‘Why would I believe this?’ – Resistance Strategies Used
against Persuasive Health Messages
Student Name: Wendy Steenbergen
Student Number: 10750495
Master’s Thesis; Graduate School of Communication
Master’s programme Communication Science
Supervisor: Marieke Fransen
Date: 26-06-2015
Master Thesis
2014 - 2015
A big challenge that health advocates face when promoting persuasive health messages is resistance against their messages. Audiences often feel threatened in their freedom whenever they encounter these types of persuasive messages and as a result they may start resisting it. Previous literature has suggested different ways to counter this resistance, however research on the type of strategies that audiences’ use is lacking. Because in order to properly tackle resistance, it should first be
established which exact strategies participants are employing.
This study fills this gap by presenting an overview of the different resistance strategies identified in the literature and test which of these strategies are being used by audiences. These strategies were tested by analyzing the dyads of participants after having been exposed to a persuasive health message about binge drinking. Hereby, this study aimed to make a clearer connection between the various resistance strategies and test whether audiences actually employ these strategies. This was turned into a versatile codebook that can be used to expand the current knowledge on resistance. The results indicate that audiences indeed resist against persuasive health messages and hereby often use multiple strategies. The denial strategies were the most commonly used, followed by the cognitive reappraisal strategies and the suppression strategies. Avoidance strategies were hardly ever used in this study. The results also show that there are significant gender differences when it comes to resistance. Moreover, the type of emotional frame used in the
persuasive message influences whether participants resist against the message. These results can be used both by academia and professionals to further expand the existing knowledge on resistance and test which strategies their target audience uses by means of the codebook.
Resistance strategies against persuasion
The main goal of health-promoting persuasive messages is to motivate target audiences to change their behavior (van ‘t Riet & Ruiter, 2013). However, a big problem that health advocates face is that their target audience is often not influenced by their messages. Recipients of health messages often perceive these messages to be threatening to their freedom and, more importantly, their health (Witte, 1992; Koningsbruggen & Das, 2009). As a result, they may resist against the message to prevent being persuaded, which may even lead to a boomerang effect whereby audiences do the opposite of the desired behavior (Ahluwalia, 2000; Jacks & Cameron, 2013).
Resistance against persuasion is not a new phenomenon and has been studied for a longer period of time in various research fields such as: communication, advertising, marketing and psychology (Helm, 2004; Fransen et al. 2015; Sharot et al., 2011). Several of these papers studied which type of strategies audiences use to resist against a message (Ahluwalia, 2000; Jacks & Cameron, 2003; van ‘t Riet & Ruiter, 2013; Fransen et al., 2015). However, a problem with the current literature on resistance is that most of the studies have only analyzed individual resistance strategies. Hereby, many used slightly different terminology and neglect to make a connection between the different strategies. Therefore, it is not possible to study which exact strategies are being employed by audiences. Thus, further research combining the different strategies is needed.
This study aims to fill this gap by providing an overview of the existing resistance strategies. Based on these strategies a codebook will be developed, which will be used to study resistance by analyzing dyads of participants after having been exposed to a persuasive video about binge drinking. It will be analyzed whether participants resist against the message and if so, which strategy or strategies they are employing. Based on these results, a final codebook will be established with all the prevalent resistance strategies. The research question is:
The findings of this study will result in a clear codebook containing all the strategies used by audiences; this can be used in later studies on resistance so that the knowledge on this topic can be further expanded. Moreover, these results are also relevant for health professionals; Fransen et al. (2015) studied how organizations could best counter the resistance of their audiences and argued that their tactics would be more successful if they were tailored to the exact strategy that the audience is employing. And this study provides organizations a way to identify this.
Theoretical Background
As mentioned in the introduction, this study aims to examine which resistance strategies audiences’ use against persuasive health messages. However, in order to study resistance it is important to first define what it exactly entails and what the different motives are that audiences have to resist against a message. Moreover, an overview will be provided of the recognized resistance strategies in the current literature. These strategies will be used to develop a codebook that will guide the further analysis of resistance in this paper. Additionally, this study will also focus on other factors that influence resistance; such as previous behavior, gender and the type of frame in the message.
Resistance Against Persuasion
The main goal of a persuasive health message is to motivate change in the behavior of their audiences (van ‘t Riet & Ruiter, 2013). It can either motivate audiences to engage in a certain behavior – for example taking vaccinations –, or motivate audiences to quit certain behaviors – for example to stop smoking –. Either way, the main purpose is to promote change, which can be problematic due to the fact that audiences are often quite reluctant to change their behavior, which can in turn lead to resistance (Friestad & Wright, 1994).
Resistance is a concept that is relatively difficult to define since it can be seen as both an outcome as well as a motive (Knowles & Linn, 2004). One the one hand, resistance can be defined
as “a reaction against change” (Knowles & Linn, 2004, p.4), in which case it is a motive. The audience feels that a message is trying to influence or change them, which makes them motivated to act against it. But on the other hand, it can also be an outcome, namely whether a message has succeeded in influencing the audience or whether resistance impeded this (Knowles & Linn, 2004). For the purpose of this paper resistance will be defined as a motive, since the focus is on studying what types of strategies consumers use in order to resist persuasion. In short, resistance is defined as the ability to resist against a persuasive attack.
Reasons for Resistance
There are two main reasons for resistance, namely threat of freedom and fear of change (Friestad & Wright, 1994). Threat of freedom explains that persuasive messages make audiences feel restricted in their freedom to act in a certain way (Brehm, 1996; Silvia, 2006). This can in turn lead to a so called boomerang effect, which means that a receiver is motivated to change their behavior in any way that is not in line with the persuasive message (Silvia, 2006). For example, if a message tells audiences to stop smoking, they feel restricted in their freedom and as a result they might actually smoke more. Fear of change explains that audiences are by nature designed to stick to their existing ways (Fransen et al., 2015). This fear of change is often even stronger for health messages, since the changes often have a big impact. For example, if a message motivates to stop smoking while someone has been smoking for over 30 years, this is causing a major impact on that person’s life. Therefore resistance may occur as audiences try to reduce their fear for that change.
In the case of health communication there is a possible third threat involved; the health threat (Fransen et al., 2015; van Koningsbruggen & Das, 2009). When persuasive health messages are considered to be very threatening and personal, audiences feel scared that something will happen to their health. And to avoid this threat, they may to resist against the message. For example, if a message argues that smoking can cause lung cancer, audiences may perceive this as very threatening. As a result, they may start resisting to reduce this cognitive dissonance.
Consumer Resistance Strategies
A problem with the current literature on resistance strategies is that it has become hard to differentiate between the different definitions: numerous strategies have been defined, of which many show some overlap but use slightly different terminology. This paper aims to order all of these different strategies by dividing them into different categories.
Van ‘t Riet and Ruiter (2013) studied resistance strategies in the context of health
communication. Hereby, they identified four main overarching strategies within which the more specific resistance strategies are said to fit: avoidance, denial, cognitive reappraisal and suppression. The strategies from van ‘t Riet & Ruiter have some overlap with those of Fransen et al. (2015), who also identified overarching research strategies but then in the context of advertising. However, for this study the terminology from van ‘t Riet & Ruiter will be used, since this paper also focuses on health messages. All of the specific resistance strategies available in the literature will be analyzed and categorized within these overarching strategies, which results in a complete overview of all the different resistance strategies identified in the literature (see appendix A).
Denial
The first strategy defined by van ‘t Riet & Ruiter (2013) is denial, which is a process whereby one consciously processes the persuasive messages, but refuses to accept them as true. Thus, audiences actively refute a message by challenging it. The specific strategies from the literature that fit this category are: counter arguing (Jacks & Cameron, 2003), message derogation (van Koningsbruggen & Das, 2009), source derogation (Jacks & Cameron, 2003), social validation (Jacks & Cameron, 2003), optimism bias (Chambers & Windschitl, 2004) and third-person effect (Douglas & Sutton, 2004) (see table 1, appendix A).
Counter Arguing
Counter arguing (Jacks &Cameron, 2003; Jacks & Devine, 2000) is a strategy whereby consumers directly argue against the claims in a message by giving arguments that say otherwise. Take for example a persuasive video from the cancer society that urges audiences to stop smoking
because smoking can cause lung cancer. An example of counter arguing in relation to this video would be when someone says: "There is no correlation between smoking and lung cancer and therefore no reason to quit smoking."
Message Derogation
Message derogation is a strategy whereby audiences resist against a message by questioning the validity of the message. Multiple studies have studied this strategy and found it to be prevalent (Witte, 1992; van Koningsbruggen & Das, 2009; Thompson et al., 2011), especially in the context of health communication because it is highly influenced by threat perception. (van Koningsbruggen & Das, 2009). Thus, the higher the perceived personal threat, the more likely audiences are to employ message derogation. An example of message derogation is when someone argues: “The quality of this video is really bad, therefore I do not believe anything that has been stated.”
Source Derogation
Source derogation is a strategy whereby consumers directly question the expertise of the source, or claim that the message is not true because it is coming from a certain source (Jacks & Cameron, 2003). Using the same example from the cancer society video, an example of source derogation would be saying: “I never really trust these messages, why would the people of the cancer society care about our health? They only act out of personal interest.”
Social Validation
The strategy social validation is about countering a message by bringing up the opinion of important others who argue differently than the message (Jacks & Cameron, 2003). An example of social validation is saying: “if smoking is bad for you, then why do so many people smoke?”
Optimism Bias
A rather robust strategy that is especially prevalent for health related messages is optimism bias (Sharot, Korn & Dolan, 2011; Shepperd et al., 2013). It suggests that audiences have a natural tendency to believe that negative things are less likely to happen to them than to others (Chambers & Windschitl, 2004; Weinstein, 1980). As a result, they tend to downplay the risk of that
(health-related) behavior or exaggerate the perception of their own ability to control the situation (Chambers & Windschitl, 2004). An example would be to say: “While smoking may cause lung cancer, I do not think that this risk is very high for me, because it does not run in my family.”
Third-person Effect
A term related to optimism bias is the third-person effect (Duck & Mullin, 1995; Douglas & Sutton, 2004; Gunther & Mundy, 1993; Weinstein & Klein, 1995), which explains the “tendency for people to believe that socially undesirable media messages influence others more than
themselves” (Douglas & Sutton, 2004, p.3). It is very similar to unrealistic optimism, however the focus is more on the effectiveness of media messages, whereas optimism bias focuses on the likelihood of the negative effect to occur. An example of third-person effect is when someone says: “While I think that it is really good that these messages exist, I do not really feel influenced by it.”
Cognitive Reappraisal
The second category from van ‘t Riet and Ruiter (2013) is cognitive reappraisal, whereby audiences accept the message and belief what it says, but reduce cognitive dissonance by using additional beliefs that make the impact of the message smaller (van ‘t Riet & Ruiter, 2013). Strategies that fall under this category are biased assimilation (Ahluwalia, 2000) and attitude bolstering (Jacks & Cameron, 2003) (see table 2, appendix A).
Biased Assimilation
Biased assimilation is a strategy whereby consumers search for information that is consistent with their own attitude; hereby audiences accept information that is consistent and reject
information that is not (Ahluwalia, 2000). An example of this strategy is: “Smoking only causes lung cancer in very extreme cases, most of the smokers never get lung cancer.”
Attitude Bolstering
When it comes to attitude bolstering, consumers do not refute the arguments made in the message, instead they generate new arguments or thoughts that are consistent with their own attitude (Jacks & Cameron, 2003). For example: “Non-smokers can also get lung cancer.”
Suppression
Another strategy from van ‘t Riet & Ruiter (2013) is suppression; hereby audiences
acknowledge the message and its arguments, but ignore the arguments that go against their existing beliefs. It is to some extent similar to cognitive reappraisal, however the main difference is that for this strategy, audiences are focused on themselves and what they are feeling, not necessarily on the content of the message (Fransen et al., 2015). The strategies that fit this concept are asserting confidence (Jacks & Cameron, 2003), reducing impact (Ahluwalia, 2000), weighting attributes (Ahluwalia, 2000) and negative affect (Jacks & Cameron, 2003) (see table 3, appendix A).
Asserting Confidence
Asserting confidence is about not paying attention to what is being said in the message but simply asserting for oneself that nothing can change ones opinion (Jacks & Cameron, 2003). For example: “It does not matter what they say in this message, nothing can make me quit smoking.”
Reducing Impact
Reducing impact is a strategy whereby one does not let one negative attribute spillover to the other attributes; not let the persuasive message cloud ones entire opinion about the subject (Ahluwalia, 2000). For example: “Just because smoking may cause lung cancer does not mean that it is completely bad for you.”
Weighting Attributes
When audiences employ weighting attributes, they weigh attributes that are in line with their opinion stronger than attributes that are not (Ahluwalia, 2000). For example: “Smoking may cause lung cancer, but I do not care. It helps me to relax and I need that.”
Negative Affect
Negative affect is a strategy whereby consumers respond to the persuasive message by becoming angry or irritated (Jacks & Cameron, 2003; Jacks & Devine, 2000). Hereby the
arguments made in the message are being ignored; instead audiences focus on their own emotions. For example: “I get so sick from these so-called health messages.”
Avoidance
Avoidance is a strategy whereby audiences ignore messages that go against ones existing beliefs (Fransen et al., 2015; Jacks & Cameron, 2003; van ‘t Riet & Ruiter, 2013). This strategy is in fact similar to the avoidance strategy of Jacks and Cameron (2003) and therefore there are no specific strategies that fall under this category. Familiar examples of avoidance are looking at ones smartphone while watching television or leaving the room to get something to drink as soon as commercials come on (see table 4, appendix A).
Codebook
The list of resistance strategies (see appendix 1) will be turned into a codebook that will be used to analyze resistance (see appendix B). The participants in this study will be exposed to a persuasive video about binge drinking, after which they will be placed in dyads. These dyads will be analyzed using the codebook. Hereby it will be assessed whether the participants resist against the message and if so, which type of strategy or strategies they are using. However, it should be noted that resistance might not only be influenced by the message itself, but also by other factors such as previous behavior, gender and the type of frame used in the video. These factors will therefore also be analyzed in this study.
Drinking Behavior
As mentioned before, a persuasive health message can produce effects opposite than intended. This process is called psychological reactance, which entails that telling audiences not to do something may lead them to perform the opposite, in an attempt to hold on to their freedom (Engs & Hanson, 1989, Ringold, 2002). This is also known as a boomerang effect, whereby audiences move in the opposite direction of what is advocated (Knowles & Linn, 2004).
In the case of health messages, psychological reactance is even stronger for audiences that have experience with the behavior that is being advertised as unsafe or unhealthy (Ringold, 2002).
They are more resistant because nothing has happened to them thus far and therefore they do not experience the risk as such (deTurck & Goldhaber, 1991; Ringold, 2002). To illustrate, in a study from deTurck & Goldhaber (1991), they placed a ‘no diving’ warning sign next to a swimming pool. The results showed that students that had a history with diving were more likely to notice the signs, but they also perceived less danger because so far nothing has happened to them (deTurck & Goldhaber, 1991). Therefore they became more critical and resistant towards the warning sign.
The persuasive message in this study is aimed at making participants aware of the dangers of excessive alcohol abuse – binge drinking –. Based on the above-mentioned findings, it is expected that participants that have more experience with drinking will feel more invincible, because so far nothing has happened to them. This may in turn make them skeptical towards the arguments in the message. Therefore it is hypothesized that participants that consume higher amounts of alcohol will have more resistance towards the message:
H1: The higher the level of drinking behavior, the more resistance audiences employ towards the persuasive message.
Gender
Previous literature suggests that gender also influences resistance (deTurck & Goldhaber, 1991; Lippa, 2002; Woller, Buboltz & Loveland, 2007); reactance is found to be stronger for men than for women. As a result, men are inclined to resist more against health messages than women (Lipa, 2002; Woller et al., 2007). This is largely explained by the fact that men are more assertive than women (Woller et al., 2007), which makes them more inclined to keep their personal
freedoms, while women are more likely to reestablish them. Moreover, women have a higher rate of conformity than men; meaning that they easily conform to the general opinion, especially when they are in the presence of peers that are of the same gender (Lippa, 2002). And since this study will be aimed at analyzing the dyads of same-gender participants, it is expected that similar results will
be found for this study. Therefore it is hypothesized that male participants will portray more resistance towards the persuasive message than female participants:
H2: Men show more resistance towards persuasive health messages than women.
Type of Frame
The participants in this study will be exposed to a persuasive health video. These videos are all the same, except for the frame used in it. Audiences are exposed to either one of the following four frames: disgust, fear, humor or informative. Previous research suggests that the type of frame influences how audiences respond to the message (Hendriks et al., 2014), therefore it may also influence whether participants resist against a message and perhaps even which type of resistance strategy they employ. Based on this, the following sub-research question has been formulated:
RQ2: What is the relationship between the type of frame used in the persuasive message and resistance against the persuasive message?
Method
This study has two main goals: the first is to establish an overview of the existing resisting strategies in the literature – which has been done in the previous part – and the second is to study whether participants use resistance against persuasive health messages and if so, which type of strategies they employ. Analyzing dyads from participants after having been exposed to a
persuasive health message will do this. A codebook has been created based on all of the resistance strategies found in the literature (see appendix B), which will be used to analyze the dyads.
Study Outline
The participants in this study were exposed to a video clip about the dangers of binge drinking, after which they were put into pairs to have a short – unguided – dyad. These dyads were
recorded by camera - for which consent was given afterwards – and transcribed. After transcribing the dyads, they were analyzed by means of their resistance.
This study has its foundation in the grounded theory approach (Knigge & Cope, 2006; Strauss and Corbin, 1998); the aim was to develop theory throughout the experiment, based on the collected data. The results of previous resistance studies (Ahluwalia, 2000; Franzen et al., 2015; Jacks & Cameron, 2003; van ‘t Riet & Ruiter, 2013) were used as a starting point for understanding the possible types of resistance strategies. These existing strategies were turned into a codebook (see appendix B) that was used to analyze the dyads.
The codebook existed of three steps; first, the coders needed to determine whether the participants were resisting against the message. If they did, the coders needed to establish which overarching strategy was being used: avoidance, denial, suppression or cognitive reappraisal (van ‘t Riet & Ruiter, 2013). And after this had been determined, the coders needed to specify which specific strategy the participant was using (see appendix A). It was possible that the coders found a strategy that did not fit any recognized strategies, in which case a new strategy ought to be created.
Next to analyzing whether participants resisted against the message, it was also analyzed whether resistance was influenced by other variables. The variables were: gender of the participant, drinking behavior of the participant and type of frame used in the message.
Method
The data in this study was derived from a former study conducted by Hendriks, van den Putte & de Bruijn (2013), who analyzed the effect of emotions on conversational valence and alcohol consumption. They did an experiment among Dutch undergraduate students from the University of Amsterdam. The experiment took place in three different waves: all of the variables that were related to the drinking behavior of the participant were measured twice during the experiment: once before the experimental condition and once after. The participants were all
with one another. Before entering the dyad, all participants were exposed to three different video clips, with one of the three being a persuasive health message. The experimental condition was integrated into the persuasive health message in the sense that they were each set in a different emotional appeal. Participants were randomly assigned to one of the four different conditions; fear appeal, humor appeal, disgust appeal or informational appeal. Participants in the same dyad were exposed to the same frame. According to Hendriks et al. (2013), the randomization of the different conditions has been done correctly, since there were no significant differences between the
conditions in terms of age, gender, or nationality. Hendriks et al. (2013) incorporated quite some variables into their study, however these will not all be taken into the analysis, as not all of them are relevant for this study. The variables from Hendriks et al. (2013) that will be used are:
Drinking_Behavior, Gender, and Type_of_Frame. It will be analyzed whether these variables
influence if participants resist against a message.
Gender
Since previous research suggested that gender has an influence on whether participants resisted against a message, this was taken into the analysis as well. It was analyzed whether the gender of the participant (1 = male, 2 = female) influenced whether the participant resisted against the message (0 = no, 1 = yes), using a chi-square test of association. Moreover, it was also analyzed whether gender influenced how many different types of resistance strategies participants used; this was done using a one-way Anova.
Drinking Behavior
The drinking behavior of participants was measured two weeks prior to exposure to the persuasive message (=T0), it was measured by asking participants how many alcoholic drinks they had consumed for the last fourteen days. This resulted into a score for the amount of alcohol drinks that every participant had consumed, ranging from 0 to 10 (MT0=1.53, SDT0=1.80).
For this study, only the drinking behavior from before the experiment (T0),
drinking_behavior, has been taken into account during the analysis. It was analyzed if the amount
of drinking behavior influences whether participants resist against the message (0 = no, 1 = yes), this was done using a chi-square test for association. Moreover, it was also analyzed whether the level of drinking behavior influenced how many different types of resistance strategies participants used; this was done using a one-way anova.
Type of Frame
The variable type_of_frame indicates which frame was used in the persuasive health message. Two weeks after the first round of the experiment (T1), participants watched three short videos. It has deliberately been chosen to show two random videos along with the experimental video so that it would seem like a short commercial break. The videos were: a sports commercial, a dental hygiene commercial and the persuasive health message in the form of an anti-alcohol
commercial in either one of the four different frames. The four different anti-alcohol messages were designed specifically for the study and were based on the results of an extensive pilot study in which the successfulness of the videos was tested (Hendriks et al., 2013). Even though the four different videos had a different type of frame, they all had similar content and all addressed the same consequences of binge drinking (1 = fear, 2 = disgust, 3 = humor, 4 = informative).
Resistance
This paper aims to study whether participants resist against the persuasive message and if so, what type of resistance strategy they employ. This has been done by analyzing the dyads between the participants after having been exposed to the video. These dyads were coded sentence by sentence with the help of the codebook that has been developed based on previous literature (Appendix B). The dyads were content analyzed to see whether they displayed any type of
fitted this resistance or whether it was a new type of strategy. Hereby the coders needed to mark every sentence that indicated any type of resistance so that it could be used for later reference.
Two coders were trained in using the codebook; one coder analyzed 100% of the sample, while the other coder analyzed 20% of the sample in order to determine the Intercoder Reliability. Cohen’s κ was run to determine the agreement between the two coders’ judgment of the type of resistance strategies used (Landis & Koch, 1977). The results showed that there was a good
agreement between the two coders, κ = .66, p < 0.0005 (Landis & Koch, 1977). Thus, the codebook provides a good description to help coders in the coding process.
The Sample
The sample existed of 202 Dutch undergraduate students from the University of Amsterdam. These participants were selected using random sampling: students following the communication science track were invited to take part in the experiment and they then voluntarily subscribed to participate in the experiment, for which they received a monetary incentive or experiment points. The coders analyzed the dyads of all the participants (n = 202) by means of their resistance against the message. All of the participants were students at the university of Amsterdam, studying
communication science. There were 164 female participants and 38 male participants, and they were aged between 18 and 29 (M = 20.26, SD = 2.07).
Results
The aim of this paper was to study whether audiences resist against persuasive health messages and hereby provide an explorative overview of the type of strategies that they use. From the 202 participants, 155 (76.70%) resisted against the message and from this group, 97 (48%) used more than one strategy. The amount of strategies that the participants used ranged from 0 to 7, with the median being 2 (M = 1.99, SD = 2.00). The participants have used all of the four overarching
strategies; denial was used the most, followed by cognitive reappraisal, suppression and avoidance. Moreover, the participants employed most of the specific strategies (see figure 1).
Denial
The overarching resistance strategy that was used the most is denial, which was used by 72.8% of the participants. From this group, 38.6% used the strategy more than once. Message derogation was the most popular of the denial strategies, having been used 117 times, followed by optimism bias that was used 97 times, counter arguing 44 times and third-person effect 31 times (see figure 1). Source derogation and social validation were never employed
Counter Arguing
Counter arguing, arguing against a message by providing arguments that counter the claims made in the message, was used quite often (see figure 1). Most of the time, participants argued that something was not true because they never experienced it themselves, or they argued that the information in this video clip was exaggerated (see table 5).
Table 5
Examples Counter Arguing Personal
experience Countering
“I do not believe that 70% regrets having been drinking the night before, I never regret it.”
“I do not believe this, I think that this video highly exaggerates the numbers.”
44 117 97 31 60 15 2 29 14 4 6 Figure 1
Strategies used in the experiment
Message Derogation
From all of the resistance strategies combined (see figure 1), the most popular strategy was message derogation, which was used 117 times. Message derogation is defined as arguing against a message by questioning the validity of that message. Hereby the participants aimed to reduce the effect that the video had on them. To do so, they used different types of arguments; the most occurring was that participants tried to suppress the influence that this message had on them by arguing that the message was false, unoriginal or of a poor quality (see table 6).
Table 6
Examples Message Derogation Sarcasm
Not interesting Bad quality
“Oh yes, I always get extremely aggressive when I get drunk and I always start hitting people.”
“It’s not like the video is telling us anything new or exciting.” “I felt like the video was of low quality and quickly put together.”
Optimism Bias
Optimism bias was also used quite often, 97 times (see figure 1). Optimism bias refers to the tendency to downplay the risk of a negative event occurring. Many participants argued that the events in the clip were not going to happen to them because they know better or because they are too smart for it; using their status as a university student as an argument (see table 7).
Table 7
Examples Optimism Bias Boundaries
Status
“This will never happen to me, I know my boundaries when I am drinking” “I think these things are more likely to happen to lower educated people”
Third-person Affect
Third-person effect was also employed quite often. It is rather similar to optimism bias, but then aimed at the message; thus downplaying the effect that the message has on them. Participants
however they do not feel that it applies to them or they simply state that they do not feel like this message is aimed at them (see table 8).
Table 8
Examples Third-person Affect Influence
Appeal
“While I do believe that it is good that these type of videos are made, I do not really feel influenced by it”
“These type of messages never really appeal to me”
Cognitive Reappraisal
The second most used overarching strategy is cognitive reappraisal, which was used by 23.8% of the participants. Within this category, biased assimilation was used the most, 60 times, followed by attitude bolstering, which was used 15 times (see figure 1).
Biased Assimilation
Biased assimilation was used quite often (see figure 1). Biased assimilation is known as interpreting the message in such a way that it conforms to ones own opinion. Hereby arguments that are in line with this are being accepted, while arguments that are not in line are being rejected. Some participants noted that the clip only showed extreme cases of drinking behavior (see table 9).
Table 9
Examples Biased Assimilation Extremes
Non-representative
“This movie is only about idiots that step into the car while they had a drink or girls who are practically getting raped; it only shows extremes.”
“It really depends on your personality .The people in this video are probably fools that never really drink a lot and already go crazy from one drink.”
Attitude Bolstering
Attitude bolstering was used sparsely by a couple of participants (see figure 1). It is rather similar to biased assimilation, however instead of countering against some elements of the message,
most of the participants did was arguing that the things that happened in the video clip could also be blamed on things other than alcohol (see table 10).
Table 10
Examples Attitude Bolstering Replacing the
blame
“Accidents like this don’t only happen because of alcohol; the chance that you get a car accident without drinking alcohol is in fact pretty big.”
Suppression
Suppression was used almost as often as cognitive reappraisal, namely by 19.8% of the participants. Negative affect was used most often (29 times), followed by weighting attributes (14 times), reducing impact (4 times) and self assertion (2 times) (see figure 1).
Negative Affect
Negative effect, resisting against a message by becoming frustrated or irritated towards it, was used quite often, especially when it came to the disgust frame. Most of the participants
responded by stating that they felt engrossed by the message (in the case of the disgust frame) and a few of them argued that the tune in the clip annoyed them (see table 11).
Table 11
Examples Negative Affect Engrossed
Annoyed
“I was looking at this video and I thought ieuuw, that was just really gross. I really did not want to see it.”
“I heard the tune in the video and I could only think turn it off, turn it off!”
Weighting Attributes
Weighting attributes has been used by a couple of participants (see figure 1). It is a strategy whereby participants accept the arguments made in the message, but then weigh arguments that are in line with their opinion stronger than those that are not in line with their opinion. Some
Table 12
Examples Weighting Attributes Priorities
Not caring
“Of course I sometimes feel bad after drinking too much and regret some of the things I have done, but at least I had a good time.”
“It is true that these things happen, but hey living on the edge right? And at least you have a fun story to tell later.
Reducing Impact
A strategy that is rather similar to weighting attributes is reducing impact, which was used only a couple of times (see figure 1). Hereby the participant does acknowledge the information in the message, but ensures to not let this one negative attribute cloud their entire opinion. Some participants argued that drinking is not necessarily a bad thing, because not everybody drinks and drives (see table 13).
Table 13
Examples Reducing Impact
Perception “It is not like everybody who drinks causes fatal accidents, as long as you do
not drink and drive everything is fine.”
Self Assertion
A theory that was hardly ever used is self assertion, which was used only twice. When employing this strategy, participants ignore what is being said in the message and instead convince themselves of the fact that nothing can persuade them. Hereby they ignore the message and simply argue: “nothing can make me change my behavior”.
Avoidance
Avoidance has only been used 6 times, which accounted for 3% of all of the resistance strategies used (see figure 4). This strategy was only employed in combination with the disgust frame. While participants did not actively walk away from the video clip (they did have the
opportunity to stop), some of them did report that they had put their hands in front of their eyes because they were simply too engrossed by the video.
Gender
A chi-square test for association was conducted between gender and resistance to study whether gender influences if a participant resists against the message. Hereby resistance was rated as no resistance (0) or resistance (1). All expected cell frequencies were greater than five. There was a statistically significant association between gender and resistance, χ2(1) = 3.94, p = .033. This association was relatively small, φ = .14, p = .047. From the male participant 89.2% resisted against the message, while from the female participants only 73.9% resisted against the message. Therefore H1 was accepted; male participants offered slightly more resistance than female participants.
A one-way Anova was conducted to determine whether gender also influenced the amount of resistance strategies the participant employed. Hereby the amount of strategies participants used ranged from 0 to 7. The data was normally distributed for each group, as assessed by visual
inspection of Normal Q-Q Plots and there was homogeneity of variances, as assessed by Levene’s test for equality of variances (p = .873). The results of the Anova showed that there was no
significant difference between men and women F(1, 200) = 2.04, p = .155, ω2 = 0.01. Thus, gender does not influence the amount of resistance strategies used.
Drinking Behavior
Another chi-square test was conducted, this time between drinking behavior and resistance to study whether the level of previous drinking behavior influences if the participant resists against the message. Hereby resistance was rated as no resistance (0) or resistance (1) and drinking
behavior ranged from 0 to 10. All expected cell frequencies were greater than five. There was no statistically significant association between gender and resistance, χ2
(8) = 6.83, p = .56. Thus, H1 is rejected; the drinking behavior of a participant does not influence their resistance against a message.
A one-way Anova was conducted to determine whether drinking behavior influenced the amount of resistance strategies the participant employed. Hereby the amount of strategies participants used ranged from 0 to 7 and drinking behavior again from 0 to 10. The data was
normally distributed for each group, as assessed by visual inspection of Normal Q-Q Plots and there was homogeneity of variances, as assessed by Levene’s test for equality of variances (p = .958). The results of the Anova show that the amount of strategies used by the participant was not
influenced by drinking behavior F(8, 193) = .86, p = .551, ω2 = 0.03. Thus, drinking behavior does not influence the amount of resistance strategies used.
Type of Frame
As mentioned before, the participants in this study were exposed to either one of four
different frames: disgust (n = 50), fear (n = 52), humor (n = 50) or informative (n = 50). First, it was analyzed whether the type of frame influenced if participants resisted against the message, this was done by means of a chi square test for association. After this it was tested whether the type of frame also influenced the amount of resistance, which was done through an anova with a LSD Post Hoc test. Then, four different anovas were run for each overarching strategy to determine whether the type of frame also influenced the type of overarching strategy that participants used.
Resistance – Chi-square A chi-square test for association was conducted between type of frame and resistance to study whether the type of frame influences if a participant resists against the message. Hereby resistance was rated as no resistance (0) or resistance (1). All expected cell frequencies were greater than five. There was a statistically
0 0,5 1 1,5 2 2,5 3 3,5
Disgust Fear Humor Informative
Figure 2
significant association between type of frame and resistance, χ2
(3) = 21,26, p < .0005. This was a strong association, φ = .32, p < .0005. The disgust frame evoked the most resistance (94.0%), followed by the fear frame (84,6%). Both the humor (38.4%) and the informative (38.4%) frame evoked the least resistance. Therefore H1 was accepted; the type of frame significantly influences whether participants resist against a message.
Resistance - Anova
A one-way ANOVA was conducted to determine whether the type of frame influenced the amount of resistance strategies that the participants employ. Hereby the amount of strategies participants used ranged from 0 to 7, and participants were exposed to either one of four frames: disgust (n = 50), fear (n = 52), humor (n = 50) and informative (n = 50). There were no outliers in the data, as assessed by inspection of a boxplot. The data was normally distributed for each group, as assessed by visual inspection of Normal Q-Q Plots. There was homogeneity of variances, as assessed by Levene's test of homogeneity of variances (p = .193). The amount of resistance was significantly influenced by the different type of frames, F (3, 198) = 31.83 p < .0005, ω2 = 0.16. The disgust frame evoked the most resistance (M = 3.12, SD = 1.76), followed by fear (M = 1.96, SD = 1.62), informative (M = 1.38, SD = 1.32) and humor (M = .58, SD = .50), in that order (see figure 2). Tukey HSD post hoc analysis (see table 14) revealed that the disgust frame evoked significantly more resistance than fear (1.16, 95% CI [.34, 1.98], p = .002), informative (1.64, 95% CI [.91, 2.57], p < .0005) and humor (1.74, 95% CI [.81, 2.47], p < .0005). No other group differences were statistically significant.
Table 14
Multi Comparisons Table
Mean Difference Std. Error Sig.
Disgust Fear 1.16 .32 .002
Informative 1.64 .32 < .0005 Humor 1.74 .32 < .0005
Resistance Strategy
Now that it has been assessed that the type of frame influences whether participants resist against a message, it should also be determined whether it influences the type of resistance strategy that they use. Participants were again exposed to either one of four frames: disgust (n = 50), fear (n = 52), humor (n = 50) and informative (n = 50). There were some outliers, as assessed by boxplot, however it has been determined that these are expected since some strategies are used more often than others. The data was normally distributed for each group, as assessed by visual inspection of Normal Q-Q Plots. However, the results show that there was multicollinearity amongst the strategies; therefore it has been decided to run a separate Anova for each of the four overarching strategy instead of one combining Manova.
Denial
A one-way ANOVA was conducted to determine if the strategy denial was used differently per frame. The amount of times the participants employed the denial strategy ranged from 0 to 6 (M = 1.43, SD = 1.33), There was homogeneity of variances, as assessed by Levene's test of
homogeneity of variances (p = .201). The use of the strategy denial was significantly different for the different type of frames, F(3, 198) = 2.69, p = .048, ω2
= 0.04. The denial strategy was used the most for disgust (n = 50, M = 1.80, SD = 1.34), then fear (n = 52, M = 1.56, SD = 1.47), and the least for informative (n = 50, M = 1.20, SD = 1.34), and humor (n = 50, M = 1.16, SD = 1.08), in that order. LSD post hoc analysis (see table 15) revealed that denial was used significantly more for disgust than for humor (.64, 95% CI [.12, 1.16], p = .016), and informative (.60, 95% CI [.08, 1.12], p = .024). There were no other significant differences between the items.
Table 15
Multi Comparisons Table
Mean Difference Std. Error Sig.
Disgust Humor .64 .26 .016
Cognitive Reappraisal
Another one-way ANOVA was conducted to determine if the strategy cognitive reappraisal was used differently per frame. The amount of times the participants employed the cognitive reappraisal strategy ranged from 0 to 2 (M = .27, SD = .52). There was no homogeneity of variances, as assessed by Levene's test of homogeneity of variances (p = < .0005). The use of the strategy cognitive reappraisal was significantly different for the different type of frames, Welch’s F(3, 198) = 6.05, p < .0005, ω2 = 0.10. The cognitive reappraisal strategy was used the most for disgust (n = 50, M = .54, SD = .68), then fear (n = 52, M = .23, SD = .47), and the least for informative (n = 50, M = .22, SD = .47), and humor (n = 50, M = .10, SD = .30), in that order. Games-Howell post hoc analysis (see table 16) revealed that cognitive reappraisal was used significantly more for disgust than for fear (.31, 95% CI [.01, .61], p = .044), humor (.44, 95% CI [.16, ,72], p < .0005), and informative (.32, 95% CI [.02, .62], p = .035). There were no other significant differences between the items.
Table 16
Multi Comparisons Table
Mean Difference Std. Error Sig.
Disgust Fear .31 .16 .044
Humor .44 .10 < .0005 Informative .32 .17 .035
Suppression
Another one-way ANOVA was conducted to determine if the strategy suppression was used differently per frame. The amount of times the participants employed the suppression strategy ranged from 0 to 1 (M = .30, SD = .17). There was no homogeneity of variances, as assessed by Levene's test of homogeneity of variances (p = < .0005). The use of the strategy suppression was significantly different for the different type of frames, Welch’s F(3, 198) = 8.37, p < .0005, ω2
(n = 52, M = .17, SD = .43), and the least for informative (n = 50, M = .06, SD = .24), and humor (n = 50, M = .12, SD = .06), in that order. Games-Howell post hoc analysis (see table 17) revealed that suppression was used significantly more for disgust than for fear (.49, 95% CI [.14, .83], p = .002), humor (.54, 95% CI [.20, ,88], p < .0005), and informative (.60, 95% CI [.28, .92], p < .0005). There were no other significant differences between the items.
Table 17
Multi Comparisons Table
Mean Difference Std. Error Sig.
Disgust Fear .49 .13 .002
Humor .54 .13 < .0005 Informative .60 .12 < .0005
Avoidance
The amount of times the participants employed the avoidance strategy ranged from 0 to 3 (M = .25, SD = .56). From the analysis it can be seen that avoidance has been used only in combination with the disgust frame. There is thus no need to conduct a one-way ANOVA, since it is not being employed for any of the other frames.
Discussion
The aim of this study was partly exploratory and partly experimental. The first exploratory part was aimed at creating a list of existing resistance strategies from previous literature and analyze which of these strategies audiences are employing when resisting against persuasive health
messages. Hereby, this study aimed to make a connection between the various resistance strategies available in the literature and provide a codebook that researchers can adopt to study resistance.
The second experimental part was aimed at analyzing which variables influence resistance. Participants in this study were all exposed to a persuasive video about binge drinking. These videos
were all the same, except for the frame used in the video; they were exposed to either one of the following four frames: humor, disgust, fear or informational. It was analyzed whether the type of frame influenced the resistance of participants against the message. Moreover, it was examined if gender and previous drinking behavior influenced whether participants resisted against the message.
Strategies used by the participants
As mentioned before, Van ‘t Riet and Ruiter (2013) identified four overarching resistance strategies: denial, suppression, cognitive reappraisal and avoidance. From the results it could be seen that participants employed all of the four overarching strategies. The most used strategy was denial, followed by cognitive reappraisal, suppression and avoidance, which was used only a couple of times and only in combination with the disgust frame.
Use of the Denial Strategy
Denial was by far the most employed strategy; many participants resisted against the message by claiming that the statements made were not true, or only true for some people but not for them. This strategy was used the most for the disgust and fear frame and the least for the informative and humor frame.
A possible explanation for the fact that denial was employed this often is that the sample existed solely out of university students; perhaps they are better capable of arguing against the message than lower-educated audiences. A possible interesting future study would be to do a similar study, but then among audiences of different education levels to analyze if education influences whether audiences resist against the message and which strategy they are employing.
Counter Arguing
From the denial strategies, counter arguing was employed quite often. It is defined as
resisting against a message by providing counter arguments. This strategy has also been found to be one of the most prevalent used strategies in the studies from Jacks and Cameron (2003) and Jacks
and Devine (2000), who studied commercial messages. Thus, counter arguing is a rather robust strategy that is employed for both commercial and health messages.
Message Derogation
The most popular from all strategies combined was message derogation, which is defined as arguing against a message by questioning the validity of that message. These findings are in line with previous research of van Koningsbruggen and Das (2009), who also found that participants employed message derogation a lot. However, it is opposed to the results from Jacks and Cameron (2003), who concluded that participants did not employ message derogation.
A possible explanation for this difference in results is that Jacks and Cameron studied commercial messages, while van Koningsbruggen and Das argue that message derogation is used mainly for health messages, because it is highly influenced by the level of perceived fear. Thus, perhaps message derogation is a strategy that is only being used in the context of health
communication, however this ought to be investigated further in order to make significant claims. Perhaps by conducting a study whereby one condition is a persuasive commercial message and one a persuasive health message and then analyze the differences between the employed strategies.
Moreover, it should be noted that the persuasive clips used in this experiment have been made purely for the sake of the study. While this provides the opportunity to manipulate the different frames, it is also problematic in the sense that the quality of the clips is not as good as those of professional clips. Thus, participants may have used message derogation this often as a result of the somewhat poorer quality of the clip; perhaps message derogation is not as often employed for professional video clips. This provides an interesting angle for future research; for example conducting a similar study but then analyzing professional video clips.
Optimism Bias and Third-person Effect
Optimism bias refers to the tendency to downplay the risk of a negative event occurring. Third-person effect is rather similar, but then aimed at the message; thus downplaying the effect that the message has. Both optimism bias and third-person effect were used quite regularly, which
is in line with previous research, since both of these strategies are often encountered for health messages (Chambers & Windschitl, 2004; Douglas & Sutton, 2004).
Source Derogation and Social Validation
From all of the strategies discussed in this study, only two were never used by any of the participants, namely source derogation and social validation. This is contrary to findings of Jacks and Cameron (2003), who found that participants did employ these two strategies.
Source derogation is defined as diminishing the effect of a video by questioning the expertise of the source. It is not very surprising that this strategy has not been found in this study, since it was not clear for the participant who the source of the video was. Therefore it was not possible for them to argue that the source did not have any expertise about the topic. The reason that social validation, arguing against a message by bringing up the opinion of important others that argue against the message, was not used is perhaps because of the topic. Many of the participants are probably aware of the dangers of drinking alcohol and therefore it is hard to bring up an argument of someone who argues differently. Perhaps this strategy will be found statistically significant in a similar study with a different topic. Thus, these two strategies should not be written off simply because the participants in this study did not employ any of these two strategies. Instead, more similar research should be conducted with different topics and different types of messages in order to study whether it is employed in other situations and conditions.
Use of the Cognitive Reappraisal Strategy
Cognitive reappraisal was the second most used strategy; hereby participants did pay attention to the message, but then gave their own interpretation or added extra information in order to change the message in such a way that it favors ones own opinion. This strategy was used the most in relation with the disgust frame.
Biased Assimilation
Biased assimilation is defined as countering a message by accepting the information that is in there, but then arguing that it is not very likely to happen to you. Overall, this strategy was
employed quite often, which is in line with the results from Ahluwalia (2000), who argues that this strategy has been found to be quite robust.
Attitude Bolstering
Attitude bolstering is defined as generating thoughts that are consistent with one’s own attitude without directly arguing against the claims made in the message. As noted by Jacks and Cameron (2003), there are quite some similarities between this strategy and counter arguing. The main difference lies in the fact that for counter arguing participants specifically counter against a claim made in the message, while for attitude bolstering the arguments are accepted, but twisted in such a way that it is in line with ones own opinion. An example of attitude bolstering would be saying that the car accidents could have been attributed to other things than alcohol as well, while counter arguing would be saying that these car accidents were not caused by alcohol. Jacks and Cameron (2003) found that participants rated the perceived likelihood of using attitude bolstering higher than the likelihood of using counter arguing, however when it came to their actual behavior, they were in fact more likely to use counter arguing. The results confirm these findings, as the participants in this study were more likely to employ counter arguing than attitude bolstering.
Use of the Suppression Strategy
The third-most used strategy is suppression; ignoring the things being said in the message and instead become irritated towards it or claim that you cannot be influenced under any
circumstance. This strategy was again used the most in relation with the disgust frame.
Negative Affect
The strategy negative affect, resisting against a message by becoming angry or irritated towards it, was used quite often. Most of the times it was in combination with the disgust frame, which made participants grossed out. In the study of Jacks and Cameron (2003), participants also responded by getting upset or irritated about the content of the message. However, that did not happen in this study; participants only employed negative affect when they were engrossed by to the disgust frame or annoyed by the tune. This may again be contributed to the fact that Jacks and
Cameron (2003) used a commercial message; most audiences are exposed to a vast amount of commercials and this can become annoying at some point, which may cause audiences to resist by using the negative affect strategy. And perhaps health messages are considered to be less annoying. However, another study comparing commercial messages and health message should be conducted in order to make significant claims about this.
Weighting Attributes and Reducing Impact
Weighting attributes is a strategy whereby audiences weigh the arguments that are in line with their opinion stronger than those that are not. For example, some participants argued that they did not care about the risks of drinking because they believed that it was more important to have fun. Reducing impact is quite similar; hereby the participant does not let one negative attribute spillover to the others. For example arguing that while binge drinking may cause traffic accidents, this does not mean that it is completely bad for you.
Both these strategies were used only sporadically, which is contrary to the results of
Ahluwalia (2000), who found that both these strategies were employed quite often. This difference may be due to the fact that the participants in this study were exposed to a health message. Most participants probably do care about their health, therefore it would be contradicting for them to use a strategy whereby they claim differently. It may be the case that these two strategies are used more for commercial messages, however this should be investigated further in order to make significant claims. For example by conducting a similar study for commercial messages and test whether participants in this condition make more use of weighting attributes and/or reducing impact.
Self Assertion
Self assertion is defined as denying the arguments made in a message by ensuring oneself that one cannot be persuaded by anything, this strategy was used very sparsely; only twice. This is in line with the research from Jacks and Cameron (2003); they also resulted that this strategy was only used a couple of times.
A problem with this strategy is that it is very similar to optimism bias and third-person effect, which makes it hard to distinguish this strategy. The main differentiation lies in the fact that participants that use the self assertion strategy are focused on themselves and on the fact that they cannot be influenced by anything, whereas the other two strategies are focused on the message and the arguments made by it. Moreover, since it has found to be hardly used in both this study and the study by Jacks and Cameron, maybe a future study should investigate whether audiences actually employ this strategy or whether the terminology should be adjusted.
Use of the Avoidance Strategy
Avoidance, ignoring the message altogether, was used the least of all the overarching strategies. It was used only a couple of times and only in combination with the disgust frame; this was the only time that participants specifically mentioned that they looked away from the message because they could not watch it any more. These findings are contrary to the study of Jacks and Cameron (2003), who found that avoidance was employed quite often.
A possible explanation for this and something that should be noted is that avoidance could not be measured very well in this study. In reality audiences are likely to walk away during a commercial break or turn to their smartphone whenever they loose interest. While the participants in this study did have the opportunity to exit the experiment, this step was a lot bigger than when they would have been watching the same clip in their own room. Moreover, this study only analyzed whether respondents said that they did not pay attention, while they may not have been paying attention to the message without saying so. Thus, in reality avoidance may have been used more. Therefore it is interesting for future researchers to study avoidance more in depth, for example by doing an eye tracking study to see what participants are focusing on.
Gender differences
This study also analyzed whether other factors also influenced resistance. The results showed that gender significantly influences whether participants resist against a message. Hereby,
male participants are more likely to resist the message than female participants. However, there were no significant gender differences in the amount of strategies employed; men and women used an equal amount of resistance strategies.
The finding that men use more resistance than women is in line with previous literature (deTurck & Goldhaber, 1991; Lippa, 2002; Woller, Buboltz & Loveland, 2007). Lippa (2002) argues that this is due to the fact that men are more assertive than women and therefore more
inclined to resist against persuasive messages. Another possible explanation is that women are more inclined to conform to the standard than men (deTurck & Godlhaber, 1991). And since the
participants in this study where put into dyads after being exposed to the video, there is a big chance that this might have influenced the results as well. Perhaps female participants were more inclined to conform to the opinion of the other. This would be an interesting angle of research for future researchers; to study to what extent conformity influence whether audiences resist against a message and the type of strategies they employ.
Drinking Behavior
Another argument for the gender difference could be that men tend to drink more alcohol than women (Hendriks et al., 2013) and therefore feel more personally appealed by the message, which in turn makes them more inclined to resist against the message (deTurck & Goldhaber, 1991). However, contrary to the findings in previous literature (deTurck & Goldhaber, 1991), the results of this research showed that drinking behavior did not significantly influence whether participants resisted against the message.
A possible explanation for this, and a limitation of this study, is that participants needed to self-report their drinking behavior of the past two weeks (Hendriks et al., 2013). This time frame is rather long for participants to remember exactly how much they have been drinking; therefore the results may lack in precision and prove to be somewhat unreliable. Moreover, participants in this study were first told what was considered to be binge drinking (more than 3 drinks for women,
more than 5 drinks for men) and were then asked to report their own drinking behavior. Through this, a standard is being set for participants, which may influence how they report their own
drinking behavior and thereby lead to socially desirable answers. These two factors combined may have led to incorrect data for drinking behavior and therefore produce false results. Perhaps it would have been better to either limit the time frame so that it is easier for participants to remember and/or tell the binge drinking quota after they have filled in how much they have been drinking.
Type of Frame
The type of frame being used in the message significantly influenced whether participants resisted against the message. Moreover, it also influenced the amount of resistance strategies that participants used. As seen from the results, the disgust frame evoked the most resistance, followed by the fear frame, the informative frame and the humor frame. This was found to be rather
consistent over the different overarching strategies, as all of the different overarching strategies were used the most during the disgust frame.
According to Hendriks et al. (2013), the emotion fear induced the most conversations with a negative valence. And since resistance often results from a negative attitude towards the message, it would be expected that the frame that induces the most fear also evokes the most resistance.
Hendriks et al. (2013) argue that the disgust frame induces the most fear compared to the other three frames. And in line with this, the results of this study showed that the disgust frame evoked significantly more resistance than any of the other frames.
The Codebook
As mentioned before, the end goal of this study was to provide a clear and comprehensible codebook that could be used by future researchers to analyze resistance. The first codebook that was used to analyze the data in this study (see appendix B) has been adjusted somewhat based on the analysis and the data. The layout considering the coding of the variables has been adjusted as
well, since the analysis proved that this would have been more convenient. The result is a clear codebook containing the most prevalent resistance strategies, including a clear description of what the strategies entail exactly (see appendix C).
This codebook can serve as a guide for future researchers that study resistance; hereby the aim should be to keep the research exploratory. The strategies in this codebook are not set in stone; it may be the case that there are some strategies missing or not used in certain situations, as has been elaborated on previously. Moreover, it should be kept in mind that one of the main limitations of this study – and resistance studies in general –, is that there is room for a lot of interpretation. This codebook tries to reduce this by providing very clear descriptions of the strategies and clear distinctions between them, however proper training of coders remains very important.
Conclusion
Through an in-depth analysis of audience’s responses to a persuasive health video, this study aimed to examine resistance towards persuasion. The results showed that a relatively large
percentage of the participants resisted against the message. The most popular overarching strategy was denial; resisting against the message by directly countering it. From all of the strategies combined the most used strategy was message derogation; arguing against the message by questioning the validity of the message. The results further showed that gender and the type of frame used in the message had a significant influence on resistance.
The results of this study are a great addition to the recent study of Fransen et al. (2015), who studied tactics to counter resistance against persuasion. They argue that resistance-neutralizing tactics are more effective whenever they are tailored to the specific resistance strategy that the audience is using. Therefore it could be a great set up for companies that aim to tackle resistance to first use the codebook of this study to determine the type of resistance that their audiences use, and then use the results of Fransen et al. (2015) to determine how they should counter this resistance.
Reference List
Ahluwalia, R. (2000). Examination of Psychological Processes Underlying Resistance to Persuasion. Journal of Consumer Research, 27(2), 217-232.
Brehm, J. W. (1966). A theory of psychological reactance. New York.
Chambers, J.R. & Windschitl, P.D. (2004). Bias in Social Comparative Judgements: The Role of Nonmotivated Factors in Above-Average and Comparative-Optimism Effects. Psychological
Bulletin, 130(5), 813-838.
DeTurck, M.A. & Goldhaber, G.M. (1991). A developmental analysis of warning signs: The case of familiarity and gender. Journal of Products Liability, 13(55), 65-78.
Douglas, K.M. & Sutton, R.M. (2003). Right about others, wrong about ourselves? Actual and perceived self-other differences in resistance to persuasion. British Journal of Social
Psychology, 43(4), 585-603.
Duck, J.M. & Mullin, B. (1995). The perceived impact of the mass media: reconsidering the third person effect. European Journal of Social Psychology. 25(1), 77-93.
Engs, R. & Hanson, D.J. (1989). Reactance Theory: A Test with Collegiate Drinking.
Psychological Report, 64(3), 1083-1086.
Fransen, M.L., Verlegh, P.W.J., Kirmani, A. & Smit, E.G. (2015). A typology of consumer strategies for resisting advertising, and a review of mechanisms for countering them.
International Journal of Advertising, 34(1), 6-16.
Friestad, M., and P. Wright. 1994. The persuasion knowledge model: How people cope with persuasion attempts. Journal of Consumer Research, 21(1), 1-31.
Gunther, A.C. & Mundy, P. (1993). Biased Optimism and the Third-Person Effect. Journalism &
Mass Communication, 70(1), 58-67.
Helm, A. (2004). Cynics and skeptics: consumer dispositional trust. Advances in Consumer
