1
The influence of puffery on product evaluation, willingness to pay and ad
credibility, and the moderating influence of feeling awe, consumer’s
intelligence and consumer’s prior knowledge
Master Thesis, Msc. Marketing Management
University of Groningen, Faculty of Economics and Business
June 22
,2016
Vincent Adema
Student number 2159791
Boterdiep 64D
9712LS Groningen
tel: +31(0)6 57744184
e-mail:
vincentadema@gmail.com
First supervisor
Yannick Joye
Second supervisor
2
Table of content
Abstract 3 Introduction 3 Literature review 4 Puffery 4 Feeling awe 6 Intelligence 8 Prior knowledge 8 Hypotheses 9 Methodology 11Participants and design 11
Intelligence 11
Emotional state 12
Puffery 12
Product evaluation, willingness to pay and ad credibility 13
Prior knowledge 13 Results 14 Manipulation checks 14 Correlations 14 Main results 15 Puffery 15 Emotional state 17
Interaction between puffery and the emotional group 18 Moderating influence of intelligence 19 Moderating influence of prior knowledge 21
Discussion 24
Discussing the results 24
Future research 25
Limitations 26
Implications for management and government 26
3
ABSTRACT
In this paper, I have studied how the level of puffery in an advertisement affected product evaluation, willingness to pay and credibility of the advertisement. I also studied how the feeling of awe, the prior knowledge of the respondent and the intelligence of the
respondent affected these relationships. Based on prior research, I expected puffery to have a positive impact on the product evaluation and willingness to pay, while the impact on the credibility of the ad would be negative. I expected a moderating influence of the feeling of awe, intelligence and prior knowledge. The results were mixed. High levels of puffery positively influenced the product evaluation, negatively influenced ad credibility, but had no significant effect on the willingness to pay. The moderating influences were mixed, finding no significant direct effect or moderating influence of the feeling of awe, while finding mostly significant influences of intelligence and prior knowledge. Not all of the moderating influences are as predicted.
INTRODUCTION
Losing twenty pounds in a month, growing back all your hair, keeping the skin of a twenty year old, while making thousands of dollars from the comfort of your home. If
advertising claims always told the truth, life would be much easier. Unfortunately, advertising does not always work like that. Advertising is a domain where a lot of puffery is present. Claims, both those presented as factual and those presented as subjective, can be puffed to persuade consumers to buy the advertised product or service. And unfortunately, this behavior is often both effective and legal.
Puffery, or exaggeration, in advertising is defined by regulatory organizations like the Federal Trade Commission as [something] frequently used to denote exaggerations
reasonably to be expected of a seller as to the degree of quality of his product, the truth or falsity of which cannot be precisely determined and that consumers are able to identify these claims as they are, exaggerated or fake.
Rotfeld and Rotzoll (1980) argue that “the legal logic with puffery is that even if such claims are communicated to consumers, no one would seriously believe them. In other words, even if communicated, they cannot possess a tendency to deceive since they will not be believed and therefore cannot cause actual deception” (p. 16). But they found that “the communicated puffery claims were very likely to be believed” (p. 19).
And as Cowley (2006) has shown in her research about the processing of puffed claims, there are a lot of factors influencing the belief and effectiveness of puffery. According to her research, part of the population is not able to identify puffery claims as fake or puffed. And even when consumers are able to identify claims as puffery, they are not always able to label the information as puffery. This means that, upon retrieval of the information, not everyone is able to remember that the claim is fake or puffed. And finally, even when the consumer identifies a claim as puffery, the puffed brand is still rated higher than the brand without puffed claims (Cowley, 2006).
4
1980, Cowley, 2006, Goldberg & Hartwick, 1990) on both consumer attitudes towards the advertised product and consumer behavioral intentions, like willingness to buy and pay, the importance of understanding when puffery does and doesn’t work has been increasing.
One factor that may play a key role in the effectiveness of puffery is the emotional state of the consumer. Howard and Gengler (2001) have shown that the emotional state the consumer is in can influence how we evaluate a certain product, service or experience, even if the emotional state has nothing to do with the product, service or experience. One of these emotional states, awe, will be the focus of this paper. Because the feeling of awe can create feelings of smallness and feeling frozen (as in not being reactive to circumstances), puffery may work more efficiently when this emotion is primed because individuals may feel unable, or too weak to argue against the persuasive puffed messages.
Besides puffery and the emotional state of the consumer, the moderating influence of certain consumer characteristics is studied. Both the intelligence of the consumer and the prior level of knowledge of the consumer their influence as a moderator in the relationship between puffery and the dependent variables is tested. Based on previous research, I argue that higher levels of intelligence and prior knowledge could weaken the effectiveness of puffery.
Because of this, the research question of this paper is as follows. What is the effect of
puffery in advertising on product evaluation, willingness to pay and ad credibility, and how is this influenced by the feeling of awe, the intelligence of the consumer and the level of prior knowledge of the consumer.
The structure of this paper is as follows. First, I will review current research on the influence of puffery in advertising and the influence of the feeling of awe on product evaluation and willingness to pay. I will also look at how these variables influence the credibility of the ad. Then, I introduce the two moderators of this model, intelligence and prior knowledge. Based on this, I propose a collection of hypotheses with the corresponding conceptual model. After describing both the methodology and results of the study, I will discuss the results of the study, my ideas for future research, the limitations of the study and the implications for managers and government.
LITERATURE REVIEW Puffery
Puffery in advertising has been present as long as advertising itself. Whether it is in claims presented as factual, claims presented as subjective, or present in the images or videos associated with the advertisement, current and past research has shown repeatedly that puffery is effective.
5
Marks and Kamins (1988) have shown that, while direct experience is more influential in changing both consumer attitude and behavior, the impact of puffed claims plays an
important role. For example, when a consumer is exposed to a (puffed) positive claim and has a negative direct experience afterwards, their overall rating of the product is higher than when they only have the negative direct experience.
As discussed in the introduction, the research by Rotfeld and Rotzoll (1980) has shown us that puffed claims are believed by a relatively large number of respondents and were very likely to be believed. They found that up to 69% of the puffs were believed by their respondents, and in discussing their results, talk about the relatively high level of education of the respondents. They argue that because of this, their sample “(..) might be seen as more intelligent, more perceptive, and / or skeptical of advertising messages. This could mean that even higher levels of belief could be found in the general population” (p.20).
Finally, Cowley (2006) has shown that although puffed claims are seen as less credible than real claims, they are still credible. Cowley also found that even after identifying a claim as puffed, the rating of the advertised object continues to be higher than before and that when the exposed consumer is interrupted during studying or reflecting on the puffed claim, the claim is seen just as credible as a non-puffed claim.
But why does puffery work? First of all, consumers are not always able to identify claims as puffed. Shimp and Preston (1981) find that consumers are not always able to recognize puffed claims for what they are. There are multiple reasons for this, like consumers lacking the cognitive resources or level of knowledge that is necessary to identify claims as puffed. If you don’t know what you can realistically expect of a certain product, or have no knowledge of a product category, recognizing claims as puffed is extremely difficult. In this case, respondents may use external reasons to believe, as discussed by Sperber (2010) in his discussion about the guru-effect. If people lack the ability to evaluate or comprehend a claim, they may use external reasons to believe, like the reputation of the source.
This is also the explanation of Goldberg and Hartwick (1990). They quote Fishbein and Azjen’s (1975) work, who argue that when a source is seen as highly credible, receivers of the message may accept the claims easier and change their attitude more strongly. This argument is confirmed by the study of Goldberg and Hartwick (1990), as can be read in the beginning of this section.
6
“people are quite poor at rejecting, ignoring, or failing to believe comprehended [information]” (Cowley, 2006 p. 729).
Although both schools of thought have a different explanation of why puffery is effective, the one does not exclude the other. It may be that when a person is unable to accurately comprehend a puffed claim, external reasons to believe the claim are used as the reason to believe the claim and make the person change his or her attitude. On the other hand, if a person is able to identify a claim as puffed, the framework of the second school of thought may be the most important reason why puffery continues to be effective. Even though the person may recall the claim as puffed or fake, the claim itself still influences his or her attitude towards the advertised object.
Feeling of awe
On the 26th of September in 1960, the debate between presidential candidates John F. Kennedy and Richard Nixon was held. This was the first U.S. presidential debate ever to be broadcasted on television. After this debate, an interesting thing occurred. People who watched the debate on television, gave Kennedy the edge, while people who listened on the radio gave Nixon the edge. What could be the reason of this result? While radio listeners could only listen to the arguments of the debaters, people who followed the debate on television also saw Kennedy and Nixon their physical appearance. While Kennedy looked well-groomed and confident, Nixon looked less stellar, which seems to have influenced the balance in favor of Kennedy (History.com). Although there are plenty of other reasons why the difference in ratings could have occurred, the anecdote itself a good example of the influence of how the state of awe or admiration can induce powerful effects in people their perception of the situation.
7
As discussed in the opening section of this paper, Howard and Gengler (2001) have found that the emotional state a consumer is in can influence his or her evaluation of the product, service, or experience, even if the emotion itself doesn’t come from the product, service or experience. Based on this reasoning, the feeling of awe, itself a powerful emotion, can have a powerful effect on the evaluation of a product. If a consumer is in the presence of something powerful, feels small and humbled, and has a difficulty comprehending
information, this feeling may affect product evaluation, willingness to buy and ad credibility. The moderating influence of awe on the relationship between puffery and the dependent variables may be even more powerful. The aforementioned effects of the feeling of awe can significantly influence how puffery works and how effective it is.
There has been a couple of studies that look how awe influences people their behavior. Important here is the work of Griskevicius, Shiota and Neufeld (2010), who have found that when people feel awe, weak arguments are much less persuasive than when people feel other emotions, like amusement or anticipatory enthusiasm, and also less persuasive than people in the control condition. Because there were large differences between the persuasiveness of weak and strong arguments for the feeling of awe, but only minor and insignificant differences between the persuasiveness of weak and strong arguments for other positive emotions like amusement or anticipatory enthusiasm, Griskevicius et. al. (2010) argue that the feeling of awe makes people process information deeper and in a more systematic way. Awe also gave rise to a lower attitude towards the message than other positive emotions.
More evidence of the impact of awe can be found in the work of Joye. Two papers are relevant here. First of all, Joye and Bolderdijk (2015) studied the influence of extraordinary nature on emotions, moods and prosociality and show that awe-evoking pictures of
extraordinary nature improve mood, emotions and prosociality. In the paper of Joye and Dewitte (2016), the effect of awe-evoking high buildings on behavior is studied. They find that awe can lead to the feelings of smallness, freezing and entrapment and that it can decrease response time, all behaviors that create a state of being passive.
The findings of Griskevicius et. al. (2003) and the work of Joye (with Bolderdijk, 2015, and with Dewitte, 2016) seem to offer two perspectives of the influence of awe on product evaluation, willingness to pay and ad credibility. Looking at Griskevicius et. al. (2003) their findings, awe creates deeper processing and less positive attitudes. Based on this, I would argue that the effect of awe on my dependent variables is negative. But Joye and Bolderdijk (2015) show mostly positive effects of awe. A better mood, more positive emotions and more prosociality could imply a positive effect of awe on the dependent variables in this study. The passivity implied by and Joye and Dewitte (2016) could also imply a more positive effect (or less harmful effect) of puffery on the dependent variables in this study.
8
type of awe that advertising would try to prime is mostly positive, my hypotheses imply a positive effect.
Intelligence
There is no research about the moderating influence of intelligence on the relationship between puffery and product evaluation, willingness to pay and ad credibility. However, research about the effects of intelligence and education on information processing capabilities are plentiful.
The work of Shimp and Preston (1981), as discussed in the puffery section, argues that intelligence may play a crucial role in the effectiveness of puffery. They argue that puffery may not be detected because consumers lack the intelligence or prior knowledge that is necessary to identify a claim as puffed.
The findings of the work of Frederick (2005) are also important for my hypotheses concerning the effects of intelligence. His work focused on the influence of cognitive reflection on decision making. Two interesting findings are relevant for this paper. The first one is that people with a higher IQ often take more time to process information cautiously. Secondly, people with higher IQ’s are often better in delaying gratification. These findings could mean that intelligent people are less interested in quick fixes (where puffery often reigns supreme) and are less gullible to puffery because they take more time to process and reflect on the information.
The work of Schroder, Driver and Strenfert (1967) and Henry (1980) confirms the idea that intelligence plays a large factor. Both papers find that there are large differences in
people’s ability to assimilate, retain and integrate information. And these differences have a significant influence on the accuracy when reproducing information (Henry, 1980).
Besides this, one can also argue that the level of education of a consumer increases his or her ability to find and process information, based on a healthy foundation of education and studying in one’s youth. And as will be described in the following section, the ability to find and process information has a significant impact on the way consumers make decisions.
Based on these findings, I expect that consumers with a higher level of intelligence find it easier to use and process the information in an puffed claim and detect the puffery for what it is. This would make puffery less effective and the advertisement less credible. One important side note is that the research from Cowley (2006) has shown us the importance of having the time to reflect on the information. When consumers are interrupted while
processing the information in a puffed claim, they consider the claim as credible as a real one. So having enough time to process the claim completely is important to find the differences as suggested by the hypotheses.
Prior knowledge
9
First of all, I want to refer to the work of Shimp and Preson (1981) again, who argue that one needs to have enough prior knowledge to identify a claim as puffed. This make sense. And it has been shown that consumers with a high level of knowledge (or “experts”) are better able to think deeply about information than equally motivated but less knowledgeable consumers (or “novices”) (Alba & Hutchinson, 1987), are better able to acquire new
information (Brucks, 1985, Johnson & Russo, 1984) and that they process information in different ways (Maheswaran & Sterntal, 1990).
Hui, Zhao, Fan and Au (2004) have shown that, when consumers lack knowledge or experience, they may use heuristics, or simple cues, to evaluate whether a service is good or not.
Finally, Gregan-Paxton and John (1997) have shown that experts can process
information in terms of product attributes, while novices have more difficulty with this way of processing information.
Based on these findings, I argue that prior knowledge of the consumer has a
significant impact on how they process information about products or services. It seems that consumers with a low level of knowledge have multiple difficulties in creating objective ideas about an advertised product or object, while consumers with a high level of knowledge have the ability to rate products on their attributes, process information more easily and find new information if this is necessary. It also allows them to rate the credibility of the claims with more precision. One important notation to this argument is that Johnson and Russo (1984) found that when consumers had to pick an one product out of a collection of products, the relationship between the level of knowledge and decision making showed an inverted U shape. In this case, this means that consumers with a high level of knowledge only use their own knowledge to pick one of the products (and collect less new knowledge), while
consumers with a moderate level of prior knowledge also used the information that was given in the product overview.
But because high levels of knowledge generally allow for easier processing of information and rating products based on important product attributes, I expect that high levels of knowledge will decrease lower the product evaluation, willingness to pay and ad credibility when puffery increases.
HYPOTHESES
The first group of hypotheses focus on the effects of the level of puffery on the product rating, willingness to pay and the rating of ad credibility. Based on the arguments made in the literature, I propose the following hypotheses.
H1a: Puffery has a positive impact on the evaluation of the advertised object.
H1b: Puffery has a positive impact on the willingness to pay for the advertised object. H1c: Puffery has a negative effect on the credibility of the ad
10
expect that the feeling of awe will have a positive effect on the positive relationship between puffery and product evaluation and the willingness to pay. However, I believe that the feeling of awe negatively moderates the relationship between puffery and ad credibility.
H2a: Feeling awe has a positive impact on the evaluation the advertised object.
H2b: Feeling awe has a positive impact on the willingness to pay for the advertised object. H2c: Feeling awe has a positive impact on the credibility of the ad
H2d: Feeling awe positively moderates the positive impact of puffery on the evaluation the advertised object.
H2e: Feeling awe positively moderates the positive impact of puffery on the willingness to buy for the advertised object.
H2f: Feeling awe negatively moderates the negative impact of puffery on the credibility of the ad.
The third group of hypotheses focus on the effects of the respondent’s intelligence as a moderator on the relationship between puffery and product evaluation, willingness to pay and ad credibility. I expect that a higher level of intelligence decreases the strength of the
relationship between puffery and product evaluation and willingness to pay. For the expected negative relationship between puffery and credibility, I expect intelligence to be a positive moderating variable.
H3a: Intelligence of the respondent negatively moderates the positive relationship between puffery and the evaluation of the advertised object.
H3b: Intelligence of the respondent negatively moderates the positive relationship between puffery and the willingness to pay for the advertised product.
H3c: Intelligence of the respondent positively moderates the negative relationship between puffery and the credibility of the ad
Finally, the fourth group of hypotheses focuses on the moderating effects of prior knowledge on the relationship between puffery and product evaluation, willingness to pay and ad credibility. I expect that higher levels of prior knowledge about the advertised product will be a negatively moderator for the relationship between puffery and product evaluation and willingness to pay and that prior knowledge positively moderates the relationship between puffery and ad credibility.
H4a: Prior knowledge about the type of product advertised negatively moderates the positive relationship between puffery and the evaluation of the advertised object.
H4b: Prior knowledge about the type of product advertised negatively moderates the positive relationship between puffery and the willingness to pay for the advertised product.
H4c: Prior knowledge about the type of product advertised positively moderates the negative relationship between puffery and the credibility of the ad.
11
METHODOLOGY Participants and design
For this study, I recruited 228 respondents with the help of Amazon Mechanical Turk. 123 of these were male, 105 female. Their age ranged between 18 and 72 (M = 36.2, SD = 10.8). Their education level ranged from not finishing high school (1), finished high school (80), a bachelor’s degree (123), a master’s degree (23) to a PhD or higher (4).
The study was done with the online questionnaires, consisting of 34 questions. The respondents completed the questionnaire digitally. The average payment for completing the survey was $0.91. The study had a 3 by 3 between-subjects design, priming either the feeling of awe, the feeling of amusement, or a neutral (control) feeling. Afterwards, they were presented with an advertisement with either a low level of puffery, a moderate level of puffery, or a high level of puffery.
Figure 1. Conceptual Model
Intelligence
After asking for respondent’s age and gender, I measured the intelligence of the respondents in two ways. First, I asked them what their highest completed level of education was. The options were “I haven’t finished high school”, “high school”, “bachelor’s degree”, “master’s degree” and “PhD or higher”. Second, I used the Cognitive Reflection Test (CRT) from Frederick (2005), asking each respondent the following three questions.
(1) A bat and a ball cost $1.10 in total. The bat costs $1.00 more than the ball. How much does the ball cost?
(2) If it takes 5 machines 5 minutes to make 5 widgets, how long would it take 100 machines to make 100 widgets?
12
takes 48 days for the patch to cover the entire lake, how long would it take for the patch to cover half of the lake?
Frederick (2005) finds significant differences between the group of respondents who answer incorrect and the group that answers correct on all three questions. The results of the CRT test are compared with six other cognitive tests, like SAT scores, ACT (American College Training) scores and the WPT (Wonderlic Cognitive Test) score. The first two of these tests are used in the United States for Undergraduate Admissions. The latter one is used to measure cognitive abilities and is often used for job interviews. On the four different domains (intertemporal choice and three different choice under uncertainty tests), CRT test scores were either the best of second best predictor of good behavior. Frederick (2005) argues that, “for researchers interested in separating people in cognitive groups, the CRT is an attractive test” (p. 36-37).
Emotional state
Following the intelligence tests, one of following three emotional states were primed using scenarios taken from Griskevicius et. al. (2010).
Awe. Please try to recall an event in your life when you saw a particular panoramic view for the first time. Some examples might be seeing the Grand Canyon, seeing the view from high up on a mountain, or seeing the skyline of a big city for the first time. Please recall a specific event when you saw this view for the first time, rather than a general period of time.
Amusement. Please try to recall an event in your life when you heard a funny joke or when something funny happened. Please recall a specific event that was funny, rather than a general period of time
Control. Please try to recall the last time you did the laundry, the dishes, or another household chore.
After remembering the event, respondents were asked to describe this memory in detail, by writing a couple of sentences. Following this, they had to indicate, on a scale from 1 to 7 (1 = Does not describe my feelings – 7 = Clearly described my feelings) how strongly they felt “Anger”, “Awe”, “Disgust”, “Fear”, “Pride”, “Sadness” and “Happiness”. These were taken from Griskevicius et. al. (2010). On the same scale, they were also asked to describe whether the statements “I feel a strong sense of smallness”, “I feel like I am in the presence of something greater than myself”, “I feel part of some greater entity” and “I feel like I am in the presence of something grand” described their feelings or not. This was to measure feelings of smallness and was taken from the work of Piff, Dietze, Feinberg, Stancato and Keltner (2015).
Puffery
13
Low level of puffery. This newly introduced program to lose weight quickly and efficiently, contains everything you need and is rated as a good weight loss program. You will learn about the different exercises that are part of the program, about how to eat healthy and get some of our weight loss supplements. With this program, you are able to lose around 2 pounds (or 1 kilogram) each week.
Moderate level of puffery. This newly introduced program to lose weight quickly and
efficiently, contains everything you need and is rated as a great weight loss program. You will learn about the different exercises that are part of the program, about how to eat healthy and get some of our weight loss supplements. With this program, you are able to lose around 5 pounds (or 2.5 kilograms) each week.
High level of puffery. This newly introduced program to lose weight quickly and efficiently, contains everything you need and is widely regarded as the best weight loss program in the world. You will learn about the different exercises that are part of the program, about how to eat healthy and get some of our weight loss supplements. With this program, you are able to lose around 10 pounds (or 5 kilograms) each week.
These advertisements were inspired by the advertisements of Goldberg and Hartwick (1990). The goal was to have a level of contrast on multiple levels (how well the product was rated and how much weight you are able to lose).
Product evaluation, willingness to pay and ad credibility
After introducing the puffery manipulation, respondents were asked to indicate how high they expected the weight loss program would be placed in a ranking by a consumer agency, between position 1 (best) and position 50 (worst). This measure has been taken from Goldberg and Hartwick’s (1990) work on advertising credibility and evaluation.
The questions regarding the credibility of the ad were also taken from Goldberg and Hartwick (1990). In these questions, consumers were asked to compare the advertisement they just saw with most other advertisements they saw on a 7 point Likert scale. They were asked whether they found the advertisement either deceptive, misleading, dull,
unprofessional, unsophisticated, boring and unrealistic (rated as 1) or honest, sincere, exciting, professional, sophisticated, interesting and realistic (rated as 7). These seven questions can be combined in one factor, which had a Cronbach’s alpha of .89.
Finally, the willingness to pay measure asked consumers which price was the
maximum they would pay for the program, before they couldn’t afford it anymore or wouldn’t think the price was fair. They were also asked to indicate the minimum price they would pay for the program, before they had the feeling the quality of what they would get in return would be subpar. This scale is taken from Breidert, Hahsler and Reutterer (2006). Prior knowledge
14
know a lot about healthy nutrition”, “I know a lot about fitness and exercise”, “I know a lot about losing weight” and “There has been a period in my life in which I consciously dieted, exercised and lost weight”.
The first five of these six questions could be combined in one factor. The Cronbach’s alpha of this new factor is .93.
RESULTS Manipulation checks
First of all, I tested whether priming the different emotional states had a significant effect on the level of awe respondents felt. A one way ANOVA returns a significant
difference in the feeling of awe between the three groups with F(2, 225) = 74.71, p = .00. In the awe group (M = 5.00, SD = 0.20), participants felt more awe than the participants in the amusement group (M = 2.57, SD = 0.21) and control group (M = 1.72, SD = 0.19). The difference between the feeling of awe between the awe-primed and amusement-primed group is significant, p = .00, just as the difference between the amusement-primed and control group, with p = .00.
Correlations
I start off with an overview of the observed correlations between the different continuous variables. The seven measured emotions (anger, awe, fear, sadness, happiness, pride, disgust) and the five dependent variables (product ranking, maximum price, minimum price, overall liking and ad credibility) have all been included. The results can be seen in table 1. The correlations shows some interesting phenomena.
First of all, with only a few exceptions, almost all of the dependent variables correlate significantly together. For example, when the product ranking increases (and gets closer to position #1), the maximum price respondents are willing to pay increases, the overall liking of the program increases, and the ad credibility increases. A similar thing is found with the overall liking of the program. The higher the overall liking of the program, the more
respondents are willing to pay and the higher they rate the credibility of the advertisement. Another interesting finding are some of the significant correlations between the how the respondents feel and how they score the five dependent variables. The correlations surrounding the feeling of awe are minimal, although it is interesting to see that the
correlation between the feeling of awe and the maximum price respondents are willing to pay is significant. As can be seen in the matrix, correlations between the dependent variables and both the feeling of pride and happiness give the most significant results. Higher levels of pride go together with a higher product ranking, higher liking of the program and a higher ad
15
Table 1
Correlation Matrix of the different Independent and Dependent Variables
Measure Product ranking Maximum price Minimum price Overall liking Ad credibility Feeling awe Feeling anger Feeling disgust Feeling fear Feeling pride Feeling sadness Feeling happiness Product ranking Maximum price -.25** Minimum price -.12 .44** Overall liking -.55** .32** .12 Ad credibility -.43** .26** .14* .76** Feeling awe .01 .18** .02 .03 .08 Feeling anger .10 .01 .06 -.09 -.03 .04 Feeling disgust .10 -.01 .06 -.01 .04 -.04 .78** Feeling fear .10 -.01 .03 -.07 -.01 .16* .55** .39** Feeling pride -.14* .08 -.00 .22** .27** .42** .03 -.03 .20** Feeling sadness .18** -.07 -.02 -.08 .05 .01 .57** .41** .59** .08 Feeling pride -.19** .17** .07 .29** .36** .42** -.16 -.18** .06 .54** -.17* *p < .05. **p < .01. Main results
Multiple two way ANOVA tests have been conducted with puffery and the primed emotion as independent variables. The dependent variables were respondents’ product evaluation, willingness to pay, overall liking of the product and ad credibility. The results of these tests are summarized in table 2. In general, I use an alpha level of .05 as cut-off level for significance, although some results with an alpha level of .10 are also discussed as moderately significant. Below I discuss the direct effect of puffery and the feeling of awe, after which I discuss the moderating influence of the feeling of awe, intelligence and prior knowledge. Puffery
To test the effect of puffery on the dependent variables, a one way ANOVA has been used. An overview of the effect of puffery on the dependent variables can be seen in table 2. The differences between the different levels of puffery for each of the dependent variables can be seen in table 3.
16
14.31. This is depicted in graph 1.Where I expected to see either a curvilinear effect or an inverted U-curve, the results show an unexpected U-curve. But because the overall effect is significant and a high levels of puffery improve the product ranking, I consider the effect proven. Hypothesis 1a is supported.
Table 2
The Direct Effect of the Emotional Group, Puffery, and the Interaction between these two variables on Product Evaluation, Willingness To Pay and Ad Credibility
Independent Variable
Dependent Variable df F P
Emotional group Ranking one to fifty 2 1.94 .15
Maximum price 2 0.76 .47
Minimum price 2 1.65 .19
Liking of the product 2 1.97 .14
Credibility of the ad 2 0.52 .59
Puffery Ranking one to fifty 2 4.51 .01***
Maximum price 2 2.06 .13
Minimum price Liking of the product Credibility of the ad 2 2 2 2.19 0.73 4.33 .12 .48 .01*** Interaction emotional group and puffery
Ranking one to fifty Maximum price Minimum price Liking of the product Credibility of the ad 4 4 4 4 4 0.29 0.65 0.14 0.60 0.17 .89 .63 .97 .66 .95
Hypothesis 1b predicted that puffery would have a positive influence on the
willingness to pay. Both the effect of puffery on the maximum price (F(2, 227) = 2.06, p = .13) and the effect on minimum price (F(2, 227) = 2.19, p = .13) is insignificant. As can be seen in table 3, the maximum and minimum price does seem to increase with the level of puffery in the advertisement. The differences between the ad with low puffery and the ad with high puffery are significant (p < .05) on both the maximum price and minimum price
measure, but the overall effect is insignificant. Hypothesis 1b is not supported.
Hypothesis 1c predicted that puffery would have a negative influence on the ad
17
level of puffery ads is significant (at the p < .05 level). The difference between the medium and high level of puffery ads is insignificant. Hypothesis 1c is supported.
Figure 1
Product Evaluation for Different Levels of Puffery
Table 3
The Mean and Standard Error of Product Evaluation, Willingness To Pay and Ad Credibility for different levels of Puffery
Product evaluation Maximum price Minimum price Overall liking Ad credibility M SE M SE M SE M SE M SE Low puffery 16.83 1.46 73.70 13.27 21.34 9.23 4.65 0.17 0.28 0.12 Mid puffery 20.51 1.48 84.74 13.48 33.24 9.43 4.36 0.18 -0.17 0.12 High puffery 14.31 1.45 110.75 13.24 48.70 9.26 4.57 0.17 -0.11 0.15
M(I-J) SE M(I-J) SE M(I-J) SE M(I-J) SE M(I-J) SE
Difference low(I) – mid(J) -3.68* 2.08 -11.04 18.92 -11.91 13.24 0.29 0.25 0.45*** 0.16 Difference low(I) – high(J) 2.52 2.06 -37.05** 18.75 -27.36** 13.12 0.08 0.25 0.38** 0.16 Difference mid(I) – high(J) 6.20 2.07 -26.02 18.89 -15.45 13.22 -0.21 0.25 -0.63 0.16 *p < .10. **p < .05. ***p < .01. Emotional state
To test the effect of the different emotional states on the dependent variables, a one way ANOVA has been used. An overview of the effect of the emotional state on the
0 5 10 15 20 25
Low puffery Mid puffery High puffery
18
dependent variables can be seen in table 2. The differences between the different emotional states for each of the dependent variables can be seen in table 4.
Hypothesis 2a predicted that the feeling of awe would have a positive effect on the product evaluation. The results returned insignificant, with F(2, 227) = 1.94, p = .15. The differences between awe and amusement are minor, while the difference between awe and control is only significant at the p < .10 level. Hypothesis 2a is not supported.
Hypothesis 2b predicted that the feeling of awe would have a positive effect on the willingness to pay. The results for both the maximum price (F(2, 227) = 0.76, p = .46) and minimum price (F(2, 227) = 1.65, p = .19) are insignificant. Even though the results are insignificant, it is interesting to see that both the maximum price of the awe group is the highest, while the minimum price of the awe group is the lowest, as can be seen in graph 2. Hypothesis 2b is not supported.
Hypothesis 2c predicted that the feeling of awe would have a positive impact on the credibility of the ad. This effect is also insignificant, with F(2, 227) = 0.52, p = .59.
Hypothesis 2c is not supported.
Interaction between puffery and the emotional group
The interaction between puffery and the emotional groups has also been tested, with a two way ANOVA. The moderating influence of awe on the effect of puffery on the dependent variables has been tested with the PROCES macro from Hayes with model 1. The results of the interaction can be seen in table 2. The effects of the test for moderation can be seen in table 6. A comparison of the means for the different emotional groups, given the level of puffery, can be seen in table 5.
Figure 2
Willingness To Pay for the Different Emotional Groups
Hypothesis 2d predicted that that awe would positively moderate the positive relationship between puffery and product evaluation. In this relationship, there was no interaction between puffery and the emotional state (F(4, 227) = 0.29, p = .89). The
0 20 40 60 80 100 120
Awe Amusement Control
19
moderating effect was also insignificant (F(3, 224) = 1.72, p = .16). Hypothesis 2d is not supported.
Hypothesis 2e predicted that awe would positively moderate the positive relationship between puffery and product evaluation. In this relationship, there was no interaction between puffery and the emotional state (for the maximum price F(4, 227) = 0.65, p = .63, for the minimum price F(4, 227) = 0.14, p = .97). The moderating effect was also insignificant (for the maximum price F(3, 224) = 1.78, p = .15, for the minimum price F(3, 224) = 1.77, p = .15). Hypothesis 2e is not supported.
Hypothesis 2f predicted that that awe would negatively moderate the negative relationship between puffery and ad credibility. In this relationship, there was no interaction between puffery and the emotional state (F(4, 227) = 0.17, p = .95). The moderating effect was also insignificant (F(3, 224) = 1.82, p = .15). Hypothesis 2f is not supported.
Moderating influence of intelligence
To test the moderating influence of intelligence, I used the PROCESS macro from Hayes with model 1. The results of these tests can be seen in table 6. Hypothesis 3a predicted that intelligence would negatively moderate the positive relationship between puffery and product evaluation. The result is insignificant, with F(3, 224) = 1.06, p = .37. Hypothesis 3a is not supported.
Table 4
The Mean and Standard Error of Product Evaluation, Willingness To Pay and Ad Credibility for the Awe, Amusement and Control Condition
Product evaluation Maximum price Minimum price Overall liking Ad credibility M SE M SE M SE M SE M SE Awe 18.47 1.46 101.76 13.30 24.43 9.31 4.24 0.18 -0.09 0.12 Amusement 18.24 1.54 77.93 14.02 48.22 9.81 4.71 0.18 0.08 0.12 Control 14.94 1.39 89.51 12.63 30.63 8.84 4.62 0.17 0.01 0.11
M(I-J) SE M(I-J) SE M(I-J) SE M(I-J) SE M(I-J) SE
Difference awe(I) – amu(J) 0.24 2.12 23.84 19.32 -23.78* 13.52 -0.47* 0.25 -0.17 0.17 Difference awe(I) – con(J) 3.54* 2.01 12.25 18.35 -6.20 12.84 -0.37 0.24 -0.10 0.16 Difference amu(I) – con(J) 63.30 2.07 -11.58 18.87 17.59 13.20 0.10 0.25 0.07 0.16
Note. Amusement is shortened as amu and control as con.
20
Table 5
The Mean and Standard Error of Product Evaluation, Willingness To Pay and Ad Credibility, given their Emotional Group, for Different Levels of Puffery
Product evaluation Maximum price Minimum price Overall liking Ad credibility M SE M SE M SE M SE M SE Awe Low puffery 19.43 2.38 86.07 21.65 13.50 15.15 4.32 0.28 0.12 0.19 Mid puffery 20.39 2.38 116.71 21.65 26.89 15.15 4.36 0.28 -0.21 0.19 High puffery 15.60 2.81 102.50 25.61 32.90 17.92 4.05 0.34 -0.18 0.22 Amusement Low puffery 17.48 2.62 60.87 23.88 29.09 16.71 4.87 0.31 -0.45 0.21 Mid puffery 22.67 2.96 68.83 27.00 47.17 18.89 4.44 0.35 -0.19 0.24 High puffery 14.57 2.38 104.07 21.65 68.39 15.15 4.82 0.28 -0.02 0.19 Control Low puffery 13.58 2.57 74.17 23.38 21.42 16.36 4.75 0.31 0.26 0.20 Mid puffery 18.47 2.30 68.67 20.91 25.67 14.63 4.27 0.28 -0.11 0.18 High puffery 12.76 2.33 125.69 21.62 44.79 14.89 4.83 0.28 -0.12 0.19
Hypothesis 3b predicted that intelligence would negatively moderate the positive relationship between puffery and willingness to pay. The results are significant, with F(3, 224) = 4.14 and p = .01 for the maximum price and F(3, 224) = 4.26, p = .01 for the minimum price. The coefficient of the effect of puffery on the maximum price is much higher for
respondents with a level of intelligence below average, than for respondents with a level of intelligence that is at or above the average. This can be seen in graph 3. Hypothesis 3b is supported.
21
Moderating influence of prior knowledge
To test the moderating influence of intelligence, I used the PROCESS macro from Hayes with model 1. The results of these tests can be seen in table 6. Hypothesis 4a predicted that prior knowledge would negatively moderate the relationship between puffery and product evaluation. The effect of the moderation is significant, with F(3, 224) = 3.99, p = .01, but the opposite of what I expected. The coefficient of puffery on product evaluation for respondents with an above average level of prior knowledge is higher (-2.38) than the coefficient for respondents with an average level of prior knowledge (-1.22) or a below average level of prior knowledge (-.06). This is also depicted in graph 4. Hypothesis 4a is not supported.
Hypothesis 4b predicted that prior knowledge would negatively moderate the relationship between puffery and willingness to pay. The effect is significant for the maximum price (F(3, 224) = 2.91, p = .04, but not significant for the minimum price (F(3, 224) = 2.07, p = .11. The effect was as predicted, as respondents with a higher level of prior knowledge had a lower coefficient of puffery on maximum price (15.99) than respondents with an average level of prior knowledge (16.77) or low level of prior knowledge (17.56). These differences are small, but the overall effect is significant. Hypothesis 4b is supported.
Finally, hypothesis 4c predicted that prior knowledge would positively moderate the negative relationship between puffery and ad credibility. The moderation effect is significant, with F(3, 224) = 3.88, p = .01, but the effect was the opposite of what I predicted.
Respondents with a level of prior knowledge that was below average had a higher coefficient of puffery on ad credibility (-.24) than respondents with an average level of prior knowledge (-.20) or above average level of prior knowledge (-.16). This is also depicted in graph 5. Hypothesis 4c is not supported.
Figure 3
The Effect of Puffery on the Maximum Price (blue) and Minimum Price (orange) Respondents are Willing To Pay for Different Levels of Intelligence
0 5 10 15 20 25 30 35
22
Figure 4
The Effect of Puffery on Product Evaluation for Different Levels of Prior Knowledge
Figure 5
The Effect of Puffery on Ad Credibility for Different Levels of Prior Knowledge -2,5 -2 -1,5 -1 -0,5 0
One SD below average Average One SD above average
Coefficient of puffery on product ranking
for different levels of prior knowledge
-0,25 -0,2 -0,15 -0,1 -0,05 0 0,05 0,1 0,15 0,2
One SD below average Average One SD above average
23
Table 6
The Moderating Influence of the Emotional Group, Intelligence and Prior Knowledge on the Relationship between Puffery and Product Evaluation, Willingness To Pay and Ad Credibility
df F p β SE Ranking one to fifty Emo group 3 1.72 .16 0.54 1.27 Intelligence 3 1.06 .37 1.01 0.96 Prior knowledge 3 3.99 .01 -0.76 0.65
Max price Emo group 3 1.78 .15 9.23 11.41
Intelligence 3 4.14 .01 -9.31 8.51
Prior knowledge
3 2.91 .04 -0.52 5.93
Min price Emo group 3 1.77 .15 -0.03 8.00
Intelligence 3 4.26 .01 -8.65 5.96
Prior knowledge
3 2.07 .11 1.94 4.18
Overall liking Emo group 3 0.73 .53 0.08 0.15
Intelligence 3 1.58 .20 -0.08 0.11
Prior knowledge
3 3.05 .03 0.11 0.08
Credibility Emo group 3 1.82 .15 -0.02 0.01
Intelligence 3 5.22 .00 0.03 0.07
Prior knowledge
3 3.88 .01 0.03 0.05
24
DISCUSSION Discussing the results
Although not all of the hypotheses have been significant, plenty of interesting results have been found.
First of all, I have found that puffery in advertising have significant effects. Puffery can increase the evaluation of a product. My results say that the product that is advertised with a high level of puffery is rated the highest, followed by the products that are advertised with a low level of puffery, and finally the medium level of puffery. This finding is interesting, because I predicted either a (curvi-)linear relationship or an inverted U shape. Instead, I found a U shape, where the medium level of puffery is rated the lowest. Although I’m unable to prove this with my findings, it could be that the negative effects of puffery are only
counteracted by the positive effects of puffery when the puffery is high. Second, I have shown that ads that contain puffery are indeed seen as less credible, all other things being equal. This finding was as expected.
The effect of the three different emotional states that were used in this paper were mostly insignificant. Although the group that was primed with the feeling of awe rated their feelings of awe significantly higher than the amusement or control group, this feeling of awe did not have an effect on the four dependent variables. Some moderately significant results were found in the differences between the three groups, for example the group that felt awe rated the product higher than the control group and the amusement group had a higher minimum price for the product and liked the product better overall than the awe group. But these differences are only moderately significant. And awe did not have a moderating influence on the relationship between puffery and the four dependent variables.
There are more significant influences of the intelligence and prior knowledge
moderators. Intelligence moderated the relationship between puffery and the maximum and minimum prices. Intelligence also moderated the relationship between puffery and ad credibility. Some very interesting results were found. More intelligent respondents were slower to decrease the rating of the ad’s credibility than less intelligent respondents. This results is the opposite of what I expected to find and it is difficult to predict why this is the case. It could be that the respondents with a lower level of intelligence were quicker and harsher to judge than more intelligent respondents, but this is just a guess. More research is needed to explain this finding.
What is interesting is that even after finding the that unexpected result, higher levels of intelligence did decrease the willingness to pay for the product. Respondents with a lower level of intelligence have a higher willingness to pay, both wanting to pay a higher maximum price and a higher minimum price when puffery is present. This finding is as predicted, but is difficult to reconcile with the effect of intelligence as moderator in the relationship between puffery and ad credibility, as explained above. It may be that more intelligent respondents have a better idea about an honest price.
Prior knowledge moderated the relationship between puffery and the product
25
again, multiple unexpected results came back. Higher levels of prior knowledge made
respondents increase the evaluation of the products with high puffery, improved their overall liking of the product and gave lower decreases to the evaluation of the credibility than
respondents with lower levels of prior knowledge. All three results were the opposite of what I predicted. But once again, these respondents have shown a lower willingness to pay than respondents with a low level of prior knowledge. This was as expected, and could be because they have a better idea what the average price for a weight loss program is and base their maximum willingness to pay on those figures.
Future research
The results of this paper presents us with multiple avenues for future research. First, I have confirmed earlier findings (e.g. Goldberg and Hartwick, 1990 and Rotfel and Rotzoll, 1980) that puffery in advertising can work. Respondents that saw the advertisement with a high level of puffery evaluated the advertised product as better than the product advertisement that contained low or medium levels of puffery, even though they also think the advertisement is less credible than the advertisement with a low level of puffery. Just like Cowley (2006) found in her paper, the fact that the advertisement is seen as less credible does not mean that respondents think less of the advertised product. Both Cowley (2006) and I found products that were evaluated higher.
I also find that personal characteristics have a moderating influence on the effectiveness of puffery. More intelligent respondents and respondents with more prior knowledge about the market of the advertised product (in this study, losing weight) are less affected by puffery in the advertisement. It would be interesting to see if similar effects of prior knowledge and intelligence are found in other markets. The weight loss industry is an industry where puffery is used a lot. It may be that the majority of the people don’t know how fast the average person is able to lose weight.
It would also be interesting to see more work focus on the characteristics of people on which puffery works best. What is the influence of age? Or of the consumer’s involvement in the product category? Their level of concentration while watching the advertisement? Or their awareness of the presence of puffery in advertising?
26
product). For happiness, there is also a positive and significant correlation with the maximum price respondents are willing to pay.
I think these results are very interesting and are worth investigating further, because it shows that different emotional states can influence consumers liking of products, ranking of products and even the price they are willing to pay for the product. Further research about the effects of the different emotions (in relation with puffery) and the reason why would help us understand if priming certain emotions can help companies influence (or manipulate,
depending on your perspective) consumers more effectively. Limitations
Finally, this study had a couple of limitations. First of all, the intelligence of the respondents was measured with the help of a proxy, like the level of education and the cognitive reflection test. Although both are reasonably valid tools, they are not IQ tests. Having a more trustworthy measure like an IQ test may influence the results.
Secondly, the emotional state was primed externally and not induced by either the advertisement or the product itself. For the purposes of this study this should be sufficient, as Howard and Gengler (2001) have shown us that the existing emotional state can be
transferred to the product or service and this way of priming emotions has been used in plenty of research. Still, it would be interesting to see if the results are different when the emotions come from the product or advertisement itself.
Finally, the condition in which the respondent watched the advertisement was quite clinical, in that it wasn’t watched in a natural setting. I suspect most respondents to be aware of the fact that the advertisement was important for the study, even though they did not know what I wanted to research. This may have affected the way they processed the information or how concentrated they were. The fact that the advertisement was seen for the purpose of a study also affects the respondent’s interest in the product. If someone who was actively looking to lose weight had seen the advertisement, their interest and concentration could be significantly different than the respondents in the survey. This may have had an influence on the results.
Still, I have found that the three primed emotions significantly affected the way respondents felt, that I had a good idea about the respondent’s intelligence and that the three different advertisements brought about significant differences between the groups and that my results are valid overall.
Implications for managers and government
The results of this study show that puffery works and that its effectiveness depends on the characteristics of the consumer. I found that prior knowledge about the product category and the intelligence of the consumer have a significant moderation effect on the effectiveness of puffery. This is important to know.
First of all, government and regulatory bodies should take note of the confirmation of the effectiveness and the moderating influence of intelligence and prior knowledge.
27
income and more difficulties with being self-sufficient. Educating consumers about most important product- and service markets could decrease their susceptibility against puffery. Taking more active measures against limiting puffery could be more effective.
28
REFERENCES
Alba, J.W. & Hutchinson, J.W. 1987. Dimensions of consumer expertise. Journal of
Consumer Research, 13: 411-454.
Bjork, R.A. 1972. Theoretical implications of directed forgetting. Melton, AW, Martin.E. Breidert, C., Hahsler, M., & Reutterer, T. 2006. A review of methods for measuring willingness-to-pay. Innovative Marketing, 2(4): 8-32.
Brucks, M. 1985. The effects of product class knowledge on information search behavior. Journal of Consumer Research, 12: 1-16.
Crowley, E. 2006. Processing exaggerated advertising claims. Journal of Business Research,
59: 728-734.
Fishbein, M., & Azjen, I. (1975). Belief, attitude, intention, and behavior: an introduction to
theory and research. Reading, MA: Addison-Wesley.
Frederick, S. 2005. Cognitive reflection and decision making. Journal of Economic
Perspectives, 19(4): 25-42.
Gilbert, D.T., Krull, D.S., & Malone, P.S. 1990. Unbelieving the unbelievable: some problems in the rejection of false information. Journal of Personality and Social
Psychology, 59(4): 601-613.
Goldberg, M.E., & Hartwick, J. 1990. The effects of advertiser reputation and extremity of advertising claim on advertising effectiveness. Journal of Consumer Research, 17: 172-179.
Gregan-Paxton, J., & John, D.R. 1997. Consumer learning by analogy: a model of internal knowledge transfer. Journal of Consumer Research, 24: 266-284.
Griskevicius, V., Shiota, M.N., & Neufeld, S.L. 2010. Influence of different positive emotions on persuasion processing: a functional evolutionary approach. American
Psychological Association, 10(2): 190-206.
Henry, W.A. 1980. The effect of information-processing ability on processing accuracy.
Journal of Consumer Research, 7: 42-48.
History.com. (http://www.history.com/this-day-in-history/kennedy-and-nixon-square-off-in-a- televised-presidential-debate). Accessed on the 21st of June, 2016.
Howard, D.J., & Gengler, C. 2001. Emotional contagion effects on product attitudes. Journal
of Consumer Research, 28: 189-201.
Hui, M.K., Zhao, X., Fan, X., & Au, K. 2004. When does the service process matter? A test of two competing theories. Journal of Consumer Research, 31: 465-475.
Johnson, E.J., & Russo, J.E. 1984. Product familiarity and learning new information. Journal
of Consumer Research, 11: 542-550.
29
Joye, Y., & Dewitte, S. 2016. Up speeds you down. Awe-evoking monumental buildings trigger behavioral and perceived freezing. Journal of Environmental Psychology, 47: 112-125.
Keltner, D., & Haidt, J. 2003. Approaching awe, a moral, spiritual, and aesthetic emotion. Cognition and Emotion, 17(2): 297-314.
Maheswaran, D., & Sternthal, B. 1990. The effects of knowledge, motivation, and type of message on ad processing and product judgments. Journal of Consumer Research,
17: 66-73.
Marks, L.J., & Kamins, M.A. 1988. The use of product sampling and advertising: effects of sequence of exposure and degree of advertising claim exaggeration on consumers’ belief strength, belief confidence, and attitudes. Journal of Marketing Research, 25: 266-281.
Piff, P.K., Dietze, P., Feinberg, M., Stancato, D.M., & Keltner, D. 2015. Awe, the small self and prosocial behavior. American Psycholocial Association, 108(6): 883-899.
Rotfeld, H.J., & Rotzoll, K.B. 1980. Is advertising puffery believed? Journal of Advertising,
9(3): 16-21.
Schroder, H.M., Driver, M.J. & Strenfert, S. 1967. Human Information Processing. New York, NY: Holt, Rinehart & Winston. Inc.
Sperber, D. 2010. The guru effect. Review of Philosophy and Psychology, 1: 583-592. Schul, Y., & Bernstein, E. 1985. When discounting fails: conditions under which individuals use discredited information in making a judgement. Journal of Personality and Social
Psychology, 49: 895-903.
Wyer, R.S., & Budesheim, T.L. 1987. Person memory and judgments: the impact of information that one is told to disregard. Journal of Personality and Social
