The role of message framing and health vs. ecological
claims in promoting meat substitutes
Master thesis 2015/2016
Master Track: Persuasive Communication Igor Tschernizki, 10110208
Thesis supervisor: Kim Brandes Date: 24th of June 2016
1 Abstract
The world’s population has roughly doubled over the last forty years. This has had a huge impact on our environmental footprint: food production has had to increase to sustain this growth. This unsustainable food production has to be tackled, and one of the most effective ways is by decreasing the consumption of meat. Furthermore, recent studies have shown that consumption of red and processed meat increases the chance of getting diseases like stomach cancer and type-2 diabetes. It is clear that the consumption of meat is not a vital necessity, because all the necessary nutrients found in meat can be obtained from plant-based products. One of options is to replace meat with meat substitutes, which are prominently plant-based. In order to find the most effective message which could persuade people to buy meat substitutes instead of real meat, this study conducts an experiment with messaging types in promotion of a meat substitute product, testing the effects of the type of message (health vs. ecological) and message framing (positive or negative, i.e. a loss-frame vs. gain-frame) on consumer attitudes towards and the willingness and intent to buy a meat replacement product. The experiment showed no significant effect of the type of claim towards ad attitude, brand attitude, and purchase-intention. The interaction effects of framing and the type of claim were also not significant. Analysis of framing showed a significant effect on consumer attitude towards the ad, but not towards the brand or purchase-intention.
2 Index
1. Introduction 3
2. Theoretical Background 6
2.1 Claims in context of the EPPM model 6
2.2 Framing 9
2.3 Interaction between the claim and type of frame 11
3. Method 12
3.1 Experimental design 12
3.2 Pre-test 13
3.2.1 Results of pre-test 14
3.3 Stimulus material 15
3.4 Procedure and sample 15
3.5 Variables 16 3.5.1 Independent variables 16 3.5.2 Dependent variables 17 3.6 Analyses 18 4. Results 19 4.1 Respondents 19 4.2 Randomization check 19 4.3 Manipulation check 20
4.4 Analyses of the effects of the claim type 20
4.5 Analyses of the effects of message framing 21
4.6 Analyses of interaction effects between the claim type and framing 22
5. Conclusion and discussion 23
5.1 Conclusion 23
5.2 Research limitations 24
5.3 Implications and future research 24
References 26
Appendix A: Pre-test 31
Appendix B: Experiment Qualtrics 37
3 1. Introduction
The world population has grown incredibly quickly – roughly doubling over the last forty years. This has had a huge impact on our environmental footprint: food production has had to increase to sustain this growth, and now accounts for up to 30% of the Earth’s land use, but also for at least 30% of the greenhouse gas emissions (Garnett, 2011). This development is hardly sustainable, and may prove disastrous: people may therefore have to change their dietary intake in order to stop or even reverse the damage.
As convincingly argued by much scientific research, the most effective way to accomplish this would be to cut our consumption of meat (Hallström, Carlsson-Kanyama & Börjesson, 2015). Driven not only by population growth but also by economic development – meat consumption per capita rose 25% (Henchion et al., 2014) – global meat consumption has grown almost 60% between 1990 and 2009. At least 80% of Dutch citizens consume meat three times or more per week (Hoek, Luning, Weijzen, Engels, Kok & de Graaf, 2011). This exacerbates the environmental impact of human development in two important ways: first of all, the production of meat uses a fast amount of our earth’s resources, like land, water, and food; furthermore, the animals themselves produce waste which is one the leading causes of climate change (Marlow et al., 2009). This meat consumption pattern poses a huge threat to the Earth’s climate (EC, 2006). When looking at climate change for example, greenhouse gas emissions from agriculture can be brought back to and stabilized at 2005 levels (a target that all international climate change policies have accepted) if people would consume meat twice a week, with a maximum of 80 grams per serving, of which only half may be red meat (McMichael, Powles, Butler & Uauy, 2007).
This increase in meat consumption does not only have a negative impact on our earth’s environment; it is also detrimental to our health. In the last few decades, multiple studies have shown that eating red and processed meats can be dangerous. Consumption of red and processed meat has been linked to obesity, cardiovascular disease, different forms of cancer, and Type-2 diabetes (Boada, Henríquez-Hernandez & Luzardo, 2016; Satija et al., 2016). This cost to the environment and our health seems incredibly wasteful: consumption of meat is not a vital necessity, because all the necessary nutrients found in meat can be obtained from plant-based products (Davies & Lightowler, 1998). Even if meat would be the only source available for the intake of nutrients like protein, the consumption of this product is far in excess of what is needed.
4 One possible solution to the waste and the drain on the Earth’s resources due to meat
production is to replace this meat consumption with meat substitutes. Not only does the production of these plant-based products require fewer resources, but it also results in less harmful waste. Meat substitutes for instance use the vegetable soy as their main ingredient. To produce 1 gram of soy, which contains the necessary proteins that our body needs, there is less water and land needed than to produce 1 gram of meat (Pimentel & Pimentel, 2003). Furthermore, changing to a vegetarian diet or prudent “in-between” diets that allow only a small quantity of red meats has proven to be effective in reducing the risk of the diseases mentioned above (McEvoy, Temple & Woodside, 2012).Although meat substitutes have been available for a while now, at the moment they still can be considered a niche product. A study conducted on the purchase of meat substitute products in the Benelux has shown that only 36% of households purchase it once every few weeks (GfK Panel Services Benelux, 2009). This number may seem considerable, but when taken into consideration that meat is used on a daily basis, we can state that the consumption of meat substitutes has to obtain a more prominent place in people’s diets. Unfortunately, people seem to be reluctant to consume meat substitutes instead of real meat, and in countries like China or Brazil the consumption of meat has been even on the rise (Clonan, Roberts & Holdsworth, 2016).So what would it take to increase the consumption of this product? The product category has so far received little attention in the academic research field. Because of the immense
importance of triggering someone to actually replace his or her intake of meat products by meat substitutes, this thesis will try to make a first attempt at investigating the best way to promote meat substitutes.
This study focuses on the drivers for consumer meat substitute use, and will try to identify ways in which these products can be promoted most effectively in order to increase their consumption. The promotion will happen in the form of an advertisement, which will attempt to stimulate a positive attitude from the respondent towards the ad itself and the product’s brand. Also, the purchase intention will be measured, which one would expect to increase if the consumer’s attitude is positive. The ad will be tested on a number of axes. As mentioned above, industrial-scale meat production and consumptions poses two threats: one to the planet (climate change) and one to our health – which entails not only a larger intake of calories than needed, but also a higher risk of developing various forms of cancer (Larsson, & Orsini, 2014; McEvoy, Temple & Woodside, 2012). Therefore, two different claims will be used to persuade the consumer to purchase the product, one with respect to health one with respect to ecological welfare. The Extended Parallel Process Model (EPPM) will be used to
5 evaluate these claims (Witte & Allen, 2000). Furthermore, the claims will be framed either as a loss-frame or a gain-frame. This form of framing of a message has shown to have different effects on changing undesired behavior. Whereas gain-frames emphasize the benefits of choosing a safer option, the loss-frame focusses on the costs of failing to take action
(Rothman, Bartels, Wlaschin & Salovey, 2006). Therefore, even though the claims will have the same outcome, the expectations should be different.
The experiment set-up is therefore as follows: consumers are exposed to an advertisement that contains either a message with a health claim or an ecological welfare claim, with both claims also framed as a loss-frame and a gain-frame. After exposure, purchase intention is measured in order to examine the willingness of the participants to buy the product. Furthermore, it is important to see how the messages influence attitude towards the brand vis-à-vis those towards the ad, because these attitudes could have an effect on the purchase intention. In this regard the study will aim to answer the following central question:
“What are the effects of a health benefit claim or an environmental claim in an advertisement for a meat substitute product when presented in a gain- versus loss-frame message, on brand attitude, ad attitude, and the consumer’s intent to purchase?”
6 2. Theoretical Background
2.1 The claims in context of the EPPM model
An examination of the two (kinds of) claims – with respect to health and to ecological welfare – reveals that the motives behind them differ several ways. First, the claims activate different emotions and appeal to different needs. Secondly, they evoke fear, because the messages contain information about diseases and ecological disasters. In order to predict which claims would have a greater effect on the consumer’s attitude towards the
advertisement, toward the brand, and purchase intention, the Extended Parallel Process Model (EPPM) can be used. When this model has been used successfully, it has been proven to be effective for promoting attitudinal, intentional, and/or behavioural change (Lewis, Watson & White, 2013). When we look at the claims from the perspective of the Extended Parallel Process Model (EPPM) there are a few things that can be observed.
The EPPM states that there are two different cognitive appraisals that a person gets when confronted with a message that is perceived as threatening (Witte, 1992). The message in question must not only contain elements of threat (dangers of performing an undesired behaviour or the dangers of not performing an undesired behaviour) but also of efficacy (the strategies to overcome or prevent the threat). The recipient of the message first makes an estimation of the threat appraisal: in other words, does the person perceive the message as threatening? To do so, the person evaluates the severity of the threat: “how severe or serious would the consequences of the undesired behaviour be?” Furthermore the person’s
susceptibility to the threat is also judged, which means that the person looks at how big the risk is that the he or she could actually be affected by the threat stated in the message (Witte, 1992).
If both parts are strong enough and the recipient experiences enough fear he moves on to the second appraisal. This appraisal is about the efficacy of the recipient towards the message and consists of two parts: self-efficacy and response efficacy. The self-efficacy can be explained as the recipient’s perceived ability to actually perform the desired behaviour that is stated in the message. The second part of the appraisal, the response efficacy, indicates whether the recipient perceives the strategies given in the message are actually useful and effective towards reducing or diminishing the threat. When there is high self-efficacy and high response efficacy the recipient initiates danger control. This entails that the message is accepted and they are likely to perform the desired behaviour. On the other hand, if either self-efficacy or response efficacy (or both) are low, then the recipient initiates fear control. In
7 this case the message is not accepted and the person won’t perform the desired behaviour but will engage in defensive processing. If this occurs there is a chance that the recipient will perform the undesired behaviour even more (Witte, 1992).
The EPPM model can be applied to both claims. These two claims address serious health and environmental topics, and contain the necessary elements to use the EPPM framework. Using the model enables us to explain the persuasive process and to predict which of these two claims would have a greater effect on the consumer’s attitudes towards the advertisement, towards the brand, and their intent to purchase. When we look at the claims from the perspective of the EPPM, there are a few things that can be observed.
To apply this model to the two different claims we first have to take a close look at the information that will be given in the advertisement messages. The claims contain either a message directed at health or at ecological welfare. The messages will entail information that drives people to obtain a meat substitute product. There are numerous reasons why a person should consume less meat. First of all, by eating meat substitutes, a person lowers its risk of type-2 diabetes, high blood pressure, high cholesterol, diverticular disease and certain types of cancer: colorectal, ovarian and breast cancers. In addition, a vegetarian diet may increase someone’s life expectancy up to three or four years (Singh, Sabate & Fraser, 2003). In the case of ecological welfare, the claim contains information that relates to environmental issues, like the excessive and dangerous amount of waste that the livestock population is producing, and the rapid expansion of land that is not only used for the livestock itself, but also to accommodate the animal’s feed supply (Hoek et al., 2011).
Health claims have proven to be effective in different circumstances. A study conducted on the usage of health claims for food products showed that consumers preferred to use products with health claims, and that it helps people to make a better food choice (Khurshid, Ahmad & Saeed, 2013). The message above contains two obvious threats: the chance of getting diseases like type-2 diabetes, and stomach cancer. These illnesses would not only be perceived as severe, but as diseases which could affect anyone and everyone. Previous research has shown that the threat appraisal consisting of the perceived severity and
susceptibility of the threat stated in the message would be perceived by the recipient of the
message as high in the case of a disease like cancer (Stephenson & Witte, 1998). When considering the efficacy appraisal, it can be stated most people cutting down on their meat consumption and instead buying meat substitutes would seem like a doable task.
8 instead of real meat, in this case beef, and you will decrease your chance of getting severe diseases like stomach cancer and type-2 diabetes.”
The second type of message that will be used in the experiment contains a claim about ecological welfare. Most people nowadays know the harm that we are doing to our planet and the effects of climate change are noticeable all around the world. In the UK for instance, only 1% of the people have not heard about climate change, global warming, or the greenhouse effect, and see the harm it is doing to our planet as a serious topic (Lorenzoni, Nicholson-Cole & Whitmarsch, 2007; Lorenzoni & Pidgeon, 2006). This entails that both the severity and the susceptibility of climate change (the threat) in the threat appraisal will be judged to be high and the recipients will move on to the second appraisal. The self-efficacy will be the same: the behaviour is the same as with the other type of claim. On the other hand, the
efficacy appraisal of the ecological welfare message is likely to be judged differently from
the health claim message. The problem lies in the fact that people will likely not perceive the strategy given in the message as actually useful in diminishing or reducing the threat. A study done on the barriers to action showed that one of the main barriers was that the respondents felt powerless: they did not have the feeling that they alone could do something, and felt that the act itself was meaningless. Another issue was a distrust in the government and their capacity to hold their side of the bargain on climate change mitigation (Stoll-Kleemann, O’Riordan & Jaeger, 2001). Furthermore, by presenting people with a disastrous future, their fear could take over. In the case that the message does not contain a working strategy to prevent this, as perceived by the recipient of the message, there is a risk that it will trigger defensive avoidance and reactance (Good & Abraham, 2007; O’Keefe & Jensen, 2008). This denial of the message could then have a negative effect on the behaviour intention, in this case the consumption of meat substitutes instead of real meat (O’Neill & Nicholson-Cole, 2009).
It can be assumed that the ecological welfare claim message will be perceived as having a lower response efficacy than the health claim message. Therefore, the health message is likely to work better, because the recipient who is exposed to this message is less likely to activate his or her defence mechanism. The next hypotheses can be made:
H1a: A health claim in an advertisement for a meat substitute product has a more positive
effect on the ad attitude of the consumer than an environmental claim.
H1b: A health claim in an advertisement for a meat substitute product has a more positive
9 H1c: A health claim in an advertisement for a meat substitute product has a more positive
effect on the purchase-intention of the consumer than an environmental claim.
H1d: A health claim in an advertisement for a meat substitute product is perceived to contain a
greater response efficacy by the consumer than an environmental claim.
2.2 Framing
Advertisements often make use of a promotional message to persuade people to buy a product. Framing a message has proven to be an effective strategy to persuade people into acting a certain way. The framing theory states that different ways of framing the same information can have an impact on the decision making of the recipient. The base for this thought lies in the prospect theory that talks about the fact that people’s desires and
preferences can vary depending on how for example a message is put together (Tversky, & Kahneman, 1981). The thought behind this is that – according to Tversky & Kahneman (1981) – people do not always behave in a consistent way and make rational decisions. Whenever someone is presented with a ‘decision problem’, the ‘decision maker’ will respond differently to the problem depending on the ‘decision frame’. This frame is explained as the respondents’ “conception of the acts, outcomes, and contingencies associated with a
particular choice” (Tversky & Kahneman, 1981). When the ‘decision maker’ is presented with a problem he or she adopts a specific frame. The way the problem is formulated and the habits, norms, and characteristics of the ‘decision maker’ all influence how the frame is composed. The formulation of the problem can be manipulated, because most of the time it is possible to formulate a problem in more than one way. Because humans are not rational creatures, changes of perspective can “reverse the relative apparent size of objects and the relative desirability of options (Tversky & Kahneman, 1981). And so, although the outcome of the problem would be the same, different formulations could create different expectations with the recipient. Rothman et al. (2006) have built on this framing theory and developed a new approach for the development of health messages. This theoretical approach is used to make as effective a message as possible that would encourage people to perform a more healthy behavior. This approach focusses on two different ways in which a health related message can be framed: in term of benefits (gains) or costs (losses). When framing a message with an emphasis on the benefits, the approach is called a “gain-frame”; when the focus lies on the costs of not acting the way the message is framed, it is known as a “loss-frame” (Rothman & Salovey, 1997).
10 Gain-frame appeals used in health related messages put the emphasis on the benefits of taking action. This approach is most effective and therefore used in situations when people have the idea that their behavior involves a relatively low risk of an unpleasant outcome. In this case, the behaviour that someone has to perform according to the message is risk-averse, because they have to take action to avoid risk of an unpleasant outcome (Rothman et al., 2006). The gain-frame is therefore used to persuade someone into performing certain behaviour to prevent these ‘unpleasant outcomes’. These gain-frame appeals have proven to be most effective in prevention behaviours, for example promoting the use of sun-screen. When people use sun-screen, they are less likely to get sunburned, and thereby decrease their chance of getting skin cancer in the long run. Firstly the use of sun-screen emphasizes the prevention of a short-term risk, and then the long term risk (skin cancer). The use of sun-screen is in this case the ‘safer option’ while not using it can be seen as a risky behaviour, because it increases the chance of getting ill (Rothman et al., 2006).
When a health message is aimed at persuading people into a behaviour that is ‘risk-involved’, a loss-frame should me most effective. The loss-frame puts the emphasis on the costs of not taking any action, and not the benefits of acting, as in the case of the gain-frame. In circumstances where the outcome would be perceived as negative, this frame would be more suited into motivating people to take action. Rothman et al. (2006) stated that this approach would be suited best for the detection of a health problem, because this involves a risky decision.
In this study, the health claim is aimed at persuading people to use the product due to the benefits the meat substitutes have over real meat with regard to a person’s health, or the disadvantages of not using the meat substitutes and using real meat instead. The benefits can for instance be that meat substitutes contain fewer harmful elements than real meat, and decrease the chance of getting diseases like type-2 diabetes and stomach cancer (Hoek et al., 2011). The message that contains an ecological welfare claim aims to persuade people that purchasing and consuming meat substitutes is far less harmful to the ecological health of the planet, which can be seen as the gain-frame. The message could also emphasize that not consuming meat substitutes and instead keep buying real meat, has negative consequences for the ecological welfare of the planet, which can be seen as a loss-frame. The ecological
welfare of Earth is directly linked to the future survival of the human race and the way people will be able to live on this planet. Due to the fact that buying meat substitutes can be seen as a preventative act with respect to negative consequences concerning health and ecological welfare, it can be stated that a gain-frame would have more positive effects. Furthermore, a
11 gain-frame is more positive and therefore would create more positive associations.
Consequently, this would have a more positive effect on the attitude toward the ad, brand attitude, and purchase intention of the people. Therefore the next hypotheses can be made:
H2a: A gain-frame message in an advertisement for a meat substitute product would have a
greater effect on the ad attitude of consumers than a loss-frame message.
H2b: A gain-frame message in an advertisement for a meat substitute product would have a
greater effect on the brand attitude of consumers than a loss-frame message.
H2c: A gain-frame message in an advertisement for a meat substitute product would have a
greater effect on the purchase-intention of consumers than a loss-frame message.
2.3 Interaction between the claim and type of frame
By consuming meat substitutes instead of real meat, a person can decrease the chance of getting diseases like type-2 diabetes and different forms of cancer. The consumption of meat substitutes can be seen as a preventive behaviour and therefore a gain-frame message about the product, versus a loss-frame, would be more persuasive according to Rothmann and Salovey (1997). A gain-frame describes the benefits of performing a healthy behaviour that is risk-averse.
When looking at the claim about ecological welfare, previous research on climate change mitigation has found that gain-frames have a stronger effect on desired behaviour that is pro-environment, in this case the pro-environmental behaviour being the consumption of meat substitutes instead of real meat (Spence & Pidgeon, 2010). Other research has shown that presenting with a message about climate change in a positive frame (gain-frame), increases intention to act on the desired behaviour vis-à-vis the same message in a negative frame (loss-frame). Thus a gain-frame should have a stronger effect on persuading the respondents to take action than the loss-frame (Morton, Rabinovich, Marshall & Bretschneider, 2011; Scannel & Gifford, 2011).
If this interaction of the frames is combined with framing, it is fair to assume that the health claim would have a greater effect than the ecological welfare claim. According to the study done by Covey (2010), perceived severity and susceptibility did not have an effect on the outcomes. On the other hand, increasing response efficacy and/or self-efficacy could have an effect on the outcomes, and be the deciding factor in the question whether a certain frame would or would not work (Covey, 2010). Due to the fact that a message containing a health
12 claim would be perceived as having more response efficacy, it could be hypothesised that this type of claim would have a greater effect on ad attitude, brand attitude, and
purchase-intention. Furthermore, the message that is framed in a more positive manner (gain-frame) would have greater effect than the message that is framed more negatively (loss-frame). As stated earlier, due to the preventive nature of the behaviour and the fact that a loss-frame is a more negative message, the gain-loss-frame should have a more positive effect on ad attitude, brand attitude and also on purchase-intention for both of the claims. In combination with the perceived higher response efficacy of the messages that contain the health claim the next hypotheses can be stated:
H3a: A health claim in an advertisement for a meat substitute product has a more positive
effect on the consumer’s attitude towards the ad than an environmental claim, and this effect will be more pronounced for the gain-frame messages.
H3b: A health claim in an advertisement for a meat substitute product has a more positive
effect on the brand attitude of the consumers than an environmental claim, and this effect will be more pronounced for the gain-frame messages.
H3c: A health claim in an advertisement for a meat substitute product has a more positive
effect on the purchase-intention of the consumers than an environmental claim, and this effect will be more pronounced for the gain-frame messages.
13 3. Method
3.1 Experimental design
For this study, an experiment has been conducted using a 2 x 2 between-subject factorial design. The experiment consists of two experimental factors: framing and claims (see Table 1). Both these factors consist of two levels: the independent variable for framing is whether it is a loss-frame or a gain-frame, while the second independent variable represents the specific claim, i.e. health vs. ecological welfare. Thus, the between-subject design had four different groups, and the respondents were randomly assigned to these four conditions. The
respondents in the first and third condition received a message with either a health or ecological welfare claim, both constructed in a gain-frame. The respondents in the second and fourth condition received a message with either a health or an ecological claim, both constructed in a loss-frame.
Claim Frame
Health claim Ecological welfare claim
Gain-frame Condition 1 (n = 31) Condition 3 (n = 32)
Loss-frame Condition 2 (n = 29) Condition 4 (n = 31)
Table 1. Experimental design (N = 123) 3.2. Pre-test
Before starting the experiment, a pre-test was conducted, where participants had to evaluate the four messages (see Appendix A) which would be used in the actual experiment. In total, 20 respondents filled out the pre-test survey, twelve of whom were female (60%) and eight male (40%). The average age of the respondents was 25 years (range 18-39). In terms of their level of education, twelve respondents (60%) had completed education on a University level, while three (15%) respondents completed higher vocational education. The five other
respondents completed a Secondary general education, Secondary vocational education, higher general education, Pre-University education, or other.
The purpose of this pre-test was to examine whether the four messages would be interpreted the way they were intended to. Each message contained either a health or
ecological welfare claim, and was constructed in a gain- or loss-frame. The participants first received questions to evaluate the framing of the specific message followed by questions
14 about the two claims. The framing manipulation was pretested by three questions (Lanting, 2010):
- The first question concerning if the tone of the message was negative or positive, which was measured with a 7-point Likert scale (1 = Negative to 7 = Positive).
- The second questions asked if the message they were exposed to put on the benefits of buying ‘meat substitutes’ or the disadvantages of buying ‘real beef’, which could also be answered with a 7-point Likert scale (1 = Disadvantages of ‘real beef’ to 7 = Advantages of ‘meat substitutes’).
- The third and final question was asked to see if the respondents perceived the message to put the emphasis on the consequences of risky behaviour or the impact of safe behaviour, which as before could be answered with a 7-point Likert scale (1 = Consequences of risky behaviour to 7 = Impact of safe behaviour).
The scores of these three questions were added up to measure the overall evaluation of the message framing. A total score average of 1 would mean that the participants perceived the construct of the message as a loss-frame. Conversely, a total score average of 7 would imply that the construction of the message was perceived as a gain-frame.
Furthermore, the respondents had to answer two questions about claims, which consisted of a statement that could be answered with a 7-point Likert scale (1 = Totally disagree to 7 = Totally agree). The first statement indicated that the message contained a claim about a person’s health; the second indicated the opposite, namely that the message contained a claim about ecological welfare. Asking these two questions allowed to check whether the respondents indeed interpreted the claims accordingly.
3.2.1 Results of pre-test
The results of the pre-test had to be analysed with a Paired-Samples T-Test. The respondents from the pre-test perceived the messages containing a gain-frame significantly (p < .001) more positively (M = 5.20, SD = .76), than the message containing a loss-frame (M = 2.00,
SD = 1.12). These results showed that framing of the messages was perceived as intended,
and they were usable for the main experiment. Furthermore, the pre-test was designed to test the type of claim; the pre-test respondents thought that the health message indeed contained a health claim (M = 5.68, SD = 1.10), and not an ecological welfare claim (M = 2.03, SD = 1.53), and the difference was significant (p <.001). The respondents also thought that the
15 ecological welfare message contained an ecological welfare claim (M = 6.55, SD = .48), and not a health claim (M = 2.30, SD = 1.25). This difference was also significant (p < .001). The results from the pre-test for the type of claim showed that the type of claim in the message was perceived as it was intended.
3.3 Stimulus material
Four advertisements (see Appendix C) were used in the experiment as stimulus materials, which were visually identical except for the added message (manipulation) underneath the image of the product itself. To keep the material homogenous in all aspects except for the message, it was necessary to keep the rest of the advertisements elements identical. The advertisement consisted of an existing image and a message that was developed specifically for this experiment. To make sure that the ecological validity of the experiment was
maintained, the stimulus material was made of not only an existing meat substitute brand, but also of a ‘real’ visual advertisement. The brand used in the advertisements,‘Beyond Meat’, is a fairly new brand that is not sold in The Netherlands. An unfamiliar brand was used to minimize the chance of any pre-existing attitudes the respondent could have had towards the product.
3.4 Procedure and sample
The experiment was executed in the months of May and June of 2016, and the data collection for the pre-test and the main experiment took approximately two weeks Respondents for the first part of the experiment, the pre-test, were obtained via two different methods. A portion of the collected participants received a private message through a social medium (specifically Facebook) with the request to answer a small digital survey (pre-test) about meat substitute advertisements. The other participants were acquired by going to the University of
Amsterdam and handing out the survey to students. Before the respondents had to fill out the survey, they received background information about the experiment and how much time filling out the survey would approximately take. The only criterion that was used for the pre-test sample was that the participants had to be 18 years or older.
The survey of the main experiment (see Appendix B) was developed with Qualtrics, online survey/experiment software. The advertisement message and the survey itself were designed in English, which ensured the maximum number of respondents. Due to the fact that most of the participants were highly educated, language was not a problem for the
16 questions. In the case of the main experiment, just as with the pre-test, people were informed that they had to be at least 18 years old to participate. All of the respondents were either approached through a private message, in which they were informed about the experiment and asked to participate, or through a general post on Facebook containing some general information about the experiment. All of the participants were given the following
information: filling out the survey would take approximately 10 minutes, the purpose/topic of the experiment, and that there would be a cash prize of €25,- for two of the participants selected at random. The survey itself also contained an introduction which stated that entering the competition for the cash prize was not mandatory, and that their answers would be
detached from their email address to guarantee their anonymity. The participants had to submit a consent form that entailed the conditions set up by the Ethics Committee. After this, participants were presented with a couple of questions about general demographic
characteristics, such as gender, age, education, and their dietary lifestyles. This was followed by the actual advertisement (stimulus), where each respondent was randomized into one of the four conditions and received a message containing one of the following: gain-frame health, loss-frame health, gain-frame ecological welfare, or loss-frame ecological welfare. The advertisement had a time lock of 15 seconds. The advertisement was followed by three sets of questions about ad attitude, brand attitude, and purchase intention. The survey
continued with questions regarding the EPPM model and the manipulation check. At the end of the survey, the respondents could decide if they wanted to participate in the contest to win the €25,- prize. If the answer was “yes”, then they were redirected to another form, where they could fill in their name and email address. In this way, anonymity of the answers they had given was ensured. Later on, two of the respondents who were interested in the contest were selected at random and received the €25,- euro cash prize.
3.5 Variables
3.5.1 Independent variables Claims
The first independent variable was the specific claim used in the advertisements’ message. Two of the advertisements used in the experiment contained a health claim, while the other two contained an ecological welfare claim. These specific claims were chosen because they are the prominent reasons for someone to consume ‘meat substitutes’ instead of ‘real meat’.
17
Framing
The second independent variable was the frame in which the advertisements’ message was constructed. The first form in which the message was framed was the gain-frame, which emphasised the benefits of buying ‘meat substitutes’ and the impact of this safe behaviour. The second perspective was the loss-frame, which emphasised the disadvantages of ‘real beef’ and the consequences of risky behaviour.
3.5.2 Dependent variables
In total, three dependent variables were measured in the experiment: attitude towards the advertisement, brand attitude, and the purchase-intention.
Attitude towards the ad
To measure the respondents’ attitude towards the advertisement, a number of questions were asked after the respondents had been exposed to the stimulus material. The questions used a 7-point semantic differential scale consisting of four items: very bad/very good, not at all likeable/likeable, unfavourable/favourable, and unappealing/appealing (Macinnis & Park, 1991). The results from the exploratory factor analyses show that the scale explains 79.62% of the variance and that the scale is unidimensional. The reliability analysis shows a
Cronbach’s Alpha of .91 (M = 15.40, SD = 5.14).
Brand Attitude
The second dependent variable measures the respondent’s attitude towards the brand that has been used in the advertisement. To achieve this, a number of questions were asked after the respondents had been exposed to the stimulus material. The questions will use a 7-point semantic differential scale consisting of four items: very bad/very good, not at all
likable/likable, unfavourable/favourable, unappealing/appealing (Macinnis & Park, 1991). The results show that this scale explains 90.50% of the variance and that the scale is
unidimensional. The reliability analysis on this scale show a Cronbach’s Alpha of .96 (M = 16.07, SD = 5.79).
Purchase-intention
The final dependent variable measures the respondent’s intention to purchase the product after being exposed to the stimulus material (Yoo & Donthu, 2001). The participants were presented with two statements concerning the product, which then had to be judged with the help of a 7-point Likert scale (1 = Totally disagree to 7 = Totally agree). The two statements
18 were: ‘I would like to buy a product from Beyond Meat’, and ‘I intend to purchase a product from Beyond Meat’. The results of the exploratory analyses show that the scale explains 94.82% of the variance and that here too the scale was unidimensional. The reliability analysis conducted shows a Cronbach’s Alpha of .95 (M = 8.59, SD = 3.43).
3.6 Analyses
First, a descriptive analysis was conducted using the general information questions to describe the sample’s demographic characteristics. The second step was conducting a randomization check for the following variables: age, education, and dietary lifestyle. To achieve this randomization check, a Chi-Square analysis has been carried out.
Furthermore, a manipulation check has been conducted to see if the respondents interpreted the manipulations as they were intended. Both the framing of the message and the claim that the message entailed had to be checked. In order to check the manipulation of the type of claim, two one-way ANOVA analyses were conducted. First, the manipulation of the message containing a health claim was tested. A one-way ANOVA was carried out with the type of claim as the independent variable and the statement that the message contained a health claim as the dependent variable. Secondly, the manipulation of the message containing an ecological welfare claim was tested. A one-way ANOVA was carried out with the type of claim as the independent variable and the statement that the message contained an ecological welfare claim as the dependent variable. The second manipulation that had to be tested was the framing of the message. Did the respondents perceive the message to be framed in a gain- or a loss frame? In order to find that out a one-way ANOVA had to be conducted, with the type of frame as the independent variable and the manipulation check of the framing as the dependent variable.
After conducting the necessary steps described above the main analyses could be carried out. All the hypotheses were tested separately, each using a two-way ANOVA analysis.
19 4. Results
4.1 Respondents
The initial sample consisted of 189 respondents, 138 (73%) of whom filled out the survey completely. Of these 138 respondents, 10 had to be excluded because they had answered the question about their dietary lifestyle as being vegetarian or vegan. Of the remaining 128 respondents, a further 5 respondents indicated that they either knew or may have known the brand Beyond Meat. These five respondents have also been excluded from the main analyses to prevent any biased answers. After excluding these two sets of respondents, the final sample size was 123.
An additional check was conducted to exclude outliers to see whether some respondents scored very differently from the rest of the sample. In order to do this, the Z-scores of every dependent variable were assessed. For the dependent variables that consisted of more than one item, the Mean of these items was used to find the outliers. An analysis of the Z-scores revealed that none of the 123 respondents had a score above 2, and that therefore none of the respondents had to be excluded from the experiment. The final sample of the experiment ended up at a total of N = 123.
Of the 123 respondents in the final sample, 69 (56.1%) were male and 54 (43.9%) were female. The average age of the participants was 27 years old, ranging between 19 and 55. In terms of their level of education, the sample consisted for the main part out of respondents who completed University, 63 (51.2%) and 36 (29.3%) respondents who completed Higher vocational education.
4.2 Randomization check
In order to check if the randomization was successful, the following independent variables were tested: gender, education, and dietary lifestyle. The results from the Chi-Square
analyses showed that there was no difference between the conditions in gender distribution X2 (3, N = 123) = .58, p = .902. This was also the case for the distribution of educational level, with X2 (24, N = 123) = 25.55, p = .326. The last control variable, brand familiarity, also showed no difference in the conditions regarding the distribution, with X2 (3, N = 123) = 1.94,
p = .586. The randomization check was therefore successful for all of the control variables, so
20 4.3 Manipulation check
Claim
A manipulation check was carried out to test if the respondents interpreted the manipulations as intended. The first manipulation to be tested was the sort of claim that the message
contained: health or ecological welfare. The respondents who had received a health claim message had M = 4.78, and the group with an ecological welfare claim message M = 3.03. The variance analyses show that the conditions concerning the claim differed significantly, F (1, 121) = 38.79, p < .001. Secondly, the manipulation of the message containing an
ecological welfare claim was checked. The respondents who had received a health claim message had M = 2.33, and the group with an ecological welfare claim message M = 5.46, this difference was statistically significant, F (1, 121) = 140.40, p < ,001). From these analyses we can conclude that the manipulation of framing was successful.
Frame
The group that was exposed to a gain-frame message had M = 3.97, while the group that received the loss-frame message had M = 2.43. The results of the analysis over variance show that the difference between the two conditions (groups) was significant with F (1, 121) = 41.23, p < .001. These results indicate that the manipulation was successful and the two groups interpreted the framing of their message the way in which it was intended.
4.4 Analyses of the effects of the claim type
The first three hypotheses stated that the advertisements containing a health claim would have a positive effect on the ad attitude, brand attitude, and a greater purchase intention vis-à-vis advertisements containing an environmental claim.
- The first hypothesis (H1a) predicted that the claim type would have an effect on people’s
attitude toward the ad. The results of the first two-way ANOVA analysis did not establish a significant effect, F (1, 119) = .50, p = .479.
- The second hypothesis (H1b) stated that the type of claim would have an effect on people’s
brand attitude. Once again, the results show that there is no significant effect with F (1, 119) = .57, p = .453.
- The third hypothesis (H1c) talked about the effects of the claim type on purchase intention.
As before, the analysis results show that type of claim had no significant effect on the purchase intention with F (1, 119) = .03, p = .857 either.
21 - The fourth and last hypothesis (H1d) stated that the response-efficacy for the health claim
would be perceived as greater than for the message that contained ecological welfare claim. As before, the results do not show a significant effect with F (1, 121) = .31, p = .581.
Therefore, all the hypotheses about the direct effects of the claims have been rejected. The type of claim does not appear to lead to a more positive attitude towards the brand, attitude towards the brand, or a higher purchase intention.
4.5 Analysis of the effects of message framing
The next set of three hypotheses stated that respondents who received an advertisement that contained a message constructed in a gain-frame would have a more positive effect on their attitude towards ad attitude, brand attitude, and a greater purchase intention than the
respondents who received a loss-frame message.
- The first hypothesis (H2a) stated that framing would have an effect on ad attitude. The
analysis results show a statistically significant effect, F (1, 119) = 5.68, p = .019. The effect of the gain-frame message was higher (M = 4.11, SD = 1.33), than the loss-frame message (M = 3.58, SD = 1.19), with
indicating a weak effect size.- The second hypothesis (H2b) that message framing had an effect on the respondent’s brand
attitude. The results indicate that the type of frame has no significant effect for brand attitude,
F (1, 119) = 2.50, p = .117.
- The last hypothesis (H2c) stated the effects of framing on the respondent’s intention to
purchase the product. As with the previous one, the analysis results show that the frame type has no significant effect on the variable of purchase intention, F (1, 119) = 1.26, p = .264.
One the results indicated a significant effect, namely the effect of message framing on ad attitude. The other two hypotheses had to be rejected. The type of frame did only lead to a higher attitude towards the ad.
4.6 Analysis of the effects of the interaction between the claim type and framing The third and final set of hypotheses was designed to test the effects of the interaction between the type of claim and message framing.
22 - The first hypothesis (H3a) was designed to test the interaction effects of the type of claim
and the framing on people’s attitude towards the advertisement. The results do not show a significant effect, F (1, 119) = 1.82, p = .180.
- The second hypothesis (H3b) stated that there would be interaction effects of the type of
claim and message framing on people’s attitude towards the brand. Likewise, no significant effect is observed, F (1, 119) = .90, p = .344.
- The final hypothesis (H3c) predicted that there would be interaction effect of the type of
claim and message framing on people’s intention to purchase the product. The analysis shows no significant effect on the dependent variable purchase intention, F (1, 119) = .11, p = .738.
None of the above results show a significant effect, and all three hypotheses have to be rejected. This means that the interaction between the type of claim and framing (gain- versus loss-frame) did not have a significant effect on all of the three dependent variables, i.e. ad attitude, brand attitude, and purchase intention.
23 5. Conclusion and discussion
5.1 Conclusion
This study attempted to answer the following question: What are the effects of a health
benefit claim or an environmental claim in an advertisement for a meat substitute product when presented in a gain- versus loss-frame message, on ad attitude, brand attitude, and people’s intent to purchase?
The first main effect examined, the type of claim (health or ecological welfare)
showed no significant effect on the respondent’s attitude towards the ad, or purchase-intention. It was expected at the outset that a message with a health claim would have had a more positive effect than a message with an ecological welfare claim. This was expected due to the fact that a health claim would have had a higher perceived response efficacy, and therefore more positive effect on the respondent’s acceptance of the message. Mitigation of ecological welfare had proven to be complicated, because people would often feel helpless when presented with this issue. Therefore, the expectation was that the respondents who were presented with an ecological welfare message would perceive the threat appraisal as high, but would not have the feeling that switching from meat to meat substitutes would be an effective strategy to minimize or diminish this threat. Testing the perceived response efficacy of the four messages however did not show a significant effect. This means that this assumption had proven to be untrue and the response efficacy of the messages (health vs. environmental welfare) was not perceived as different by the respondents. Consequently, it then also makes sense that there is no significant effect of the type of claim on ad attitude, brand attitude, and people’s intent to purchase.
The second main effect examined was the effect of message framing (gain-frame versus loss-frame) on ad attitude, brand attitude, and purchase-intention. The results showed that there was no significant effect of framing on the respondent’s attitude towards the brand or purchase-intention. A significant (but small) effect of framing on ad attitude was found, indicating that that the gain-frame did indeed lead to higher attitude towards the ad. As expected the gain-frame works better when the desired behaviour is of a preventive nature (Rothman, 2006).
Finally, this study also looked into the interaction effects of framing and the type of claim on the respondent’s ad attitude, brand attitude, and purchase-intention. The analysis of the survey results shows no significant interaction effects. This could have been expected,
24 because none but one of the main effects were found to be significant. This was also due to the fact that the response efficacy of the different claims had no significant effect.
5.2 Research limitations
The first limitation to this research was the way the experiment was conducted. Because the experiment was done with the help of Qualtrics, most people participated in a real-life environment: their home, at work, at school, and so forth. Although this was beneficial for the ecological validity of the experiment, it also had its drawbacks. There were limitations to controlling the conditions of the experiment, and people for example could participate in the experiment while doing some other task. On the other hand, even the conditions in which the respondents were exposed to the advertisement were not completely real-life. People had to participate in the experiment on their computer, and were exposed to advertisement through this medium. An advertisement in another setting, like a magazine or a billboard, would have been more natural. Furthermore, the respondents were only exposed to the advertisement once. In most cases, a product will be shown on more than one medium, and for a period of time, increasing the chance that people would be exposed more than once to the
advertisement and would take it in fully. This could have been one of the reasons why there was only on significant effect found. For example, the mere-exposure effect states that if people would see the advertisement more than once it would increase their attitude towards the ad (Bornstein & Robert, 1989).
Another limitation of this experiment was the fact that there was no pre-test done on people’s attitudes towards the meat substitutes. The respondents could have had pre-existing ideas and feelings towards this product category, which could have had an effect on not only the attitude towards the ad and brand, but also purchase-intention. If people already
consumed meat substitutes on a regular basis they would be more open and positive towards the advertisement. The only way to control this was to exclude people with a vegetarian or a vegan lifestyle, but this way the issue that the respondents could have been biased was only partly resolved and may have biased the results in the other direction – after all, people most likely to be purchasing meat substitutes were purposely excluded from the sample.
5.3 Implications and future research
Meat consumption has proven to be a major issue for the ecological welfare of this planet and our health. Finding a solution to this growing problem is a necessity, and switching from consumption of meat to meat substitutes is one of them. The category of meat substitutes
25 needs more attention and people need to be persuaded to purchase it. This research made an attempt to see what the most effective way would be to promote it.
The different benefits for health or ecological welfare that meat substitutes have over real meat did not have a significant effect on the respondents’ attitude towards the ad, attitude towards the brand, or intention to purchase the product. Although these claims were expected to have different effects, the experiment showed otherwise. Furthermore, due to the size of the experiment, not all benefits could be taken into account. Animal welfare for examples is also an important factor that people seem to consider when cutting their consumption of meat (Hoek et al., 2011).
The strategy of framing has proven to be an effective way of persuading people to take action and this was tested in the context of meat substitutes. Because the consumption of meat substitutes is an act that could prevent an unpleasant outcome, it was expected that the gain-frame would have had more effect. This assumption was not backed up by the
experiment for the intent to purchase or the attitude towards the brand, but the analyses showed that framing of the message had an effect on consumer attitude towards the advertisement. It might therefore still be wise to take the framing of the advertisement’s message into consideration.
Future research could try to make usage of longitudinal measurements to see if the attitudes change over time, and if the advertisements actually succeed in persuading people to buy the product. Furthermore, this study only exposed the respondents to the advertisement once, and it is important to see if the effects of these messages would be significant if the experiment use more than one exposure. Finally, it would be interesting to test the effect of various message mediums such as video or audio, various designs for the ad image, etc.
26 References
Boada, L.D., Henríquez-Hernandez, L.A., & Luzardo, O.P. (2016). The impact of red and processed meat consumption on cancer and other health outcomes: Epidemiological evidences. Food and Chemical Taxicology, 92, 236-244.
Bornstein, & Robert F. (1989). "Exposure and affect: Overview and meta-analysis of research, 1968-1987". Psychological Bulletin, 106(2), 265-289. doi:10.1037/0033 2909.106.2.265.
Clonan, A., Roberts, K.E., & Holdsworth, M., (2016). Socioeconomic and demographic drivers of red and processed meat consumption: implications for health and environmental sustainability. Proceedings of the Nutrition Society, 1-7. doi:10.1017/S0029665116000100
Covey, J. (2014). The role of dispositional factors in moderating message framing effects. Health Psychology, 33, 52-65. doi:10.1037/a0029305
Davies, J., & Lightowler, H. (1998). Plant-based alternatives to meat. Nutrition & Food
Science, 2, 90–94.
European Commission (EC), 2006. Environmental Impact of Products (EIPRO): Analysis of the Life Cycle Environmental Impacts Related to the Total Final Consumption of the EU25. European Commission Technical Report. Retrieved from http://ec.europa.eu/- environment/ipp/pdf/eipro_summary.pdf
Garnett, T. (2011). Where are the best opportunities for reducing greenhouse gas emissions in the food system (including the food chain)? Food Policy, 36, 23-32. doi:10.1016/j. foodpol.2010.10.010.
GfK Panel Services Benelux. (2009). Inzichten en marktontwikkelingen in de Benelux. Dongen, The Netherlands: GfK Panel Services Benelux bv. GfK Jaargids, 1.
27 meta-analysis of measures. Health Psychological Review, 1(2), 208–229. doi:
10.1080/17437190802280889.
Hallstrom, E., Carlsson-Kanyama, A., & Borjesson, P. (2015). Environmental impact of dietary change: a systematic review. Journal of Cleaner Production, 91, 1-11.
Hoek, A. C., Luning, P. A., Weijzen, P., Engels, W., Kok, F. J., & de Graaf, C. (2011). Replacement of meat by meat substitutes. A survey on person- and productrelated factors in consumer acceptance. Appetite, 56, 662–673.
Khurshid, N., Ahmad, W., & Saeed, R. (2013). Usage of Food Health Claims and Related Consumer Understanding. Scientific Papers, 13(3), 117-120.
Lanting, M. (2010). Preventie en detectie van huidkanker: ‘Het effect van emotie en framing in voorlichtingsboodschappen’ (Master’s thesis, University of Amsterdam,
Amsterdam, The Netherlands). Retrieved from http://www.scriptiesonline. uba.uva.nl/document/193193
Larsson, S.C., & Orsini, N. (2014). Red meat and processed meat consumption and all-cause mortality: a meta-analysis. American Journal of Epidemiology, 179(3), 282–289.
Lewis, I., Watson, .B, & White, K.M. (2013). Extending the explanatory utility of the EPPM beyond ear-based persuasion. Health Communication, 28, 84-98. doi:10.1080/ 10410236.2013.743430
Lorenzoni, I., Nicholson-Cole, S., & Whitmarsh, L. (2007). Barriers perceived to engaging with climate change among the UK public and their policy implications. Global Environmental Change, 17, 445-459.
Lorenzoni, I., & Pidgeon, N. (2006). Public views on climate change: European and USA perspectives. Climatic Change, 77(1-2), 73–79.
MacInnis, Deborah, J., & Park, W.C. (1991). The differential role of characteristics of music on high- and low-involvement consumers’ processing of ads. Journal of
28 Marlow, H.J., Hayes, W.K., Soret, S., Carter, R.L., Schwab, E.R., & Sabate, J.(2009). Diet and the environment: does what you eat matter? American Journal of Clinical
Nutrition 89, 1699-1703. http://dx.doi.org/10.3945/ajcn.2009.26736Z
McAfee, A.J., McSorley, E.M., Cuskelly, G.J., Moss, B.W., Wallace, J.M., Bonham, M.P., & Fearon, A.M. (2010). Red meat consumption: an overview of the risks and benefits. Meat Science, 84(1), 1–13.
McEvoy CT, Temple N, & Woodside J.V. (2012). Vegetarian diets, low-meat diets and health: a review. Public Health Nutrition, 15, 2287-2294.
McMichael, A.J., Powles, J.W., Butler, C.D., & Uauy, R. (2007). Food, livestock production, energy, climate change, and health. The Lancet, 370 (9594), 1253-1263.
Morton, T.A., Rabinovich, A., Marshall, D., & Bretschneider, P. (2011). The future that may (or may not) come: how framing changes responses to uncertainty in climate change communication. Global Environmental Change, 2, 103–109.
O’Keefe, D. J., & Jensen, J. D. (2008). Do loss-framed persuasive messages engender greater message processing than do gain-framed messages? A meta-analytic review.
Communication Studies, 59, 51–67.
O’Neill, S., & Nicholson-Cole, S. (2009). ‘Fear won’t do it’: promoting positive engagement with climate change through visual and iconic representations. Science
Communication, 30(3), 355–379.
Pimentel, D., & Pimentel, M. (2003). Sustainability of meat-based and plant-based diets and the environment. American Journal of Clinical Nutrition, 78 (3), 660-663.
Rothman, A. J., Bartels, R. D., Wlaschin, J. & Salovey, P. (2006). The strategic use of gain- and loss-framed messages to promote healthy behavior: How theory can inform practice. Journal of Communication, 56, S202-S220.
Rothman, A. J., & Salovey, P. (1997). Shaping perceptions to motivate healthy behavior: The role of message framing. Psychological Bulletin, 121, 3–19.
29 Satija, A., Bhupathiraju, S.N., Rimm, E.B., Spiegelman, D., Chiuve, S.E., & Borgi, L. (2016) Plant-Based Dietary Patterns and Incidence of Type 2 Diabetes in US Men and Women: Results from Three Prospective Cohort Studies. PLOS Medicine, 13(6), 1- 18. doi:10.1371/journal.pmed.1002039
Scannell, L., & Gifford, R. (2011). Personally Relevant Climate Change: The Role of Place Attachment and Local Versus Global Message Framing in Engagement. Environment
and Behavior, 1-26. doi:10.1177/0013916511421196
Singh, P.N., Sabate, J., & Fraser, G.E. (2003). Does low meat consumption increase life expectancy in humans? American Journal of Clinical Nutrition, 78(3), 526-532.
Spence, A., & Pidgeon, N. (2010). Framing and communicating climate change: The effects of distance and outcome frame manipulations. Global Environmental Change,20. 656-667.
Stephenson, M.T., & Witte, K. (1998): Fear, threat, and perceptions of efficacy from frightening skin cancer messages. Public Health Review, 26, 147-174.
Stoll-Kleemannl, S., O’Riordan, T., & Jaeger, C.C. (2001). The psychology of denial concerning climate mitigation measures: evidence from Swiss focus groups. Global
Environmental Change, 11, 107-117.
Tversky, A., & Kahneman, D. (1981). The framing of decisions and the rationality of choice. Science, 221, 453-458.
Witte, K. (1992). Putting the fear back into fear appeals: The extended parallel process model. Communication Monographs, 59, 329-349.
Witte, K. & Allen, M. (2000). A meta-analysis of fear appeals: Implications for effective public health campaigns. Health Education & Behavior, 27(5), 591-615.
Witte, K., Meyer, G., & Martell, D.P. (2001). Effective Health Risk Messages: A Step-By- Step Guide. Thousand Oaks, CA: Sage Publications.
30 Yoo, B., & Donthu, N. (2001). Developing and validating a multidimensional consumer- based brand equity scale. Journal of Business Research, 52(1), 1–14.
31 Appendix A: Pre-test
Pre-test
Introduction
At the moment I’m studying a Master Persuasive Communication at the University of Amsterdam. For my thesis I will conduct an experiment on advertising meat substitutes. In the next few pages you will come across four advertising messages for meat substitutes, with each a few questions that you will need to answer. I would be very grateful if you could fill out this survey for me. It will take about five minutes to complete, and the data will be processed anonymously.
Thank you in advance,
Igor Tschernizki
Consent form
I agree, fully and voluntarily, to participate in this research study. With this, I retain the right to withdraw my consent, without having to give a reason for doing so. I am aware that I may halt my participation in the experiment at any time.
If my research results are used in scientific publications or are made public in another way, this will be done such a way that my anonymity is completely safeguarded. My personal data will not be passed on to third parties without my express permission.
If I wish to receive more information about the research, either now or in future, I can contact Igor Tschernizki (igortschernizki@gmail.com).
32 General questions
1. What is your gender?
A) Male B) Female
2. What is your age?
3. What is your highest completed education?
A) Did not follow or finish any education B) Primary school (Basisonderwijs) C) Lower vocational education (LBO) D) Pre-vocational education (VMBO) E) Secondary general education (MAVO) F) Secondary vocational education (MBO) G) Higher general education (HAVO) H) Pre-university education
I) Higher vocational education (HBO)
J) University level (Bacherlor’s or Master’s Degree) K) Other, namely
On the next two pages you will encounter four messages from an ad about meat substitutes. Each message will contain its own set of questions that you can answer. I would like to ask you to carefully look at the messages one by one, and answer the questions. There are no
33 right or wrong answers, because the questions should reflect your opinion on how you
perceive the message. It is important that you fill out the whole survey.
Message 1
“Buying our delicious meat substitutes instead of ‘real beef’ will not only make you feel healthier, but will also help you keep slim and fit. Furthermore it doesn’t contain any saturated fats or cholesterol and decreases your chance of getting certain diseases, like Type 2 diabetes and stomach cancer.”
1) To what extent do you find the tone of the first message negative or positive?
Negative 1 2 3 4 5 6 7 Positive
2) Does the first message put the emphasis on the benefits of buying ‘meat substitutes’ or the disadvantages of buying ‘real beef’?
Disadvantages of ‘real beef’ 1 2 3 4 5 6 7 Advantages of ‘meat substitutes’
3) Does the first message put emphasis on the consequences of risky behaviour or the impact of safe behaviour?
Consequences of risky behaviour 1 2 3 4 5 6 7 Impact of safe behaviour
4) To what extent do you agree that the claim in the first message is about a person’s health? Totally disagree 1 2 3 4 5 6 7 Totally agree
5) To what extent do you agree that the claim in the first message is about earth’s ecological welfare?
34 Message 2
“Buying ‘real beef’ instead of our delicious meat substitutes will not only make you feel less healthy, but will also prevent you from being slim and fit. Furthermore it contains saturated fats, cholesterol and increases your chance of getting certain diseases, like Type 2 diabetes and stomach cancer.”
1) To what extent do you find the tone of the second message negative or positive?
Negative 1 2 3 4 5 6 7 Positive
2) Does the second message put the emphasis on the benefits of buying ‘meat substitutes’ or the disadvantages of buying ‘real beef’?
Disadvantages of ‘real beef’ 1 2 3 4 5 6 7 Advantages of ‘meat substitutes’
3) Does the second message put emphasis on the consequences of risky behaviour or the impact of safe behaviour?
Consequences of risky behaviour 1 2 3 4 5 6 7 Impact of safe behaviour
4) To what extent do you agree that the claim in the second message is about a person’s health?
Totally disagree 1 2 3 4 5 6 7 Totally agree
5) To what extent do you agree that the claim in the second message is about earth’s ecological welfare?
35 Message 3
“Buying our delicious meat substitutes instead of ‘real beef’ is not only less harmful to the ecological system of this planet, but also helps us to prevent further climate change. Furthermore it uses less land to produce and increases the chance of a sustainable future for us and all the generations to come.”
1) To what extent do you find the tone of the third message negative or positive?
Negative 1 2 3 4 5 6 7 Positive
2) Does the third message put the emphasis on the benefits of buying ‘meat substitutes’ or the disadvantages of buying ‘real beef’?
Disadvantages of ‘real beef’ 1 2 3 4 5 6 7 Advantages of ‘meat substitutes’
3) Does the third message put emphasis on the consequences of risky behaviour or the impact of safe behaviour?
Consequences of risky behaviour 1 2 3 4 5 6 7 Impact of safe behaviour
4) To what extent do you agree that the claim in the third message is about a person’s health? Totally disagree 1 2 3 4 5 6 7 Totally agree
5) To what extent do you agree that the claim in the third message is about earth’s ecological welfare?
