
Hypotheses These hypotheses were derived from the existing literature
on risk-taking tendencies and the effects of self-efficacy. Five main
hypotheses with one research question were examined in this study.
H1a: Perceptions of fear will be greater in the non-self-efficacy
condition than the self-efficacy condition.
This hypothesis is driven from the assumption that self-efficacy
messages would encourage the individual's feeling of control over
their actions, reducing a feeling of vulnerability. Therefore, it is
hypothesized that self-efficacy messages will reduce the
participants' fearful response to the messages.
H1b: Perceptions of fear will be greater for low rebellious
individuals than high rebellious individuals.
Previous research (Lee et al., 2002; Lee 2003) found that
low-rebellious individuals tend to be more scared by health messages
about their drinking behaviors than high-rebellious individuals.
H2: Highly rebellious college students in the self-efficacy
condition will like the ads better than those in the
non-self-efficacy condition while the same effect is not expected
among low rebellious college students.
Encouraging self-efficacy is expected to increase the liking of the
ads by lowering the defensive reactions of rebellious individuals. It
is hypothesized that this will be true only for the high rebellious
individuals because previous studies (Lee et al., 2002; Lee 2003)
have shown that low rebellious individuals are less likely to exhibit
defensive reactions to the media messages.
H3: Rebellious participants who view the ads with the self-efficacy
messages will have higher levels of drinking refusal self-efficacy
than those in the non-self-efficacy condition.
H4: Rebellious participants' perceptions of risks involved with
drinking will be greater for individuals in the self-efficacy
condition than the non-self-efficacy condition.
H5: Rebellious participants' intention to change their drinking
behaviors will be greater for those in the self-efficacy condition
than the non-self-efficacy condition.
It is hypothesized that using self-efficacy statements to lower
rebellious participants' defensive reactions to anti-alcohol abuse
messages will lead to higher information processing, which will lead
to higher levels of intention to change their risky behaviors.
RQ1: Will alcohol expectancies be influenced by the self-efficacy
statements in the humorous, positively reinforced anti-alcohol abuse
ads based on individuals' rebellious risk-taking tendencies?
Method
This study examined whether self-efficacy statements in humorous,
positively reinforced, anti-alcohol abuse advertisements would reduce
rebellious individuals' defensive reactions to "drink responsibly"
advertisements, in turn increasing their likelihood of acceptance of
the messages and leading to a higher intention to change their
drinking behaviors. The experimental design was a post-test only
model with a self-efficacy condition and a non-self-efficacy
condition. One hundred and twenty four participants from a large
university in the northwestern United States participated and were
compensated either with extra credit or monetarily.
Dependant variables in the experiment included liking of the ads,
alcohol expectancies, intention to change their drinking behaviors,
drinking refusal self-efficacy, fear of drinking and perceptions of
risks involved with drinking.
Developing the Stimuli
Choosing the ads. Two graduate students and one undergraduate student
in communication reviewed and selected humorous television ads from
previous studies (Lee, 2003; Lee et al., 2004) and newer ads on the
Internet. Three coders (communication graduate and undergraduate
students) rated existing humorous alcohol ads using the humor/fear
scale created by Lee et al. (2002). Humorous ads were defined as ads
that use humor to gain attention and try to deter excessive drinking
or encourage responsible drinking. These ads often use light
consequences of drinking, such as gaining weight or sounding
obnoxious. All ads used were positively reinforced, meaning they
included positive verbs such as "do" or "can" instead of negative
verbs such as "don't" or "can't" (Lee et al., 2004) in taglines. In
addition, all of the ads selected were anti-alcohol abuse ads, and
the majority encouraged responsible drinking.
The coders' ratings and quality of the ads were used to determine
the final six clips that were used. The titles of the clips given by
the producers are; "You Know When To Stop," "Guys in the Desert,"
"Drink Responsibly," "Slash," "Responsibility Matters" and "Pit Crew."
Self-Efficacy. The self-efficacy messages were created by two
graduate students, an undergraduate student and a professor in
communication and were added to the end of each ad for the
experiment. Two tapes were created for the experiment. The
non-self-efficacy condition's tape was not edited or changed at all
from how the ads appear on television, and the self-efficacy
condition's tape had a graphic inserted onto the ad which contained a
self-efficacy statement. The slogans were added onto the existing ad
instead of on a separate screen to make the ads seem as realistic as possible.
The messages were "Only You Control the Situation," "You Can Break
the Myth," "Drink Responsibly, You Can Change," "You Can Make a
Better Choice," "You Can Make a Difference Too," and "You're in Charge."
"Only You Control the Situation" was used in the "You Know When to
Stop" ad and was preceded by a graphic which read "Knowing When to
Stop is a Good Thing." "You Can Break the Myth" was used with the
"Guys in the Desert" ad and was preceded by a graphic which read
"Alcohol, it's Not as Cool as You Think." "Drink Responsibly, You Can
Change" was used with the "Drink Responsibly" ad and had no other
writing with it. "You Can Make a Better Choice" was used with "Slash"
and was preceded by a graphic that read "Drink Intelligently." "You
Can Make a Difference Too" was used with the "Responsibility Matters"
ad and was preceded by a graphic which read "We All Make a
Difference." "You're in Charge" was used with the "Pit Crew" ad and
was preceded by a graphic which read "Drink Intelligently."
Experimental Procedure. Upon arrival, students were greeted and
randomly assigned to one of the conditions. At the beginning of the
experiment, the participants were given a survey designed to measure
their risk-taking tendencies (Ferguson et al, 1991; Lee et al, 2002),
their general self-efficacy (Schwarzer & Fuchs, 1995), and drinking
refusal self-efficacy (Young & Oei, 1996), which was modified by a
professor and three communication students. After completing the
pretest survey, the participants were asked to watch either the tape
with the self-efficacy statements (n = 65), the self-efficacy
condition, or the tape which had not been altered (n = 59),
non-self-efficacy condition.
After watching the ads, participants were asked to answer Likert
scale questions (on a scale of 0 to 9) regarding their fear of
drinking, liking of the ads, intention to change their drinking
behavior, alcohol expectancies, perceived risks of alcohol, and
drinking refusal self-efficacy.
Data Analysis
Creating the rebellious scale Ten items from a scale created by Lee
et al. (2002) were used to create the rebellious scale. The items
were; "I like wild parties," "I am rebellious," "I often do things
spontaneously," "Life without danger would be too dull for me," "I
enjoy doing things that others find dangerous," "I sometimes like to
do things that are frightening," "I'm likely to do drugs when I
party," "I believe rules are meant to be broken," "I like driving
fast," and "I would love to have new and exciting experiences, even
if they are illegal." The Cronbach's alpha score for the ten items is
.88, with 48% of variance explained. An index of the participants'
responses to the ten items was created and their average score was
noted. Participants whose index score was between 0 and 4.9 were
classified as low rebellious, n = 76, and participants whose index
score fell between 5 and 9 were classified as high rebellious, n = 47.
Manipulation check To assure that the same level of humor was
perceived in the two groups, ten Likert-type (scale of 0 to 9)
questions from the index created by Lee et al. (2002) were asked
after viewing the ads. The items were; "One of the things I liked
about these ads was how funny they were," "These ads are amusing,"
"Those people in the ads are funny," "I found myself laughing when I
watched these ads," "I found these ads exciting and stimulating,"
"These ads truly held my interest," "I like these ads very much," "I
think the advertisements I just saw are very funny," "I enjoy the
humor used in these ads," and "I found myself feeling very good after
I watched these ads." The Cronbach's alpha score for these ten items
is .93, with 62% of variance explained. An independent t-test was
conducted to ensure that the same level of humor was perceived in
each group. As expected, there was not a significant difference, p = .46.
Dependant variables Several questions were asked to measure the
participants perceptions of fear, liking, drinking refusal
self-efficacy, risk perceptions, intention to change their drinking
behavior and alcohol expectancies. A factor analysis was conducted
and a summed factor score was created to measure each dependant variable.
Perceptions of Fear Five items loaded onto one factor was used to
measure perceptions of fear in the post test survey. The items were;
"These ads made me think a great deal about the dangers of drinking,"
"These ads scare me about the dangers of drinking," "I found myself
feeling very frightened when I watched these ads," "Ads like these
truly make me afraid to drink," and "These ads remind me of how risky
it is to drink." The Cronbach's alpha score for the five items is
.82, with 59% of variance explained.
Liking Five items loaded onto one factor was used to measure liking
of the ads. The items were; "I like these ads very much," "These ads
are cool," "I can relate myself to the ads," "The portrayals in the
ads are possible," and "I had a strong emotional reaction to these
ads." The Cronbach's alpha score for these items is .71, with 47% of
variance explained.
Drinking Refusal Self-Efficacy Three items loaded onto one factor was
used to measure drinking refusal self-efficacy. The items were; "I
can stop drinking anytime I want," "I can stop drinking even if my
friends insist that I drink," and "I can stop drinking anytime I
want." The Cronbach's alpha score for these five items is .79, with
71% of variance explained.
Risk Perceptions Five items loaded onto one factor was used to
measure individuals' perceptions of risk involved with drinking. The
items were; "I consider myself to be at risk of becoming an
alcoholic," "I drink too much," "I consider myself to be at risk of
getting in an automobile accident due to my drinking," "I consider
myself to be at risk of getting alcohol-related injuries," and "I
consider myself to be at risk of becoming an alcoholic." The
Cronbach's alpha score for these five items is .85, with 62% of
variance explained.
Alcohol Expectancies Five items loaded onto one factor was used to
measure alcohol expectancies. The items were; "Drinking facilitates a
social atmosphere," "Drinking helps relationships," "Drinking makes
people relaxed," "People who drink are relaxed, easy-going people,"
and "Drinking makes people happy." The Cronbach's alpha score for the
five items is .77, with 53% of variance explained.
Intention to Change Drinking Behavior Four items loaded onto one
factor was used to measure intention to change behavior. The items
were; "I drink too much," "I consider myself to be at risk of
becoming an alcoholic," "I plan on changing my drinking habits very
soon," and "I would very much like to change my current drinking
habits." The Cronbach's alpha score for these four items is .83, with
67% of variance explained.
Results
H1a: Perceptions of fear will be greater in the non-self-efficacy
condition than the self-efficacy condition.
There was a significant condition effect on perceptions of fear in
the ads, F (1,119) = 550.72, p < .05. Participants in the
self-efficacy condition (n = 64, M = .18, S.D. = .96) indicated
higher perceptions of fear than those in the non-self-efficacy
condition (n = 59, M = -.18, S.D. = .85). Therefore this hypothesis
was not supported; in fact the direction came out to be the opposite.
H1b: Perceptions of fear will be greater for low rebellious
individuals than high rebellious individuals regardless of the condition.
There was a significant rebelliousness effect on perceptions of fear
in the ads, F (1, 119) = 368.90, p < .05. Low rebellious individuals
(n = 76, M = .13, S.D. = .95) perceived more fear than the high
rebellious individuals (n = 47, M= -.18, S.D. = .85). Therefore this
hypothesis was supported.
----- Insert figure 1 here -----
H2: Highly rebellious college students in the self-efficacy condition
will like the ads better than those in the non-self-efficacy
condition while the same effect is not expected among low rebellious
college students.
There was no significant condition effect for liking of the ads, F
(1, 119) = 1.82, p = .41. Therefore this hypothesis was not
supported. However, it should be noted that among the low rebellious
individuals the anticipated direction was different than expected. A
simple t-test of the low rebellious individuals indicated a
significant condition effect for the low rebellious individuals, t
(1.75) = 2.14, p < .05. The individuals in the self-efficacy
condition, (n= 41, M = .19, S.D. = .85) indicated higher levels of
liking of the ads as compared to the individuals in the
non-self-efficacy condition (n = 35, M = -.24, S.D. = .89).
----- Insert Figure 2 here -----
H3: Rebellious participants who view the ads with the self-efficacy
messages will have higher levels of drinking refusal self-efficacy
than the individuals in the non-self-efficacy condition after viewing the ads.
There was no significant condition effect for drinking refusal
self-efficacy, F (1, 119) = 3.54, p = .31. Therefore this hypothesis
was not supported. However, it should be noted that the direction was
the opposite of what was predicted. The participants in the
self-efficacy condition (n = 64, M = -.11, S.D. = 1.04) indicated
lower levels of drinking refusal self-efficacy than participants in
the non-self-efficacy condition (n = 59, M = .15, S.D. = .77).
A simple t-test among the high rebellious individuals showed a
significant condition effect, t (1.46) = -1.63, p = .05. The
individuals in the self-efficacy condition (n = 23, M = -.43, S.D. =
1.13) indicated significantly lower levels of drinking refusal
self-efficacy than the participants in the non-self-efficacy group (n
= 24, M = .04, S.D. = .84).
----- Insert Figure 3 here -----
H4: Rebellious participants' perceptions of risks involved with
drinking will be greater for individuals in the self-efficacy
condition than the non-self-efficacy condition.
There was a significant condition effect for perceptions of risks
involved with drinking, F (1, 119) = 80,044.91, p < .01. Individuals
in the self-efficacy condition (n = 64, M = .16, S.D. = 1.01)
indicated higher levels of perceived risks of drinking than the
non-self-efficacy condition (n = 59, M= -.19, S.D. = .83). Therefore
this hypothesis was supported.
There was also a significant rebelliousness effect for perceptions of
risks, F (1, 119) = 143,857.53, p < .01. High rebellious individuals
(n = 47, M = .29, S.D. = .99) indicated higher perceptions of risks
involved with drinking than low rebellious individuals (n = 76, M =
-.19, S.D. = .86).
----- Insert Figure 4 here -----
H5: Rebellious participants' intention to change their drinking
behavior will be greater for those in the self-efficacy condition
than the non-self-efficacy condition.
There was a significant condition effect for intention to change
drinking behavior, F (1, 106) = 1444.31, p < .05. Individuals in the
self-efficacy condition (n = 56, M = .09, S.D. = .97) indicated
higher levels of intention to change their drinking behavior than
individuals in the non-self-efficacy condition (n = 54, M = -.14,
S.D. = .86). Therefore this hypothesis was supported.
In addition, there was a significant rebelliousness effect for
intention to change drinking behavior, F (1, 106) = 2746.97, p < .05.
High rebellious individuals (n = 45, M = .17, S.D. = .96) indicated
higher levels of intention to change their drinking behavior than low
rebellious individuals (n = 65, M = -.16, S.D. = .88).
------ Insert Figure 5 Here ------
RQ1: Will alcohol expectancies be influenced by the self-efficacy
statements in the humorous, positively reinforced anti-alcohol abuse
Ads based on individuals' rebellious risk-taking tendencies?
There was a significant interaction effect for alcohol expectancies,
F (1, 119) = 3.85, p = .05. Alcohol expectancies for high rebellious
individuals in the self-efficacy condition (n = 23, M = .18, S.D. =
.71) were lower than high rebellious individuals in the
non-self-efficacy condition (n = 24, M = .54, S.D. = .77). However,
low rebellious individuals in the self-efficacy condition (n = 41, M
= -.11, S.D. = .93) indicated higher alcohol expectancies than the
low rebellious individuals in the non-self-efficacy condition (n =
35, M = -.37, S.D. = .92).
A simple t-test was conducted on both high and low rebellious
individuals and a significant condition effect was found (t = -1.69,
(1.46), p = .05) for the high rebellious individuals. Alcohol
expectancies were higher for the high rebellious individuals in the
non-self-efficacy condition (n = 24, M = .54, S.D. = .77) than the
high rebellious individuals in the self-efficacy condition, (n = 23,
M = .18, S.D. = .71).
------ Insert Figure 6 Here ------
Discussion and Conclusion
This study examined the effect of self-efficacy statements in
humorous, positively reinforced anti-alcohol abuse television
advertisements on college students. It was discovered that adding
self-efficacy messages to humorous anti-alcohol abuse ads
significantly increased highly rebellious individuals' intentions to
change their behaviors, increased their perceived risks of drinking,
and decreased their perceived fear in the messages. In addition, the
high rebellious individuals in the self-efficacy condition had
significantly lower drinking refusal self-efficacy and alcohol
expectancies. In fact, the increase in risk perceptions and the
decrease in drinking refusal self-efficacy were contrary to the
original expectations.
Clinical studies have shown that increasing self-efficacy in
alcoholic patients leads to a decrease in drinking behavior (Annis et
al., 1988) and additional studies show that alcohol expectancies when
paired with self-efficacy also have an effect on drinking behavior
(Oei et al., 2000; Oei et al., 2004). Bandura's (1994) theory of how
self-efficacy effects human functioning could be part of the
explanation as to how perceived self-efficacy is affecting drinking
behavior. Bandura (1994) posits that self-efficacy serves to
determine the kind of life people want to live and how individuals
control negative emotions. Therefore, it logically follows that if
individuals have low self-efficacy they will set lower goals for
themselves and make poor choices regarding how to handle negative
situations. Therefore, increasing their self-efficacy will lead them
to make better choices.
In this study researchers attempted to increase individuals'
self-efficacy through media messages, in order to make them think
about the choices they can make. Perhaps, self-efficacy messages lead
high rebellious individuals to engage in self-evaluation and become
more realistic in their views of their ability to control their
alcohol consumption thereby losing their feelings of invincibility.
It appeared that when rebellious individuals were encountered with
messages that they couldn't rebel against, or they agreed with, (i.e.
You're in charge) their responses became anything but the typical
highly rebellious individual's defensive response.
Another explanation could be the relationship to alcohol
expectancies. Studies have already shown that lower alcohol
expectancies lead to a decrease in drinking behavior (Blume et al.,
2003; Oei et al., 2004) so it is possible that the speculated
self-evaluation that is triggered by self-efficacy messages is
decreasing alcohol expectancies thereby increasing intentions to
change their drinking behavior. This is an area that deserves more attention.
There are, however, several limitations that should be considered
carefully. First of all, the experiment was conducted in an
artificial setting, in a classroom with at least one moderator in the
room. This could have influenced the participants' responses, as they
knew they were being watched. Another potential limitation is that
the data were based on the participants' self-reports, which while
known to be reliable, could have been suppressed or exaggerated by
the participants.
In addition, the findings are based on the participants' short-term
responses right after they watched the ads. Any effect that was noted
in this experiment should be considered as a short-term effect.
Therefore, a longitudinal study with a quasi field experiment design
may provide useful insights in terms of long term effects of
self-efficacy statements in anti-alcohol abuse media messages.
Another potential limitation in this study was the participants. A
majority of the participants were communication students and they
could be biased by their media awareness or through their familiarity
with the researchers. It is recommended that the experiment be
conducted again with a more educationally diverse group.
In conclusion, self-efficacy statements in humorous anti-alcohol
abuse ads appears to be effective in reducing highly rebellious
individuals' defensive reactions to the messages and seem to increase
the susceptibility of recommended actions in media messages. However,
more research is needed to determine its long-term effects and should
be repeated to rule out any possible bias.
Finding messages to help reduce possible defensive responses from
this target audience while at the same time increasing the likelihood
of their acceptance of recommended actions is a very challenging
task. However, efforts to identify effective ways to communicate with
this target audience should continue. Such efforts will lead us to a
better understanding of human communication, as well as more
effective ways to offer direction to individuals, especially young
adults, for the promotion of a healthy lifestyle, without setting off
their defensive reactions.
References
Austin, E. W., & Meili, H. K. (1994). Effects of interpretations of
televised alcohol
portrayals on children's alcohol beliefs. Journal of Broadcasting &
Electronic Media, 38, 417-
435.
Bandura, A. (1994). Self-efficacy. In V.S. Ramachaudran (Ed.),
Encyclopedia of human behavior. New York: Academic Press.
Bandura, A. (1995). Exercise of personal and collective efficacy in
changing societies. In A. Bandura (Eds.), Self-efficacy in changing
society. Cambridge: Cambridge University Press.
Batra, R., & Ray, M. L. (1986). Affective responses mediating
acceptance of advertising.
The Journal of Consumer Advertising, 13, 234-249.
Blume, A.W., Schmaling, K.B., Marlatt, A.G. (2003). Predictors of
change in binge drinking over a 3-month period. Addictive Behaviors,
28(5), 1007-1012.
Donohew, L., Palmgreen, P., & Lorch, E.P. (1994). Attention, need for
sensation, and health communication campaigns. American Behavioral
Scientist, 38(2), 310-323.
Dorsey, A.M., Miller, K.I., Scherer, C.W. (1999). Communication,
Risk Behavior, and Perceptions of Threat and Efficacy: A Test of a
Reciprocal Model. Journal of Applied Communication Research, 27, 377-395.
Hasking, P.A., Oei, T.P.S. (2002). The differential role of alcohol
expectancies, drinking refusal self-efficacy and coping resources in
predicting alcohol consumption in community and clinical samples.
Addiction Research and Theory, 10(5), 465-494.
Ferguson, M.A., Valenti, J.M., & Melwani, G. (1991). Communicating
with risk takers: a public relations perspective. Public Relations
Research Annual, 3, 195-224.
Khavari, K.A., Farber, P.D. (1978). A profile instrument for the
quantification and assessment of alcohol consumption: The Khavari
Alcohol Test. Journal of Studies on Alcohol, 39(9), 1525-1539.
Lee M. J. & Chen, Y. (May, 2004). Effectiveness of positive and
negative reinforcements in humor advertisements targeting heavy and
moderate drinkers. Paper presented at International Communication
Association, New Orleans.
Lee, M. J. (2003). Effects of emotional appeals targeting college
students for the prevention of binge-drinking: Humor works.
Manuscript submitted for publication.
Lee, M. J. & Bichard, S. (May, 2002). Effective message design
targeting college students for the prevention of binge-drinking: Are
they rebels? Paper presented at International Communication
Association, South Korea.
Lee, M. J. & Ferguson, M. A. (2002). The effects of anti-tobacco
advertisements based on risk-taking tendencies: Realistic fear ads
versus gross humor ads. Journalism & Mass Communication Quarterly, 79
(4), 945-963.
Madden, T. J., & Weinberg, M. G. (1984). Humor in advertising: A
Practioner view. Journal of Marketing Advertising Research, 24 (4), 23-79.
Monahan, J. L. (1994). Thinking positively: Using positive affect
when designing messages. In E. Maibach & R. Parrott (Eds.), Designing
Health Messages (pp. 81-98). CA: Sage.
Moore, S. & Gullone, E. (1996). Predicting adolescent risk behavior
using a personalized cost-benefit analysis. Journal of Youth and
Adolescence, 25(3), 343-359.
Oei, T.P.S., & Baldwin, A.R. (1994). Expectancy theory: A two-process
model of alcohol use and abuse. Journal of Studies on Alcohol, 55, 525-534.
Oei, T.P.S., & Burrow, T. (2000). Alcohol expectancy and drinking
refusal self-efficacy: A test of specificity theory. Addictive
Behaviors, 25(4), 499-507.
Oei, T. P. S., & Morawska, A. (2004). A cognitive model of binge drinking: The
influence of alcohol expectancies and drinking refusal self-efficacy.
Addictive Behaviors, 29(1), 159-179.
Skuttle, A. (1999). The relationship among self-efficacy
expectancies, severity of alcohol abuse, and psychological benefits
from drinking. Addictive Behaviors, 24 (1), 87-98.
Wechsler, H., Dowdall, G. W., Maener, G., Gledhill-Hoyt, J., & Lee,
H. (1998). Changes in binge drinking and related problems among
American college students between 1993 and 1997: Results of the
Harvard School of Public Health College Alcohol Study. Journal of
American College Health, 47(2), 57-69.
Weinberger, M., & Gulas, C. (1992). The impact of humor in
advertising: A review. Journal of Advertising, 21(4), 35-59.
Zuckerman, M. (1979). Sensation seeking: Beyond the optimal level of
arousal. Hillsdale, NJ: Lawrence Erlbaum Associates.
Figure 1
Perceptions of fear by condition and rebelliousness
Figure 2
Liking of the ads by condition and rebelliousness
Figure 3
Drinking refusal self-efficacy by condition and rebelliousness
Figure 4
Risk perceptions by condition and rebelliousness
Figure 5
Intention to change drinking behavior by condition and rebelliousness
Figure 6
Alcohol expectancies by condition and rebelliousness
|
