You can't judge a book by its cover, or can you? : the effects of anthropomorphism on health consumption behaviour, as mediated by ideal congruity

YOU CAN’T JUDGE A BOOK

BY ITS COVER, OR CAN YOU?

The effects of anthropomorphism on health consumption

behaviour, as mediated by ideal congruity

Tabitha Richtersz, 11644648

Frauke Mattison Thompson MSc. Business Administration – Marketing track Amsterdam Business School June 22nd 2018 Final thesis 9406 words

2

Statement of Originality

This document is written by Student Tabitha Richtersz who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document are original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

3 Table of contents

1. Abstract ……….. 4

2. Introduction ... 5

3. Literature Review ………. 7

Anthropomorphism: what is it and why do we do it? ………. 7

Ideal congruity ……… 9

Research gaps ………. 10

4. Theoretical framework ………. 12

5. Methods ……….. 17

Pilot study ……… 17

Participants and procedures ……… 19

Measures ……….. 21

Analysis ……… 23

6. Results ………. 23

7. Discussion and conclusion ………. 26

Limitations and directions for future research ……… 30

Theoretical and practical implications ……… 32

Appendices ………. 35

Appendix A Manipulation of independent variable ………... 35

Appendix B Survey questions ………... 36

Dutch ……… 36

Original items in English ………. 39

Appendix C Pre-test survey ……….. 41

4 1. Abstract

Despite the frequent use of anthropomorphism in marketing, little is known about the effects of anthropomorphic packaging elements on health consumption behaviour such as consumption intention, purchase intention, and product evaluation. Prior research found positive effects of anthropomorphism on consumer behaviours such as loyalty, purchase, product evaluation, and brand preference, but failed to consider effects of anthropomorphism in the context of packaging and healthy food consumption, and failed to consider the mediating role of ideal congruity. This study therefore examined the effects of anthropomorphic packagings on health consumption as mediated by ideal congruity. In doing so, this study contributes to literature in three ways: firstly, by examining the mediating role of ideal congruity; secondly, by looking at packaging elements; and thirdly, by using only healthy foods. An experimental survey was conducted in which participants were exposed to either an anthropomorphic or neutrally designed granola packaging. Results demonstrated that anthropomorphism did not increase health consumption behaviour and that this relationship was not mediated by ideal congruity. However, ideal congruity did increase health consumption behaviour. More research is necessary to further establish the boundary conditions of these effects and to fully understand the underlying mechanisms.

5 2. Introduction

From cars with faces to talking cows in dairy commercials, companies increasingly market their brands and products through anthropomorphism, targeting not only children, but also adults (Epley, Waytz & Cacioppo, 2007; Nenkov & Scott, 2014; Lee, Chen, Chang & Huang, 2017). Marketing literature has shown that anthropomorphism, the attribution of human characteristics and behaviours to non-human entities, has become a key strategy of increasing product likeability and loyalty and establishing stronger (emotional) relationships in many different categories (toys, household, toiletries, groceries (Triantos, Plakoyiannaki, Outra & Petridis, 2016)) (Aggarwal & McGill, 2007; Nenkov & Scott, 2014 Rauschnabel & Ahuvia, 2014; Lee et al., 2017). Think for example, of salt and pepper shakers in the shape of a cuddling couple, cars with eyelashes on the headlights, or talking M&M’s that don’t want to be eaten – anthropomorphism is an indispensable part of marketing. Marketers encourage consumers to perceive products in human terms, for example by describing the product in first rather than third person, by shaping the product in a way that resembles physical human features, such as a Coca Cola bottle, by ascribing feelings to their products, or by designing an anthropomorphic spokes character to represent the brand. In fact, 54% of FMCG products contain anthropomorphic elements, whether this be in the name of the brand, brand logo, slogans, spokes characters or in packaging elements (Triantos et al., 2016).

Although there is some research on the effects of anthropomorphism on consumer behaviours such as product evaluation (Aggarwal & McGill, 2012), brand preference (Wan, Chen & Jin, 2017), consumption (Nenkov & Scott, 2014; Lee et al., 2017), purchase behaviour (Chandler & Schwartz, 2010) and loyalty (Guido & Peluso, 2015), there is a surprising lack of studies on the effects of anthropomorphism on the consumption of food, specifically on how anthropomorphism can affect the consumption of healthy foods. Considering we live in a society where obesity numbers are staggeringly high and continue to grow, this is an interesting

6

and insufficiently studied area of research. In order to fill this gap, the present study is primarily focused on the question of whether anthropomorphism can affect the consumption of healthy foods, and if so, how? Specifically, how does presenting products in an anthropomorphic manner affect the evaluation of healthy food products, and moreover, how does it affect consumption behaviour, such as purchase intention and consumption intention?

Previous research has studied anthropomorphized food shape (Lee et al., 2017) and anthropomorphized spokes characters (Kraak & Story, 2015), but not packaging elements. The present study will therefore address the effects of anthropomorphized packaging on health consumption behaviour. One key contribution of this study is to test the mediating effect of ideal congruity, the perceived similarity between the product’s personality and consumers’ ideal self-concept, on the relationship between anthropomorphism and health consumption behaviour. When it comes to health consumption, one would expect consumers to show purchase and consumption behaviour in line with their current lifestyle. However, studies have shown consumers tend to purchase products that are congruent with their ideal selves (Sirgy, 1985; Guido & Peluso, 2015) and that anthropomorphism strengthens this effect (Chandler & Schwartz, 2010). Investigating the relationship between ideal congruity and consumption behaviour is important, considering most images of people in advertising are idealized versions of the target audience.

The present study considers healthy consumption behaviour to consist of purchase intention, consumption intention and product evaluation. The relationship between these variables and anthropomorphism, as mediated by ideal congruity, was studied in an experimental survey with granola packagings. The results did not confirm any relationship between anthropomorphism and healthy consumption behaviour, however, this study did provide evidence that ideal congruity positively influences health consumption. In doing so, this study contributes to existing literature on anthropomorphism and ideal congruity and

7

provides a framework for understanding the conditions under which healthy consumption behaviour is influenced by anthropomorphic packaging elements and ideal congruity.

3. Literature review

Anthropomorphism: what is it and why do we do it?

Anthropomorphism is a concept widely used in many disciplines and fields and can be defined as the innate human tendency to attribute humanlike characteristics, such as emotions, ambitions, cognitions and intentions, to non-human entities, e.g. animals or technology, which results in a humanlike treatment of those non-human entities (Aggarwal & McGill, 2012; Delbaere, McQuarrie & Phillips, 2011; Guthrie, 1993; Triantos et al., 2016). According to Waytz, Cacioppo and Epley (2010), humans tend to anthropomorphize in two ways. Firstly, they might attribute humanlike physical features to a non-human entity, e.g. seeing a face in the clouds. Secondly, they might attribute mental capacities to non-human entities, e.g. emotions like joy or sadness, consciousness, intelligence and intentions. People may, for example, scold a technological device when it isn’t working as desired. Some even go as far as to name objects that are dear to them, for example their cars. Three cognitive and motivational psychological determinants explain when people are more likely to anthropomorphize: elicited agent knowledge, sociality, and effectance. One explanation is that the act of anthropomorphism facilitates people in making sense of the world around them (Aggarwal & McGill, 2012; Gameiro, 2017). Humans apply their existing knowledge of human behaviour and other human characteristics and ascribe this to other entities in order to explain their unfamiliarity; a cognitive process called elicited agent knowledge (Guthrie, 1993; Epley et al, 2007; Triantos et al., 2016). Children, more so than adults, are typically inclined to anthropomorphize that which they are unfamiliar with, as they have limited knowledge of their environment (Epley et al., 2007). This forces them to apply their knowledge of human characteristics to unfamiliar entities they regard as akin to human beings (Triantos et al., 2016). A second explanation, a cognitive

8

construct called sociality (Epley et al., 2007), focuses on social reasons for anthropomorphism. Humans have an innate need for social interaction, and when they perceive this need as deprived, they will seek alternative means to satisfy the lack thereof (Baumeister & Leary, 1995; Gardner, Pickett, Jefferis, & Knowles, 2005). Anthropomorphizing functions as an alternative way of satisfying this need for social connection by providing humans with the comfort of social relationships (Guthrie, 1993). This view is supported by research from Chen, Wen Wan, and Levy (2016), who found that socially excluded consumers expressed a greater preference for anthropomorphized brands. In other words, anthropomorphizing brands can help socially excluded individuals fulfil the deprived need for social affiliation by building a relationship in much the same ways in which they build interpersonal relationships. Lastly, the tendency for anthropomorphism can be explained by the phenomenon of effectance motivation: the motivation and desire to exercise control over and influence one’s environment, thus ensuring effective interaction (Epley et al., 2007; Gameiro, 2017). This includes interaction with human and non-human entities (Triantos et al., 2016). In order to ensure effective communication with non-human entities, “attributing human characteristics and motivations to non-human agents increases familiarity and confidence in predictions of these agents’ behaviours in the future” (Triantos et al., 2016, p.7). In conclusion, people will anthropomorphize when anthropomorphic knowledge is readily available in their minds, when motivated to exercise control over a situation and interact effectively, and in absence of a sense of social interaction with other humans (Epley et al., 2007).

There are three forms of anthropomorphism. In literal anthropomorphism, objects are literally shaped like humans or presented as acting human, and people literally believe these entities to be an actual person (Guthrie, 1993). This occurs, for example, when cartoon animals wear clothes and talk. Seeing a face in a building is an example of accidental anthropomorphism: when individuals consider the anthropomorphism to be accidental

9

(Guthrie, 1993; Aggarwal & McGill, 2012). Lastly, and most relevant for the current study, partial anthropomorphism occurs when people attribute human features to objects, but don’t perceive the entity as a whole as human (Guthrie, 1993). Partial anthropomorphism often occurs in a product context, when consumers attribute human characteristics to products without going so far as to believe the product to actually be a person (Guthrie, 1993; Aggarwal & McGill, 2012).

Ideal congruity

Ideal congruity originates in self-congruence theory, which states that “individual behaviour is regulated by an individual’s perceived similarities (or dissimilarities) of environmental conditions to self-image” (Astakhova, Swimberghe & Wooldridge, 2017). Consequently, the theory postulates that the congruity between the product-user image and a consumer’s self-image, in part, determines consumer behaviour (Sirgy et al., 1997). When consumers perceive the product image to be a match with their own self-image, self-congruity is high, and vice versa. The concept is a multi-dimensional construct, consisting of multiple self-perspectives: the actual self-image, the social self-image, the ideal self-image and the ideal social self-image. The actual self-image is considered a person’s perception of who they actually, realistically are, whereas the ideal self-image is an individual’s perception of the characteristics they ideally would like to possess. The social self-image is how an individual thinks others perceive them, whereas the ideal social self-image is how an individual would ideally like others to perceive them (Sirgy, 1985; Sirgy et al., 1997, Astakhova et al., 2017; Huber, Eisele & Meyer, 2018). Self-image/product-image congruity has thus been treated as multidimensional as well, such that congruity between the actual self-image and product personality is referred to as self congruity, congruity between the ideal self-image and product personality is referred to as ideal congruity, etc. The current research will address ideal congruity.

10

In creating ads, products and brands, marketers try to appeal to consumers’ actual and ideal selves, for example by picturing idealized versions of the target audience or by showing how consumers can become a better version of themselves by using product or brand X. The desired outcome of appealing to the actual or ideal self-image is to create a connection and establish an emotional bond between consumer and brand (Astakhova et al., 2017; Huber et al., 2018). Brands are important to the self-concept because they help individuals express and define their identity. Consumers consequently prefer brands that match their actual or ideal selves because it fulfils their need to maintain and enhance their self-esteem (Astakhova et al., 2017).

Research gaps

Although some studies have demonstrated that anthropomorphism increases consumption, they mainly focus on children’s consumption behaviours, not adults’ consumption behaviours (Roberto et al., 2010; Lapierre et al., 2011; Letona et al., 2014; Kraak & Story, 2015). Moreover, these studies do not take into account whether anthropomorphism has an effect on the consumption of healthy foods. As Lee et al. (2017) suggested for future research, this study focuses on healthy foods only. The few existing studies on the consumption of anthropomorphized foods provide inconsistent results. Recent research by Nenkov & Scott (2014), for example, studied the effects of anthropomorphic products on indulgent behaviour, and found that the whimsical cuteness of the product (an ice cream scoop shaped like a miniature lady) increased consumption of indulgent food (ice cream). However, because an ice cream scoop is a functional product belonging to a utilitarian product category, the authors considered this effect to extend to utilitarian consumption. Research from Lee, Chen, Chang, and Huang (2017) contradicts these findings. They found that anthropomorphized foods, such as cookies or jelly shaped like Hello Kitty, increased hedonic, but not utilitarian consumption. However, this effect only occurred when the motivation of the consumption, either hedonic or

11

utilitarian, matched with the food type, anthropomorphized or neutral. In other words, participants primed with a utilitarian health motivation ate more chocolates only when the chocolates were neutral rather than anthropomorphized, because the neutral shape fits with the expectations consumers have of healthy food. This is linked to the theory of schema congruity: a fit between the healthy schema and the product leads to satisfaction, whereas a non-fit between a healthy schema and the product leads to discomfort and thus to a lower evaluation or possible lower wish to consume (Aggarwal & McGill, 2007).

Nenkov and Scott (2014) and Lee et al. (2017) follow the same reasoning: a product’s appearance puts the consumer in a certain mind set, which then influences their consequent behaviour to match that mind set. In other words, an anthropomorphized product puts the consumer in a fun and playful mind set, which influences the consumer to act in accordance with that mind set and engage in indulgent consumption. Aggarwal & McGill (2012, p. 307) support this notion, saying that “brand exposure may elicit automatic behavioural effects consistent with the brand image”. They pose the idea that consumers acting in a consistent manner with the brand image subsequent to being exposed to it, are motivated by their desire to achieve the main characteristic communicated by the brand. Thus, when exposed to a brand with a healthy image, consumers will assimilate their behaviour to that image and, for example, make healthier food choices. However, they found that these effects only occur when a brand is anthropomorphized. Anthropomorphized brands thus lead consumers to act in accordance with the brand image.

Following the reasoning of Aggarwal & McGill (2012), exposure to an anthropomorphized product with a healthy image should increase consumers’ healthy consumption behaviour. Further research states that not only a brand’s image, but also its personality has an impact on how consumers interact with a brand or product: the more consumers perceive their own personality to be similar to a brand’s personality, the more

12

willing they are to purchase that product (Sirgy, 1985; Sirgy et al., 1997), the stronger their motivation to re-buy and the higher their loyalty (Guido & Peluso, 2015). However, these studies don’t consider the fact that most images of people in advertisements are idealized versions of the target audience. It would therefore make sense if not people’s actual, but their ideal personality impacts their consumption and purchase behaviour. In other words, a match between consumers’ ideal personality and the personality of a brand should increase consumption and purchase behaviour. Considering anthropomorphism facilitates the recognition of a personality in a product, it is surprising that extant literature fails to link anthropomorphism to ideal congruity. The current study attempts to fill this gap by considering the mediating role of ideal congruity in the relationship between anthropomorphism and health consumption behaviour.

Research has shown that most consumer decisions are made in store, based on incoming stimuli (Briggs, 2016). The first stimulus consumers get from a product in store is the packaging. Packaging is the first chance to communicate a message to the consumer. It would therefore make sense packaging has some effect on consumers’ decisions and behaviours, and research has confirmed this hypothesis (Triantos et al., 2016; Gameiro, 2017). No studies to my knowledge have researched the effects of anthropomorphic packagings on consumer behaviour. The present study is therefore one of the first to consider effects of anthropomorphic packagings on health consumption behaviour.

4. Theoretical framework

Anthropomorphism is widely used in marketing communications and marketing research has documented many positive effects of anthropomorphism on consumer behaviour. For example, anthropomorphism increases product likeability, purchase frequency, consumption and loyalty (Aggarwal & McGill, 2007; Delbaere, McQuarrie & Phillips, 2011; Rauschnabel & Ahuvia, 2014; Lee et al., 2017). Furthermore, anthropomorphism elicits positive emotions and

13

establishes positive attitudes towards a brand, resulting in brand and product preference (Delbaere, McQuarrie & Phillips, 2011). In order to understand what causes these effects, it is important to look at the way consumers interact with anthropomorphic products.

People process information by thinking in categories or schemas. When seeing a product, a schema is activated in consumer’s minds and consumers evaluate a product based on the perceived fit between the schema and the product. When interacting with an anthropomorphized product, consumers call to mind the schema that fits with the type of person represented by the product’s features. Evaluation of the product then depends on the perceived fit between its features and the schema. A good fit between schema and product creates a feeling of satisfaction, which in turn leads to a better product evaluation (Aggarwal & McGill, 2007). This would mean that a healthy anthropomorphized product should bring to mind a schema for a healthy type of person, which, when the features of the product match the schema, would result in a positive evaluation.

In social psychology, it is a common known fact that people tend to imitate behaviour from others around them (see for example Bargh, Chen & Burrows, 1996). Aggarwal & McGill (2012) propose that this effect extends to brands and products when they are perceived as people. In other words, consumers apply the same social rules and schemas used to interact with people in interaction with anthropomorphic entities. Further research supports this notion. For example, Aggarwal and McGill (2007) showed that consumers applied the social norm of smiling when you meet someone new to cars: when a car seemed to smile, it was evaluated more positively then when the car seemed to frown. Likewise, Kim and McGill (2011) showed that consumers who felt powerful towards an anthropomorphized slot machine, as opposed to consumers who felt powerless, felt a higher sense of control towards the slot machine and thus experienced a lower sense of risk. Additionally, Wen, Chen and Jin (2017) showed that consumers apply the “what is beautiful is good” principle to anthropomorphized brands, which

14

in turn increases their reliance on appearance attributes in judging a product and thus their preference for products with a superior appearance. Consumers favoured and were more likely to choose products with better appearance, superior attractiveness and with a more appealing design. Their actual purchase behaviour matched these preferences.

The above studies show that consumers treat anthropomorphic brands as they would treat humans. It would make sense if consumers also formed relationships with brands and products the way they would with other humans. Further research indicates that anthropomorphism indeed facilitates the establishing of relationships with brands and products. Anthropomorphism helps consumers to perceive a brand as more familiar, which establishes an emotional connection to a brand. This emotional attachment leads to brands being able to get close to their consumers, thus fostering strong relationships, resulting in brand love, a passionate desire to use, and an overall positive attitude (Rauschnabel & Ahuvia, 2014).

In light of the present study, these studies support the expectation that an anthropomorphized healthy product would call to mind the schema of a healthy person, which consequently would influence consumers to act in accordance with this schema and act healthily. The hypothesis is therefore as follows:

H1: there exists a positive relationship between anthropomorphic food packaging and health consumption behaviour, such that when exposed to anthropomorphic packaging, consumers will have a higher consumption intention, more positive product evaluation and higher purchase intention.

Products and brands are often perceived as having a personality. This personality is shaped by many aspects of marketing communication, such as advertising, packaging, price, endorsers or typical users (Chandler & Schwarz, 2010; Delbaere, McQuarrie & Phillips, 2011).

15

Anthropomorphism facilitates the recognition of personality traits in a product (Aggarwal & McGill, 2012). When a product is perceived as being human or humanlike, it is easier for consumers to attribute personality characteristics to that product. A product’s personality interacts with a consumer’s self-concept in such a way that consumers seem to prefer products that are similar to their own actual or ideal personality – in other words, anthropomorphism facilitates seeing personalities in products and thus facilitates perceiving the similarities between consumers’ ideal personality and the product’s personality. Thus, anthropomorphism strengthens ideal congruity (Aggarwal & McGill, 2012; Guido & Peluso, 2015; Mourey, Olsen & Yoon, 2017).

H2: anthropomorphism has a positive effect on ideal congruity, such that the perceived similarity between a product and a consumer’s ideal self is higher when the product is anthropomorphized.

Individuals use products to express and define their identity. Academic research has demonstrated that consumers favour products that facilitate the expression of their actual or ideal selves (Guido & Peluso, 2015). Consequently, brands market their products by communicating that their products help consumers get closer to becoming who they want to be. Consumers prefer products that are congruent with their ideal self because it helps them to fulfil their need to maintain and enhance their self-esteem. The higher ideal congruity, the better consumers evaluate a product and the higher their wish to purchase a product, because of the self-esteem boost the product provides (Astakhova, Swimberge & Wooldridge, 2017). Furthermore, recognition of aspects of one’s ideal self leads to establishing a deeper emotional connection with brands and products, that resemble the relationships formed with real people. This in turn increases brand attachment and loyalty, results in brand and product preference,

16

and increases sales (Rauschnabel & Ahuvia, 2014; Gameiro, 2017; Huber, Eisele & Meyer, 2017). It is thus likely that consumers whose ideal self is active, fit, and consumes mostly healthy foods will feel an emotional connection with products that match these personality characteristics, and thus show higher intention to consume and purchase.

H3: consumers experiencing a higher degree of ideal congruity will score higher on consumption intention, higher purchase intention, and show a more positive product evaluation.

Lastly, since anthropomorphism has been found to strengthen the effects of ideal congruity (Chandler & Schwarz, 2010; Aggarwal & McGill, 2012), the present study will test ideal congruity as a mediator.

H4: the positive relationship between anthropomorphism and health consumption behaviour is mediated by ideal congruity, such that when exposed to anthropomorphic packaging, consumers will have higher congruity, resulting in higher purchase and consumption intention and in a more positive product evaluation.

Studies on anthropomorphism and food to date have mainly considered anthropomorphic elements in food shape (Lee et al., 2017) and anthropomorphized spokes characters (Kraak & Story, 2015), but not anthropomorphic packaging. Previous studies have demonstrated effects of packaging on consumer behaviour (Triantos et al., 2016). The current research will therefore study the effects of anthropomorphic packagings on ideal congruity and health consumption behaviour. This study, as suggested by Lee et al. (2017) for future research, will only consider

17

the effects of truly healthy foods. The food category picked for the experiment, therefore, is granola, a food that is generally perceived as healthy.

Figure 1. Conceptual model

5. Methods Pilot study

A pilot study was conducted with three test packagings (see appendix A) that varied in their level of anthropomorphism: one packaging was designed to look neutral, another packaging was designed to look semi-anthropomorphic, and a third packaging was designed to look completely anthropomorphic. The details of the packagings can be found in the section Measures below. The product chosen to test the effects of anthropomorphism on health consumption was granola, as granola is a food that is typically believed to be healthy. The goal of the pre-test was to find out if the manipulations of the packagings were successful and to test the extent to which the packagings were perceived as anthropomorphic or neutral. A second goal of the pre-test was to compose a list of characteristics typical for a healthy type of person. These characteristics were consequently used in the survey to measure the personality of each granola box and to test whether the granola packagings were perceived as fitting with a healthy schema. Additionally, these health characteristics were used to test the extent to which participants’ own ideal self-image was healthy and to test the level of ideal congruity.

18

Respondents were recruited in the personal social circle of the author in order to ensure a timely response. The pilot test was taken by N = 14 respondents (Nfemale = 9, Nmale = 5; Nhigh school= 1, NHBO= 3, NWO= 10), of which n = 10 were aged 18-25 and N = 4 were aged 26-35. In

an open question, respondents were asked to bring to mind a typically healthy person and write down 10 characteristics that typically described this person. Of all the characteristics mentioned by the respondents, the 7 that were mentioned most were added to the survey. Next, respondents rated each of the three packagings on human characteristics, answering items such as “this packaging seems to have a face” and “this packaging is humanlike”. A short description of anthropomorphism was provided to respondents, after which they rated the packagings on their level of anthropomorphism (“this packaging is anthropomorphic”). All questions were answered on a 7-point Likert scale ranging from completely disagree to completely agree. All items used in the pre-test were based on scales from Hellén & Sääksjärvi (2013) and Guido & Peluso (2015) and had a reliability of respectively α = .92 and ρ = .81. The items were translated to Dutch by the author and were checked by an independent researcher. The entire pre-test survey can be found in appendix C in both Dutch and English.

A one-way repeated measures ANOVA was conducted with packaging as independent variable and level of anthropomorphism and human likeness as dependent variables. Results showed that the anthropomorphically designed packaging was indeed perceived as most anthropomorphic (M = 5.77, SE = 1.54). The somewhat anthropomorphic packaging was indeed perceived as being in the middle (M = 3.38, SE = 1.02). The non-anthropomorphic packaging was indeed perceived as least anthropomorphic (M = 1.31, SE = .48). This was a significant difference, F (2, 64.795) = 37.858, p < .0001. Furthermore, the anthropomorphic packaging was also perceived as most humanlike (M = 4.11, SE = .84) and the non-anthropomorphic packaging was perceived as least humanlike (M = 1.44, SE = .47), whereas the somewhat anthropomorphic packaging was perceived as being in the middle (M = 2.59, SE = .13). This

19

was a significant difference, F(2, 8.724) = 64.281, p < 0,001). Based on these findings, the anthropomorphically and neutrally designed granola boxes were used as measure of anthropomorphism in the survey.

Participants and procedure

The sample consisted of 104 participants ranging in age from 18 to 51 (M = 24.91; SD = 5.16) (Nhigh school= 16, NMBO = 10, NHBO= 27, NWO= 51). 72 participants were female (69.2%) and 32

participants were male (30.8%). All respondents were Dutch speaking, as the survey was taken in Dutch. The sample was collected through Facebook. The survey was distributed both on the author’s own personal page as well as in a Facebook group specifically meant for graduating students who were looking for respondents for their surveys. The purpose of the group was to fill in others’ surveys in return for a response on your own survey. This manner of distribution was chosen because it was a convenient way of ensuring as many responses as possible within the set time frame. Participation was voluntary and no compensation was provided.

Participants were randomly assigned to one of two conditions. In condition 1, participants were shown a picture of the anthropomorphized granola packaging. In condition 2, participants were shown a picture of the non-anthropomorphized, neutral granola packaging (for a detailed description and visuals of the packagings, see Appendix A and the section Measures below).

Before starting the survey, participants were told the objective of the study was to examine elements of product packagings and consumers’ reactions to it. Additionally, they were informed about the anonymity and confidentiality regarding all data and agreed that participation was voluntary. After answering demographic questions such as age, gender, and education level, participants were shown the granola packaging and were instructed to take a minute to have a detailed look at the packaging and really think about what they were seeing. To make sure respondents were continuously exposed to the independent variable, the picture

20

of the granola box was displayed on all pages containing questions regarding the dependent variables and the mediator. After having carefully looked at the packaging, participants answered questions about product evaluation (for example, I would recommend this product to my friends), purchase intention (for example, it is likely I will purchase this product in the near future), and consumption intention (for example, I am planning on consuming this product). To be able to judge the similarity between the product’s personality and their own ideal personality, it was important that participants first formed an idea of both these constructs. To ensure participants could make the comparison between these personalities, they were asked questions about their current actual self-image, their ideal self-image, and the personality of the product they saw. For all these questions, participants were instructed to think about the product as if it were a person. The scales for self-image, ideal self-image, and product personality contained items such as I am active/healthy, this product is energetic/balanced and the personality of this product is consistent with how I would ideally like to be (my ideal self). To measure ideal congruity, participants were asked the extent to which they thought their own ideal personality converged with that of the granola packaging. Finally, participants were asked to answer questions about the control variables health consciousness and healthy lifestyle, with items such as I think it is important to know how to eat healthy and I avoid the eating of processed foods. People who have a high level of knowledge on what is healthy and what is unhealthy for the body considering food, and people who lead an active healthy lifestyle, might evaluate a product as more attractive because of its health properties. Additionally, they might be more inclined to purchase or consume a product due to its health properties. To prevent a better evaluation and a higher purchase- and consumption intention due to factors other than the independent variable, the control variables health consciousness and healthy lifestyle were added to the survey. To end the survey, respondents were thoroughly thanked for their participation.

21

All questions were answered on a 7-point Likert scale ranging from completely disagree to completely agree. All items were adapted from English scales. The items were translated to Dutch by the author and one other independent researcher. The translations were then checked against the original English version by another independent researcher.

Measures

This section contains an overview of the most important measures used for the experiment. An extended detailed overview of all measures and items used in this study can be found in appendix B. This includes both the original items in English as well as the Dutch translations.

Anthropomorphism. Two versions of a granola box were specifically designed for the purposes of this study. The use of existing brands was avoided in order to ensure respondents did not have any associations with the manipulation prior to the study. Visuals of the granola boxes can be found in appendix A. The anthropomorphic granola box had several features that made it humanlike. Firstly, it was given an anthropomorphic name, Ola Granola, which makes it seem like the product is greeting the consumer. Secondly, the box was decorated with an anthropomorphized picture of a cactus. The cactus had eyes, a moustache, and was wearing a sombrero hat. Additionally, the cactus’ branches looked like two arms. The packaging was made to look colourful, with bright and vivid colours, to communicate fun and playfulness. By contrast, the neutral looking granola box was made to look anything but anthropomorphic. It was given a neutral name, All Natural Granola, and had a small window in which the consumer could see the granola inside the packaging. The colours on the non-anthropomorphic granola box were neutral brown and green. Both packagings included a line at the bottom saying “made with the best biological nuts, seeds, grains, and honey” to communicate the healthiness of the product.

For the pre-test, a third granola box was designed that was semi-anthropomorphic. It had a semi-anthropomorphic name, Grrranola, that seemed to make a noise. Additionally, it had

22

a window in the shape of a bear on it, with a speech balloon that said “enough nutrients for a hibernation”. This speech balloon made the image of the bear somewhat anthropomorphic, but not quite as anthropomorphic as the Ola Granola packaging.

Health consumption. The dependent variable, health consumption behaviour, was measured with three constructs: product evaluation, purchase intention and consumption intention. Product evaluation was measured using a 6 item scale based on Chan, Elaine & Anirban Mukhopadhyay (2010) with an alpha of α = .912 (e.g., I like this packaging, I have a positive attitude towards this product). A reliability analysis revealed that one item on this scale had a corrected item total correlation of below .3 (this product is tasty, r = .202). This item was therefore deleted from the product evaluation scale. Purchase intention was measured using a 3 item scale based on Yoo, Boonghee & Donthu (2001) and had an alpha of α = .928 (e.g., I intend to purchase, I expect to purchase). Consumption intention was measured using a 3 item scale based on Seo, Kim, Oh & Yun (2013) and had an alpha of α = .910 (e.g., I intend to consume, I expect to consume). Health consumption was calculated as consumption intention plus purchase intention plus product evaluation divided by three. All items were answered on a 7 point Likert scale ranging from completely disagree to completely agree.

Ideal congruity. The mediator, ideal congruity, was measured using a 4 item scale based on Malär, Krohmer, Hoyer & Nyffenegger (2011) and Guido & Peluso (2015) and had an alpha of α = .937 (e.g., I see a part of how I would like to be in this product). Additionally, actual and ideal self-image were measured using a one item scale from Landon (1974), which consisted of the question: please indicate how much you feel these characteristics describe you/how you ideally would like to be. The characteristics found in the pilot test were used as items for this question.

Control variables. Health consciousness was measured using a 10 item scale from Oude Ophuis (1989) as described in Chen (2009) and had an alpha of α = .829. Healthy lifestyle was measured

23

using a 4 item scale from Oude Ophuis (1989) as described in Chen (2009). A reliability analysis revealed that two items of the healthy lifestyle scale had a corrected item total correlation of below .3 (I work out regularly, r = .179; I often eat vegetables and fruits, r = .277). These two items were therefore removed from the scale, after which the scale had an alpha of α = .906.

Analysis

The data was analysed in SPSS version 24 (2016). To test the effects of anthropomorphism on consumers’ health consumption, as mediated by ideal congruity, a mediation regression analysis was run in PROCESS v3.0 (Hayes, 2017) with anthropomorphism as independent variable, health consumption as dependent variable, and ideal congruity as mediator. The data was tested to see if no assumptions were violated. Histograms, P-P plots and Q-Q plots showed that there were no outliers and that the data was normally distributed. Levene’s test revealed that the assumption of homogeneity of variances was not violated (F(1, 102) = ,910, p = ,342). All assumptions for an independent samples t-test were met. A total of n = 14 participants were excluded from analysis due to incomplete answers.

5. Results

Preliminary analysis was conducted on the demographic variables and revealed no significant differences between the two experimental groups in gender and education (tgender (102) = -.714,

p = .157; teducation (102) = -.429, p = .191). However, the age difference between the two

experimental groups (Manthropomorphic = 24.02, SE = 6.30; Mnon-anthropomorphic = 25.84, SE = 3.58)

was significant (t (102) = 1.823, p = .012). Since this is a very small effect (r = .18), the difference is not regarded problematic, however, the variable was still added to the mediation model as a control variable. Furthermore, the preliminary analysis did not reveal any significant differences between the two experimental groups in the control variables health consciousness (Manthropomorphism = 4.48, SE = .74; Mnon-anthropomorphism = 4.61, SE = .55), t (100) = -.983, p = .112,

24

and healthy lifestyle (Manthropomorphism = 4.27, SE = .86; Mnon-anthropomorphism = 4.46, SE = .99), t

(99) = -1.060, p = .406. Lastly, a preliminary analysis was run to check for differences between participants scores on self-image (Manthropomorphic = 5.13, SE = .82; Mnon-anthropomorphic = 4.98, SE

= .93) , ideal self-image (Manthropomorphic = 6.16, SE = .64; Mnon-anthropomorphic = 6.05, SE = .74) ,

and product personality (Manthropomorphic = 4.39, SE = 1.12; Mnon-anthropomorphic = 5.03, SE = 1.21).

No significant differences were found for any of these variables (tself-image (101) = -.842, p =

.572; tideal self-image (96) = -.814, p = .420; tproduct personality (102) = 2.820, p = .310).

A simple mediation analysis using ordinary least squares path analysis was conducted using SPSS macro PROCESS 3.0 (Hayes, 2017) with anthropomorphism as independent variable, health consumption as dependent variable, and ideal congruity as mediator. Additionally, since a significant difference in age between the two groups was found, age was included in the model as a covariate. Results on the direct effect – the effect of anthropomorphism on health consumption when participants experience the same level of ideal congruity – showed that participants in the anthropomorphic condition scored higher on health consumption than participants in the non-anthropomorphic condition (Mnon-anthropomorphic = 10.05,

SE = 2.48; Manthropomorphic = 9.29, SE = 2.85), however, this effect was not significant, b = .34, t

=. 84, p =. 40. Therefore, hypothesis 1 has been rejected. This model explained 49% of the variance in health consumption behaviour, R2 = .49. Furthermore, the results showed that anthropomorphism significantly negatively predicted ideal congruity, meaning that participants in the anthropomorphic condition, on average, scored .89 lower on ideal congruity than did participants in the non-anthropomorphic condition, b = -.89, t = -3.08, p = .003. Anthropomorphism explained 9.5% of the variance in ideal congruity, R2 = .095. This means hypothesis 2 was rejected. Independent of condition, ideal congruity significantly predicted health consumption, in such a way that an increase in ideal congruity lead to an increase in product evaluation, consumption intention and purchase intention, b = 1.29, t = 9.53, p <.001.

25

These results are in support of hypothesis 3. The total effect, the effect of anthropomorphism on health consumption without the mediator ideal congruity in the model, showed that anthropomorphism does not significantly predict health consumption: participants in the condition reported a lower health consumption than participants in the non-anthropomorphic condition, b = -.81, t = -1.49, p = .14. In this model, only 2% of the variance in health consumption behaviour was explained by anthropomorphism, R2 = .02. Lastly, a bias-corrected bootstrap confidence interval based on 5,000 bootstrap samples revealed a significant indirect effect of anthropomorphism on health consumption through ideal congruity, b = -1.15, BCa CI [-1.967, -.393]. However, since the direct effect was not significant, ideal congruity was not a mediator of the relationship between anthropomorphism and health consumption. Hypothesis 4 was therefore rejected. The confidence interval for the indirect effect was negative, meaning that the more positive ideal congruity, the more negative the effects of anthropomorphism on health consumption, or conversely, the more negative ideal congruity, the more positive the effects of anthropomorphism on health consumption.

For further understanding, independent samples t-tests were performed with anthropomorphism as independent variable and product evaluation, consumption intention and purchase intention as dependent variables. The results show an increase in both purchase and consumption

26

intention, but these results were not significant. Participants rated the non-anthropomorphic packaging as more attractive (M = 4.52, SE = 1.13) than the anthropomorphic packaging (M = 4.28, SE = 1.10). This difference was not significant t (102) = 1.061, p = .291. Participants reported a higher purchase intention for the non-anthropomorphic granola (M = 3.42, SE = 1.48) than for the anthropomorphic granola (M = 3.27, SE = 1.51). This difference was not significant t (102) = .503, p = .616. Lastly, participants had a higher consumption intention for the non-anthropomorphic granola (M = 4.39, SE = 1.23) than for the non-anthropomorphic granola (M = 3.91, SE = 1.45). This was a non-significant difference t (102) = 1.815, p = 0.072.

6. Discussion and conclusion

Although previous studies demonstrated an increase in consumption and purchase and a higher product evaluation as a result of anthropomorphic product elements (Lee et al., 2017; Aggarwal & McGill, 2007; Aggarwal & McGill, 2012; Kraak & Story, 2015), in the present study, anthropomorphism had no such effects. Additionally, previous studies provided evidence for an increase in ideal congruity resulting from anthropomorphism, however, the present study did not reflect these findings. The humanisation of granola packaging did not lead participants to increase their health consumption behaviour. On the contrary, product evaluation was more positive for the neutral, non-anthropomorphized granola box in comparison with the anthropomorphized granola box. In line with the hypotheses, anthropomorphised packaging did lead to an increase in purchase intention and consumption intention, however, these effects only occurred with ideal congruity in the model and they were not significant. Without ideal congruity in the model, negative effects of anthropomorphism on health consumption were found: product evaluation, purchase intention and consumption intention decreased as a result of anthropomorphism, however, these results were not significant.

Additionally, humanlike elements in the granola packaging did not result in an increase in ideal congruity. In other words, participants did not feel as though the personality represented

27

by the anthropomorphized granola box had similar characteristics to their own ideal personality. Interestingly, the results suggest the opposite: participants rated the perceived similarity between their ideal personality and that of the anthropomorphized product as lower than the perceived similarity between their ideal personality and the neutral, non-anthropomorphized granola box.

Furthermore, the results demonstrate that ideal congruity has a positive effect on health consumption behaviour. Product evaluation, purchase and consumption intention increase as a result of an increase in ideal congruity. In line with theorizing and the hypothesis, when consumers feel their ideal personality matches with that of the product, they are more likely to purchase and consume the product, and evaluate it more positively. Last but not least, the interaction effect of anthropomorphism and ideal congruity was significant, however, this is merely a reflection of the effect of ideal congruity on health consumption. The direct effect of anthropomorphism on health consumption was not significant, meaning ideal congruity is not a mediator of the relationship between anthropomorphism and health consumption behaviour. There are several explanations as to why anthropomorphism did not significantly increase health consumption and ideal congruity. An important question that is raised by these results is how to know if consumers successfully anthropomorphize a product? The present study pre-tested the level of anthropomorphism of the designed packagings to check if they were actually perceived as anthropomorphized. Participants received a short definition of anthropomorphism and were then asked the degree to which the packaging was anthropomorphic. Results showed that the anthropomorphized packaging was indeed perceived as such. However, according to Aggarwal & McGill (2007), mere exposure to a humanlike product is not enough to get consumers to anthropomorphize that product. Consumers process information based on schema: any new type of information is processed as belonging to a particular category or schema. Consequently, consumers evaluate products based on the

28

perceived fit between that product and the corresponding schema. For example, in their experiment, the grilles of cars were manipulated to either frown or smile. The cars were only perceived to frown or smile when a human schema was activated in consumers’ minds by introducing it as part of a product ‘family’ rather than a product ‘line’. In other words, people only anthropomorphize when they are thinking in a human schema. A human schema thus needs to be activated in the consumer’s mind before they can successfully anthropomorphize the product. This schema congruity theory proposed by Aggarwal & McGill (2007) could explain why the present study did not find the predicted results: participants were not primed to activate a human schema and thus might not have successfully anthropomorphized the granola box. Additionally, previous research demonstrated that consumers tend to classify foods into two categories: healthy, but not tasty, versus unhealthy and tasty. Anthropomorphized foods, whether they are humanlike in shape, spokes character, brand name or any other element, tend to be classified as unhealthy and are associated with fun, indulgence and good flavour (Kraak & Story, 2015; Lee et al., 2017), whereas neutral, serious looking food is usually categorized as healthy but not tasty. This is in contrast with the present study, which anthropomorphized a healthy food type. In accordance with schema congruity theory, consumers might not have been able to perceive the fit between anthropomorphism and healthy and this incongruity might have undermined the positive effects of anthropomorphism on health consumption. Schema congruity theory also explains why the anthropomorphized granola box failed to increase ideal congruity. If participants did not successfully anthropomorphize the granola box, they might have been unable to perceive the product as having a personality and consequently might not have been able to perceive similarity between the product’s personality and their own ideal personality.

Aggarwal & McGill (2007) propose an alternative explanation for product evaluation as a result of schema based thinking. Each schema or category has an affective tag that is either

29

positive or negative. Even when consumers are primed to think in a human schema and successfully anthropomorphize a product, who is to say that this human schema is a positive one? Additionally, the granola box may have activated a healthy schema in participants’ minds. Since many people struggle with eating healthy, participants might have had negative associations with the healthy schema, and thus have evaluated the granola box more negatively than they would have, had they had positive associations with the healthy schema.

In accordance with the hypothesis, the results provide evidence that ideal congruity is a predictor of health consumption behaviour. The present research did not study the mechanisms underlying these effects, but theory postulates that recognition of characteristics a person’s ideal self possesses in a product establishes an emotional connection between consumer and brand (Astakhova et al., 2017), which in turn results in brand preference, loyalty and increased purchases (Raushnabel & Ahuvia, 2014; Gameiro, 2017). Additionally, since consumers use brands and products to express their identity, purchasing and consuming products that represent characteristics of their ideal selves helps them maintain and enhance their image and self-esteem (Astakhova et al., 2017).

This study attempted to discover if anthropomorphic elements on packaging had any effects on health consumption behaviour. The results did not show any such effects, which raises the question of the role packaging played in this. Previous studies have demonstrated that packaging elements, such as colour, shape and size, and brand elements on packagings such as brand name, logo or slogan, can influence consumer behaviour (Triantos et al., 2016; Gameiro, 2017). Thinking back to schema congruity theory, perhaps, when it comes to anthropomorphism, packaging elements only are not enough to get consumers to anthropomorphize the product. Or maybe other types of packaging elements are more relevant for anthropomorphism, such as shape, size, or material. Finally, maybe the anthropomorphic packaging just did not appeal to the participants, which led them to evaluate it more negatively.

30 Limitations and directions for future research

This study had some limitations. First of all, the experiment was conducted with a relatively small sample, N = 104. Results may have turned out differently if the sample size had been larger. A small sample size has several downsides. Firstly, it reduces representativeness of the population: the smaller the sample, the less variety in characteristics represented by the sample. Secondly, a small sample size increases variability which may result in bias, consequently reducing reliability. Furthermore, a small sample size decreases accuracy of results and thus increases the likelihood of a Type II error occurring. This in turn decreases the power of the experiment. It is thus of importance that future studies use a greater sample size to increase power, precision, and reliability. Additionally, a large part of the participants were recruited via a Facebook group for students looking for respondents for their thesis survey. This is a population that is not entirely representative of all consumers, therefore, the results may have been biased and are not generalizable to all consumers. Future research on this topic should recruit a more representative sample or look at a different population to increase generalizability of the results.

Secondly, the experiment studied intended behaviour instead of actual, real life behaviour. The studying of real life behaviour falls outside the scope of this study, however, it does pose a limitation since intended behaviour is very different from actual behaviour in several ways. Most importantly, intentions aren’t actual behaviour, they are far from it. When participants indicate their intended behaviour, who is to say they would actually behave that way in a real life setting? The intention behaviour gap is a well-documented phenomenon that proves there is a large difference between people’s intentions and their actual behaviour (see for example Jackson et al., 2005). The results of this study could thus have been very different had I studied actual behaviour.

31

Furthermore, in the experiment, participants were asked to imagine the granola box to be a real life person. As a researcher, I did not have any influence over the type of person the participants called to mind. One would expect participants to imagine a healthy person, but there is no way to be sure of this. Participants may have imagined a person they had negative associations with, which may have biased the results. Additionally, participants were asked to judge the personality of the product based on a list of ten characteristics representative of a healthy person. They were asked to do this so they would have an idea of the personality of the product, which they could later compare to their own ideal personality. However, asking only for health related characteristics may have framed the personality of the product in such a way that other characteristics consumers might have ascribed to the product otherwise did not occur to them. In other words, participants might have judged the personality of the product otherwise, had it been an open question. Future research should take this into account and leave participants to judge the personality based on their own perceptions.

The findings of the present study offer several opportunities for future research. As mentioned above, the current study only tested intended behaviour, but not actual behaviour. Studying actual behaviour instead of intentions would give more accurate and reliable insight into consumer’s behaviour. Future studies could, for example, study the quantity of food consumed after being exposed to anthropomorphic packagings, or observe consumer’s preference when given the choice between an anthropomorphized or neutral looking healthy product in a supermarket setting. Furthermore, this study only tested the effects of anthropomorphism directly after exposure to the stimulus. No research to date, to my knowledge, has studied the long-term effects of anthropomorphism on consumer behaviour. Perhaps long-term exposure to anthropomorphic elements surrounding a product or brand would have stronger effects on consumer behaviour. Long-term studies could also research the effects of anthropomorphism on the emotional bonds consumers develop with brands and

32

products, and the possible mediating role of relationships on health consumption behaviour. Additionally, there is a lack of research on anthropomorphic packaging effects on consumer behaviour. The current study is one of the first to consider humanlike packagings. More research is necessary to find out the conditions under which anthropomorphic packaging elements might have a significant effect on consumer behaviour. Future research could, for example, consider different anthropomorphic packaging shapes, much like the Coca Cola bottles mentioned in the introduction, or could study the effects of anthropomorphic slogans on packagings. Future research could also consider different food types, such as salads or healthy snacks. Linking back to schema congruity theory, snacks might be more compatible to the anthropomorphic schema, because they are usually associated with indulgence and good flavour. Additionally, future studies could compare the effects of anthropomorphic packaging on healthy foods to the effects of anthropomorphic packaging on unhealthy foods.

Finally, future research could try to add elements to the research which would help consumers successfully anthropomorphize the product. For example, a short story could be written in which the product would be introduced either in first person (Hi, I’m Fred, a healthy type of granola. I am packed with healthy grains and nuts, etc.) for the anthropomorphic condition and in third person (This is All Natural, a healthy type of granola, packed with healthy grains and nuts, etc.) for the neutral condition. This might help consumers to activate a humanlike mental schema which would facilitate the successful anthropomorphizing of the product.

Theoretical and practical implications

Although the results of this study were not statistically significant, this does not mean they don’t have important implications. This study contributes to the extant knowledge on anthropomorphism in relation to food and the knowledge on anthropomorphic food packaging. Extant literature lacks in linking anthropomorphism to both healthy foods and packaging, since

33

most existing literature focuses on anthropomorphism in relation to unhealthy foods and food shape rather than packaging elements. The results of this study prove that further research is necessary to explore other factors that play a role in the effectivity of anthropomorphic packaging in influencing health consumption behaviour.

This study contributes to the literature on anthropomorphism and consumer behaviour by demonstrating results contrasting with extant literature: this study provides evidence that anthropomorphic packaging does not lead to increased ideal congruity or increased health consumption, and the relationship between anthropomorphic packaging and health consumption behaviour is not mediated by ideal congruity. More research is needed to reconcile these differences and to find the confining conditions in which anthropomorphism in packaging does and does not affect consumers’ health behaviours.

For marketing and product managers, the present study provides evidence that merely anthropomorphizing a product’s packaging alone does not affect consumers’ consumption and purchase behaviours for healthy products. Like previous studies suggest, anthropomorphism in products seems to work best for unhealthy, but tasty products, such as candy, ice cream, chips or chocolate. This implies that marketers of healthy foods should keep their packaging neutral and serious in order to increase purchase and consumption. This is in line with research from Lee et al. (2017). Considering the effects of ideal congruity on health consumption behaviour, further research is necessary to establish the conditions under which ideal congruity leads to higher consumption intentions, purchase intentions and product evaluations and to find the underlying mechanisms that cause this effect. The results of the present study indicate that although anthropomorphism may not lead to ideal congruity or health consumption behaviour, ideal congruity does have an effect on health consumption behaviour. For marketing and product managers, this means that stressing healthy characteristics in food packaging and appealing to consumers’ ideal selves works in increasing their health consumption intentions.

34

Marketers should therefore keep positioning their healthy products in a way that appeals to consumers’ ideal selves.

I would like to take this opportunity to stress the importance of non-significant results in order to prevent publication bias. Publication bias leads to a distortion in the balance of findings. It is important to know in which situations and contexts anthropomorphic packaging is effective, but also important to know in which situations and contexts it is not. The present study provides directions for the situations in which anthropomorphic packaging does not affect health consumption and in doing so contributes to the defining of context for anthropomorphic packaging. More research is necessary to further establish the boundary conditions of these effects and to fully understand the underlying mechanisms.

35 Appendices

Appendix A Manipulations of independent variable anthropomorphism

36 Appendix B Survey questions

Dutch

Beste deelnemer,

Bedankt dat u mee wilt doen aan mijn afstudeeronderzoek. Dit onderzoek gaat over elementen van productverpakkingen en hoe consumenten daarop reageren.

Het invullen van het onderzoek zal ongeveer 5 minuten duren. Houd bij het beantwoorden van de vragen in gedachten dat er geen goed of fout antwoord is. Het beste antwoord is het

antwoord dat het dichtst ligt bij uw ervaring en/of wat volgens u het best beschrijft wat gevraagd wordt. Uiteraard is het onderzoek anoniem en worden de gegevens zeer vertrouwelijk behandeld. De data (d.w.z. de informatie en antwoorden die u geeft in dit onderzoek) zal enkel gebruikt worden voor het doel van dit onderzoek. Door mee te doen aan dit onderzoek gaat u akkoord met het gebruiken van deze data voor het doel van dit

onderzoek. Deelname aan dit onderzoek is geheel vrijwillig en u kunt op ieder moment stoppen zonder daar een reden voor te hoeven geven.

Voor vragen kunt u mailen naar tabitha.richtersz@student.uva.nl.

Wat is uw geslacht?

o

o

o

Wat is uw leeftijd in jaren? (vul een heel getal in) Wat is uw hoogst afgeronde opleiding?

37

o

o

o

o

o

U krijgt nu een afbeelding van een verpakking te zien. Neem een moment de tijd om de verschillende elementen van de verpakking te benoemen en te bekijken. Beantwoord vervolgens de vragen.

Geef aan in hoeverre u het eens bent met de volgende stellingen. Product evaluation

Ik vind deze verpakking leuk. Deze verpakking spreekt mij aan.

Mijn houding ten opzichte van dit product is positief.

Hoe waarschijnlijk is het dat u dit product zou aanraden aan uw vrienden? Dit product is lekker.

Dit product is aantrekkelijk. Dit product is wenselijk. Purchase intention

Ik ga dit product zeker kopen in de toekomst.

Het is waarschijnlijk dat ik dit product koop in de nabije toekomst. Ik verwacht producten als deze te kopen n de nabije toekomst. Consumption intention

Ik zou dit product graag consumeren.

Ik ben van plan dit product te consumeren in de nabije toekomst. Ik ben bereid dit product in de toekomst te consumeren.

Actual self-image

Geef aan in hoeverre de volgende eigenschappen uzelf beschrijven.

38 Ideal self-image

Geef aan in hoeverre de volgende eigenschappen uw ideale zelf (d.w.z. hoe u het liefst zou willen zijn) beschrijven.

Sportief Gezond Fit Actief Energiek In balans Bewust Product personality

Stelt u zich dit product voor alsof het een persoon is. Geef aan in hoeverre dit product de volgende persoonlijkheidseigenschappen heeft.

Sportief Gezond Fit Actief Energiek In balans Bewust Ideal congruity

Geef aan in hoeverre u het eens bent met de volgende stelling. Houd bij iedere vraag het product dat u gezien hebt in gedachten.

De persoonlijkheid van dit product is consistent met hoe ik idealiter zou willen zijn (mijn ideale zelf).

De persoonlijkheid van dit product is een spiegelbeeld van de persoon die ik zou willen zijn. De persoonlijkheid van dit product komt overeen met het beeld dat ik van mijzelf zou willen hebben.

Ik zie een deel van wie ik zou willen zijn terug in dit product. Health consciousness

Geef aan in hoeverre u het eens bent met de volgende stellingen. Ik geef veel op voor mijn gezondheid.

Ik beschouw mezelf als zeer gezondheidsbewust.

Ik ben bereid veel te laten staan om zo gezond mogelijk te eten. Ik vind dat ik veel rekening houd met gezondheid in mijn leven. Ik vind dat het belangrijk is om te weten hoe je gezond moet eten.

Mijn gezondheid is zo waardevol voor mij dat ik bereid ben er veel voor op te geven. Ik heb het idee dat anderen meer aandacht besteden aan hun gezondheid dan ikzelf. Ik vraag mij niet continu af of iets goed is voor mijn gezondheid.

Ik wil me niet continu afvragen of de dingen die ik eet goed zijn voor mijn gezondheid. Ik sta vaak stil bij mijn gezondheid.

Healthy lifestyle

Geef aan in hoeverre u het eens bent met de volgende stellingen. Ik sport regelmatig