Anthropomorphism as selling point to increase the acceptance of flawed products

Anthropomorphism as selling point to increase the

acceptance of flawed products

A research to support the reduction of food loss

Uniquely flawed –

The effect of uniqueness-seeking on acceptance of flawed products

Anthropomorphism as selling point to increase the acceptance of flawed

products

A research to support the reduction of food loss

University of Groningen

Faculty of Economic and Business MSc Marketing Management

Master Thesis

Completion date: January 16, 2017 First supervisor: dr. Jing Wan Second supervisor: dr. Yannick Joye

Jasmin Rösch

Address: Folkingestraat 50A 9711 JZ Groningen Phone: 0177 5679634 Mail: jasmin.rosch@student.rug.nl

EXECUTIVE SUMMARY

To link the results back to the goal of the study, namely the reduction of food loss, important implications are made. To support this reduction retailer should:

• Use anthropomorphism in marketing campaigns for imperfect foods to draw attention to these kinds of food since results showed an effect of anthropomorphism on imperfect vegetables.

• Focus on the perceived quality of the imperfect vegetables since during the research the quality of the vegetables, detached from being perfect or imperfect was rated as high. Retailers should use anthropomorphism to increase quality perception of an imperfect food.

PREFACE

Food waste and food loss is an ever growing topic in today’s society and one that me personally makes very sad. Worldwide millions of people die because of hunger and in our society people throw away food because it does not have the perfect shape. Retailers even use this ridiculous reason to not sell those foods in the first place. To find a solution that counteracts this progress, I wanted to contribute to the reduction of food loss. This is why my research is of particular value. I hoped to find a simple marketing operation to oppose food loss and to open people’s minds towards this important topic. Along the track to the final thesis of my Master’s degree, some people encouraged me notably.

First of all, I want to express my deepest thank you to Dr. Jing Wan, my supervisor, who hit me on the idea to combine anthropomorphism and the acceptance of imperfect vegetables. She supported me during the whole research.

Equally, I want to thank the fellow students from my research group, especially Janina Marth, who spent days and night in the laboratory and library with me, making the work on this thesis a pleasure.

My sincere thanks goes to the laboratory team, especially Reinder Dallinga, who supported me during the conduction of my survey and helped me enjoy those busy days. This led to a professional implementation of the research.

Last but not least, I want to thank all my fellow students who took time to fill in my questionnaire.

TABLE OF CONTENT

EXECUTIVE SUMMARY ... i

PREFACE ... iii

1. INTRODUCTION ... 1

1.1 Defining the term anthropomorphism ... 6

1.2 Defining the Research Question ... 6

2. THEORETICAL BACKGROUND AND CONCEPTUAL FRAMEWORK ... 7

2.1 Theoretical Background on Anthropomorphism ... 7

2.2 Conceptual Framework - Why anthropomorphism should influence acceptance of flawed products (H1): ... 9

2.3 Introducing the Moderator ... 11

2.4. Theoretical Background on Involvement ... 12

2.5 Conceptual Framework – Why involvement influences anthropomorphism (H2) ... 13

3. METHODOLOGY ... 15

3.1 Experimental research design ... 15

3.2 Participants ... 16

3.3 Measures and procedure ... 17

3.4 Ethics ... 20

4. RESULTS ... 20

4.1 Testing Hypothesis 1 ... 20

4.2 Testing Hypothesis 2 ... 23

4.3 Testing for additional factors ... 25

1. INTRODUCTION

Flawed products are an ever growing issue in today’s society especially in the food industry, where 1.3 billion tons of food are either wasted or lost partly because of superficial flaws (Cicatiello et al., 2016). In the United Kingdom, for example, 30 percent of vegetables are not harvested because they do not meet the retailers’ standards due to physical appearance like shape, size, or color (Smithers, 2013, Lipinski et al., 2016). According to the Oxford Dictionary, there are some definitions for the word flaw. A flaw is a blemish, or another imperfection which mars an object. The Oxford Dictionary also defines a flaw as a weakness of the character of a person (Stevenson, 2016). When looking at the two definitions, it can clearly be seen that a flaw does not only appear on an object or product but also on a person. This finding provided the thought-provoking impulse for this research.

Nowadays, consumers rather choose the perfect product compared to a superficially flawed one because these days almost everything is expected to be perfect. Barry Schwartz (2004) made an interesting point in his book ‘The Paradox of Choice’ about this assumption. He states that perfection is the only medium against regret. He further mentions that the continuous search for alternatives is the only possibility to reach perfection. This might be the reason why people always strive for perfect objects in their lives. From a psychological point of view it seems that people seek perfect products because they want to compensate their own imperfection (Radwan, 2016).When entering the word ‘perfect’ in Google it becomes clear that everything needs to be perfect in peoples’ life; people want perfect clear skin, perfect looking and fitting clothes, perfect tasting coffee, perfect located and relaxing holidays, and much more. Howard (2011) states that this fact can come from the circumstance that perfection can make addicted. She states, the more people seek perfection the more it can become an obsession because it is the most people hope for. This again, is in line with the statement of Barry Schwartz (2004). The more perfect ones possessions are, the more likely it is to reach the perfect life people are looking for. Hence, it is just a matter of fact that people also expect perfect products on their sides. The only ‘objects’ in people’s life that are allowed to have flaws are the children, friends or maybe the cohabitation partner. It seems like close acquaintances are the only source of trouble that are accepted or even liked. Chhabria (2012) goes even further by saying that perfect people with perfect characters scare others. She mentions that people can identify better with others who have character flaws because they feel a similarity towards those people since oneself also might have flaws.

people need to be convinced that these foods are as good as their perfect grown fellows. This thought-provoking impulse builds the foundation for this paper and is further developed in the next paragraphs.

Since people accept people who have flaws it might be a possibility to anthropomorphize imperfect vegetables and make them look more human-like. In this way, people might accept superficially flawed foods because they remind them of humans. Aggarwal and McGill (2007) found that higher conformity between product features and the anthropomorphized product will lead to positive evaluations. When evaluating humans, people use a different set of norms compared to evaluating objects. Troutman and Shanteau (1976) found out that products are evaluated based on the quality of information and facts about it. People, on the other hand, are often evaluated based on physical cues like their smile (Wilson & Eckel, 2006). Anthropomorphism can help to build such cues by putting a happy face on the product. This leads to the assumption that flawed products, when anthropomorphized, are more accepted by people since the association with humans within the product might support the acceptance of the flaw.

further state that handmade products (compared to machine-made products) are more liked because people’s love is expected behind it (Fuchs et al., 2015). It can be seen that products, when a human can be linked to it, are more accepted and valued. This indicates that handmade products, which are associated with love and flaws of the person who crafted it, will be chosen because they reflect oneself (Gardner & Knowels, 2008). It is assumed that when people buy products, which reflect themselves, they have a stronger connection with the product. This increases the liking of the product which increases the chance that the product becomes anthropomorphized. It indicates that little flaws are more accepted when the product is handmade (reflecting the perfection and imperfection of the producer and oneself) and hence, is another reason why it is assumed that humanization of a flawed product works to increase its sales figures.

Before defining the research question it is important that the terms mentioned above are understood perfectly. This is why this research starts with defining anthropomorphism.

1.1 Defining the term anthropomorphism

To start from the beginning, the word anthropomorphism comes from the Greek and combines the words anthropos which means human and morphe which means shape/form. This indicates that anthropomorphism is not only attributing non-human entities with human-like behavior but rather describing a non-human object with human mental or physical characteristics (Epley et al., 2007). To better understand the meaning of the word, other interesting definitions are presented. Anthropomorphism is the inclination to ascribe non-human objects or entities with non-human characteristics (Triantos et al., 2016). Kim & McGill (2011) define anthropomorphism as the propensity to attribute humanlike character traits and manners to non-human objects. Epley et al. (2007) see imbuing the behavior of non-human objects with human-like emotions as the essence of anthropomorphism. Further, they say that these objects can be everything from animals to electronic devices. Epley et al. (2007, 2008) state that anthropomorphism appears out of two reasons: effectance and social motivation.

1.2 Defining the Research Question

Since the most important terms of this research are explained, the research question will now be defined. The question that is in the center of this research is:

Are products with superficial flaws more accepted when anthropomorphized?

vegetables builds the dependent variable. Anthropomorphism is used as independent variable to find out if the sales of imperfect vegetables can be increased when using it.

After defining the most important terms and the research question, in the second part, the existing literature of anthropomorphism is analyzed and combined with the hypothesis. Afterwards, a moderator will be introduced. Third, the methodology is explained and the study is elucidated. In the fourth part, the results of the study are disclosed and fifth there is a discussion part in which the results are interpreted, implementations are made, limitations are listed, and further research is suggested. The sixth part will be an overall conclusion.

2. THEORETICAL BACKGROUND AND CONCEPTUAL FRAMEWORK

2.1 Theoretical Background on Anthropomorphism

people think about it by having an emotional bonding with those products. Eisenberg (2011), a book author and passionate sales man, mentions that bonding with a product is one of the reasons why people buy a product. This indicates that anthropomorphism, since it increases the bonding with products, can influence the buying decision of a person towards the product. Aggarwal and McGill (2012) approve this assertion by saying that anthropomorphism is used in advertising to make a product more memorable and distinctive. Through anthropomorphism products are even perceived to have a soul. Therefore, it is more likely that people build a relationship with anthropomorphized products (Aggarwal & McGill, 2012).

2.2 Conceptual Framework - Why anthropomorphism should influence acceptance of

flawed products (H1):

There is a lot of literature that proves why the combination of anthropomorphism and flawed products should work to increase attention and hence increase its acceptance. This should affect the sales figures in a positive way. Chandler and Schwarz (2010) have an interesting view about anthropomorphism and the functionality of a product. Of course, objects should work properly and be accurate in their function but according to them, this theory does not concern people. For people the functionality is not important and people are allowed to malfunction. Chandler and Schwarz (2010) further state that this is due to the fact that people are not evaluated by their functionality but rather by their personality. This leads them to the assumption that when objects are anthropomorphized they are allowed to malfunction, too. Elaborating this thought in more detail, brings Chandler and Schwarz (2010) to the belief that when anthropomorphizing an object the focus is not on the functionality anymore but rather realm of interpersonal relationship. According to their statement, it is expected, that anthropomorphism has an effect on flawed products because anthropomorphized products are associated with humans and humans are allowed to have weaknesses or flaws. As Breines (2015) already mentioned above, people are liked because they have flaws and weaknesses. Even the Oxford Dictionary, as mentioned in the introduction, defines a flaw as a human characteristic. This is another reason why an effect of anthropomorphism on flawed products is expected. Epley et al. (2008) support this theory by saying that anthropomorphized objects are likelier to be treated equally to human beings. This indicates that when people are allowed to have flaws the same is true for flawed but anthropomorphized products.

products because people take higher notice towards the anthropomorphized object. Additionally, they mention in their research that products or brands with a higher attitude are chosen over products or brands with lower attitude. Park et al. (2010) define the value of attitude as the extent of positivity or negativity an object is evaluated with. They further state that a higher attitude comes from elaborate thinking about an object caused by personal relevance. The higher the attitude towards a product, the more likely people purchase the product. Kim and McGill (2011) on the other hand mention that attention, gained through anthropomorphism, leads to a higher attitude towards the anthropomorphized product, which increases according to Park et al. (2010) the purchase intention. This leads to the assumption that an anthropomorphized product is chosen over one that is not anthropomorphized.

The theories discussed in the section above lead to the first hypothesis, combining anthropomorphism and the effect on flawed products:

H1: If attention through anthropomorphism (vs. non-anthropomorphism) increases,

choice over the anthropomorphized, flawed product (vs. non-anthropomorphized, flawed product) increases.

In the next section a moderator will be introduced to find out when and under what conditions the independent variable (=IV, here anthropomorphism) influences the dependent variable (=DV, here acceptance of flawed products) (Kim et al., 2001).

2.3 Introducing the Moderator

2.4. Theoretical Background on Involvement

To define the word involvement, it is important to know that there are different kinds of involvement. According to Zaichkowsky (1985), people can be involved with ads, products, or with purchase decision. For this research, the involvement with products is of particular importance and hence, needs to be defined more detailed. The involvement with products leads to a greater cognition towards attribute differences, product importance, and commitment to brand choice (Zaichkowsky, 1985). Furthermore, product involvement is defined as the permanent perception a consumer has about the significance of the product category on the basis of consumers’ values, interest, and innate state (Bian & Moutinho, 2009). VonRiesen and Herndon (2011) found that involvement can change over time and within different types of situations. They say that involvement does not necessarily apply to an object or product directly but rather to the person’s manner towards the product.

Petty and Cacioppo (1981) made the connection between personal relevance and involvement very clear in their research. They said that involvement can be either high or low. When high involved, the persuasive message has a high personal relevance for the person and when low involved, the personal relevance of the message is low. Petty and Cacioppo (1986) mentioned that personal relevance is the most important factor in affecting a message’s liking and involvement is crucial for attitude change. Attitude change can be defined as adjustment in the evaluation of a product or object, which leads to the formation of new evaluations towards the product or object (Glaser et al., 2015).

the peripheral route. In the central route, the arguments are considered carefully before an attitude change (Fennis & Stroebe, 2016). The personal relevance of the product increases, it becomes more important, and people start thinking more precisely before forming an opinion (Petty et al. 1983). This indicates that the involvement in the central route is high. In the peripheral route, the attitude change happens without a lot of thinking. Attitude change does not occur because the person has thought about the pros and cons of the products but rather because the product is connected with positive or negative associations (Petty et al. 1983). Heuristic processing dominates the peripheral route and involvement is low (Fennis & Stroebe, 2016). Petty and Cacioppo (1986) say that increasing involvement leads to resistance to persuasion. Baker and Petty (1994) found out that persuasion is greater when arguments are stronger and smaller when the arguments are weak. This supports the theory of Petty and Cacioppo (1986). It indicates that people following the peripheral cue are more likely to be persuaded towards an attitude change. Humanization does serve as a peripheral cue as found by Lee (2010). He found out that peripheral processing operates in a holistically manner, requiring only little cognitive resources. Further, he suggests that humanization has a bigger impact on attitude change for people following the peripheral route. This supports the theory that anthropomorphism leads to attitude change towards a product.

2.5 Conceptual Framework – Why involvement influences anthropomorphism (H2)

Therefore, low involvement amplifies the effect of anthropomorphism on the acceptance of flawed products. This is why involvement is chosen as moderator.

Petty and Cacioppo (1983) say that the evaluation of a product has a bigger impact on low involved people, when they do not have to engage in product relevant thinking. This indicates that low involved people will connect better with anthropomorphized products because they do not have to evaluate the product on facts but rather on its visual effects. According to the research of Lee (2010), saying that humanization serves as peripheral cue, it is expected that people who are low involved and hence follow the peripheral route are more likely to react to anthropomorphism.

Another research that supports this expectation is the research of Basfirinci and Çilingir (2010) who found out that high involvement negatively affects the efficiency of anthropomorphism. Moreover, advertisements using anthropomorphized products have more influence on people with low involvement. Consequently, people with low involvement (following the peripheral route), usually use easy accessible and salient cues to which, according to Basfirinci and Çilingir (2010), anthropomorphized products belong to. Therefore, low involvement is expected to increase the efficiency of anthropomorphism.

These correlations between involvement and anthropomorphism lead to the assumption of the second hypothesis.

H2: If involvement decreases, the chance that the anthropomorphized, flawed product

(vs. the non-anthropomorphized, flawed product) is chosen increases.

3. METHODOLOGY

This research focuses on the acceptance of flawed products which, during the study, is depicted in form of imperfect vegetables. With this study it is aimed to find whether anthropomorphism influences people’s acceptance of flawed products. The study has the goal to prevent food loss by finding a possibility to change people’s minds about imperfect vegetables by using anthropomorphism and hence, to increase the acceptance of imperfect vegetables. Anthropomorphism will lead people to think about the product as being a human (Landwehr et al., 2011) and since humans are allowed to have flaws it is expected that the acceptance of imperfect vegetables increases when they are anthropomorphized (Breines, 2015). The second goal is to find out if product involvement has a moderating role. People with low involvement (following the peripheral route (Petty et al. 1983)) are expected to more likely accept an anthropomorphized product compared to high involved people, since anthropomorphism can be seen as peripheral cue (Lee, 2010).

3.1 Experimental research design

it can be assumed that the answers are not infl

a higher number of variables can be considered and observed (Malhotra, 2011).

The research is a 2 (humanization: anthropomorphized vs. non product) x 2 (vegetable: flawed vs. non

design. Anthropomorphism and the flaws of the vegetables Involvement is measured after the manip

towards the vegetables (see Fi

3.2 Participants

In total 259 participants take

men. The participants are students of the University of Groningen, Netherlands. They national and international student

the study. Average age of the participants

anthropomorphized vs. non-anthropomorphized product) x 2 (veg

flawed vegetable) x involvement between participant study. The participants allocated to the different conditions. Each participant

it can be assumed that the answers are not influenced by external factors. In a a higher number of variables can be considered and observed (Malhotra, 2011).

The research is a 2 (humanization: anthropomorphized vs.

non-product) x 2 (vegetable: flawed vs. non-flawed vegetable) x involvement between participant design. Anthropomorphism and the flaws of the vegetables are manipulated during the study.

after the manipulation to find out participants’ product involv (see Figure 1).

Figure 1: Conceptual framework

take part in the study. 134 of the participants

students of the University of Groningen, Netherlands. They national and international students and either get research points or money when taking part in the study. Average age of the participants is 21. The students take part in a 2 (humanization: anthropomorphized product) x 2 (vegetable: flawed vs. non flawed vegetable) x involvement between participant study. The participants

allocated to the different conditions. Each participant takes part in only one condition. H1

H2

enced by external factors. In a laboratory study a higher number of variables can be considered and observed (Malhotra, 2011).

-anthropomorphized etable) x involvement between participant manipulated during the study. product involvement

part in the study. 134 of the participants are female and 125 students of the University of Groningen, Netherlands. They are points or money when taking part in part in a 2 (humanization: etable: flawed vs. non-flawed vegetable) x involvement between participant study. The participants are randomly

3.3 Measures and procedure

To answer the hypotheses

participants sees imperfect vegetables; division is made with

anthropomorphized vegetables; half s make the conditions clear, F

conditions (see also Appendix 1).

Independent variable: The independent variable Pictures of different vegetables were edited

anthropomorphized condition

about the vegetables is additionally given in the first person to strengthen the effect of anthropomorphism. In the non

given in the third person and the vegetables

vegetables are used, when being in the flawed condition used in the non-flawed condition

into an advertisement campaign to give the participants an idea vegetables would look like (see

rocedure

To answer the hypotheses, there are four conditions that need to be tested. vegetables; the other half sees perfect shaped vegetables.

anthropomorphized vegetables. Half participants anthropomorphized vegetables; half sees vegetables which are not anthropomorphized. make the conditions clear, Figure 2 shows the factorial design including the

ppendix 1).

Figure 2: Factorial design

The independent variable (anthropomorphism)

Pictures of different vegetables were edited to give the vegetables, when being in the anthropomorphized condition, a face. In the anthropomorphized condition the information additionally given in the first person to strengthen the effect of In the non-anthropomorphized condition the ancillary information given in the third person and the vegetables do not have a face. Pictures of imperfect shape

used, when being in the flawed condition and perfect shaped vegetables flawed condition. The pictures of the different vegetables

into an advertisement campaign to give the participants an idea of how a (see Figure 3 and Appendix 1).

there are four conditions that need to be tested. Half of the perfect shaped vegetables. The same anthropomorphized vegetables. Half participants see not anthropomorphized. To igure 2 shows the factorial design including the division of

Figure 3: Manipulated advertisement campaigns per condition (using the example of potatoes)

Dependent variable: To measure the dependent variable (acceptance of flawed products), different questions are asked after each picture of the manipulated advertisement campaigns. The questions involved the participant’s evaluation about the liking of the displayed vegetable, the purchase intention, and the enjoyment of eating the displayed vegetable. Those three questions build the dependent variable (see Appendix 1). After testing the DV of all displayed vegetables, the involvement towards the vegetables is questioned.

Moderator: To get an overview about participant’s involvement, questions about how interested they are in vegetables, how vegetables help them to attain the lifestyle they strive for, how important they rate vegetables, and how long they could talk about vegetables are asked by using a 7-point Likert scale (1=Strongly Disagree, 7=Strongly Agree). These questions build the variable which is used to test the moderator, involvement.

other unrelated studies and takes place as third study. At the beginning of the study, participants are told that they take part in a research about general buying behavior in a retail environment. In the first part, participants see the names of six different vegetables. To avoid the possibility that a participant dislikes the shown vegetables, they can choose their three preferred vegetables out of the six. Afterwards, pictures of the chosen vegetables are shown. The scenario on the pictures of the vegetables differs, depending on which condition the participant is assigned to (see Figure 3 and Appendix 1).

3.4 Ethics

The research is designed in a way that participants do not suffer any physical harm or pain during the conduction. Also the questions are formulated that the respondents are not embarrassed or felt any psychological harm during the study. The study is anonymous which indicates that there is no loss in privacy at any time of the study.

4. RESULTS

4.1 Testing Hypothesis 1

effect of vegetables (F(1,192) = 94.157, p < .000) and a main effect of humanization (F(1,192) = 8.006, p < .005). The interaction effect between vegetable and humanization shows a marginal significance (F(1,192) = 2.893, p > 0.091) (see Appendix 2). To analyze this effect in more detail, a closer look into the different means is taken. Thus, a post-hoc test is made (see Appendix 3). Within the post-hoc test, it can be seen that in the non-flawed condition anthropomorphism does not statistically significant influences the acceptance of the vegetables (Mnon-anthro, non-flawed = 5.685, Manthro, non-flawed = 5,830, p > .415). However, in the

flawed condition anthropomorphism increases the acceptance of the vegetable statistically significant (Mnon-anthro, flawed = 4,202, Manthro, flawed = 4,802, p < .002) (see Table 1 and Figure

4). This indicates that hypothesis 1 is partially supported.

Dependent Variable: Acceptance of vegetables

Flaw Anthro Mean: Acceptance N

No Flaw No Anthro 5,685 48

Anthro 5,830 53

Flaw No Anthro 4,220 46

Anthro 4,802 46

Table 1: Means Acceptance of vegetable split by condition

Figure 4: Means Acceptance of vegetable split by condition 0 1 2 3 4 5 6

No Anthro Anthro No Anthro Anthro

Within the study cuteness and mood are asked as control factors. To make sure neither cuteness nor mood has an influence on the acceptance of imperfect vegetables, two separate ANCOVA analyses are conducted. First, a closer look is taken on cuteness. The question about perceived cuteness of the vegetables is averaged among all participants to build the covariate. When using this variable as a covariate, the interaction between vegetable and humanization still is marginally significant (F(1,192) = 3.172, p > .077) (see Appendix 4). This indicates that the effect of anthropomorphism on the acceptance of vegetables is not influenced by the individual’s perceived cuteness level of the vegetable.

The same measurement is made with mood as covariate. Therefore, a variable is build, combining the four mood-related questions (see Appendix 1). This variable is used within an ANCOVA as covariate. Again, the interaction effect between vegetables and humanization is marginally significant (F(1,192) = 3.073, p > .081) (see Appendix 5). It shows that participant’s individual mood does not influence the acceptance of imperfect vegetables.

4.2 Testing Hypothesis 2

To test the moderator, a 2 (humanization: anthropomorphized vs. non-anthropomorphized product) x 2 (vegetable: flawed vs. non-flawed vegetable) x involvement between subject ANOVA is conducted to find out if involvement strengthens the effect of humanization on vegetables (see Appendix 7). Again, participants who do not like the displayed vegetables are not considered. ‘Acceptance’ is used as dependent variable. The means of the asked involvement question build the moderator variable. A median split is carried out to divide involvement into high ( > 4,00) and low ( < 4,00) (see Appendix 8). Univariate analysis demonstrates that there is a main effect of vegetables (F(1,192) = 91.341, p < .000) and a main effect of humanization (F(1,192) = 8.675, p < .004). Involvement itself has no main effect. It has neither an interaction effect with vegetables nor with humanization. Also, there is no three-way-interaction between vegetables, humanization, and involvement (F(1,192) = .942, p > .333).

To understand why involvement shows no effect, a closer look on the means is taken (see Table 2). In the non-flawed condition, low involved participant show a similar acceptance of the non-anthropomorphized vegetable compared to the anthropomorphized one (Manthro, non-flawed, involvement low = 5.439; Manthro, non-flawed, involvement low = 5.762). In the flawed condition, the

difference between the anthropomorphized vegetable compared to the non-anthropomorphized vegetable is similar, when participants are low involved (Mnon-anthro, flawed, involvement low = 4.239; Manthro, flawed, involvement low = 4.720) (see Figure 5). When looking at the

low = 4.720, p < .025). Surprisingly, the same effect can be seen when participants are high

involved. For high involvement, the difference between the anthropomorphized vegetable to the non-anthropomorphized ones is again statistically significant when looking at the flawed condition (Mnon-anthro, flawed, involvement high = 4.193; Manthro, flawed, involvement high = 4.884, p < .032).

Contrary to what is expected, there is only little difference when comparing low and high involvement. This indicates that hypothesis 1 is not supported.

Dependent Variable: Acceptance of vegetable

Flaw Anthro Involvement Mean: Acceptance N

No Flaw No Anthro Low 5,439 21

High 5,877 27

Anthro Low 5,762 28

High 5,907 25

Flaw No Anthro Low 4,239 27

High 4,193 19

Anthro Low 4,720 23

High 4,884 23

Table 2: Means Acceptance of vegetable split by condition and involvement

Figure 5: Means Acceptance of vegetable split by low involvement

0 1 2 3 4 5 6

No Anthro Anthro No Anthro Anthro

No Flaw Flaw

Figure 6: Means Acceptance of vegetable split by high involvement

4.3 Testing for additional factors

Besides the testing of the two hypotheses, other variables are tested to find out if there are other factors playing a role in the acceptance of imperfect vegetables.

The perceived quality is asked after all three vegetables are shown (see Appendix 1). A new variable based on the means of the quality questions is formed. Within an ANOVA, this variable is used as DV to test how the quality of the vegetables is perceived in the different conditions. Neither vegetable nor humanization has a main effect. The interaction between vegetables and humanization is not significant (F(1,192) = 2.344, p > .127) (see Appendix 12). Nevertheless, the means show that the quality is perceived as high within all conditions (see Figure 7 and see Appendix 13). In the condition with the imperfect vegetables, anthropomorphism even raised the vegetable to the same perceived quality as a non-flawed vegetable. After this impressive finding, a post-hoc test is conducted. The test shows that the perceived quality in the non-anthropomorphized, flawed condition is significantly different from the anthropomorphized, flawed condition (Mno-anthro,flaw = 5.355 to Manthro,flaw = 5.746, p <

0 1 2 3 4 5 6

No Anthro Anthro No Anthro Anthro

No Flaw Flaw

.033) (see Appendix 14). This indicates that anthropomorphism increases the perceived quality of an imperfect vegetable to the same level as a perfect vegetable. Within the non-flawed condition such significance is not found, meaning that anthropomorphism does not have an effect on quality when the product is not flawed (Mno-anthro,no-flaw = 5.743, Manthro,no-flaw

= 5.748, p > .975). Both are interesting findings.

Figure 7: Means perceived quality split by condition

4.4 Results Discussion

Since H1 did get supported, the main finding of the first analysis shows that anthropomorphism itself has an impact on the acceptance of imperfect vegetables. This indicates that an imperfect vegetable is more accepted by people when it is anthropomorphized compared to when it is not anthropomorphized. It shows that anthropomorphism has an effect on imperfect vegetables. This is an important finding retailers can make use of. Furthermore, perceived cuteness of the vegetables and participant’s individual mood are used as control factors. Analysis shows that when controlling for cuteness and mood individually, both did not influence the interaction between vegetables and humanization. Nevertheless, when controlling for both together, hypothesis 1 is supported.

1 1,5 2 2,5 3 3,5 4 4,5 5 5,5 6

No Anthro Anthro No Anthro Anthro

No Flaw Flaw

The main finding of the second analysis is that involvement neither strengthens nor weakens the effect of anthropomorphism on flawed products. Results show a small difference in the acceptance of anthropomorphized, flawed vegetables in favor of low involvement when comparing low with high involved participants. However, hypothesis 2 finds no support.

Analysis of additional factors finds that quality is perceived as high throughout all conditions. The main finding in this analysis is that anthropomorphism increases the perceived quality of an imperfect vegetable to the same level as a non-anthropomorphized, perfect shaped vegetable. This indicates that anthropomorphism can help an imperfect vegetable to increase its perceived quality.

5. GENERAL DISCUSSION

acceptance of a flawed and anthropomorphized vegetable. The fact that anthropomorphism only has a significant effect on imperfect vegetables (compared to perfect vegetables) shows that humanization does not work for products that are perfect. The flaw itself leads to imperfection of the vegetable which can be associated with people’s own physical and character imperfections. This strengthens the effect of anthropomorphism in combination with a flawed vegetable.

In contrast to what was expected, involvement shows only very little influence on the acceptance of the vegetables when being in the flawed condition. One reason affecting this result could be the fact that when having a closer look participant’s involvement towards vegetables is neither low nor high (see Appendix 11). Since participants are only averagely involved, they do not follow the peripheral route mentioned by Petty et al. (1983) and hence, they do not see anthropomorphism as a peripheral cue as suggested by Lee (2010). This is why involvement does not affect the acceptance of the anthropomorphized vegetables as predicted. However, results show that there is a small difference between low and high involvement which indicates that the assumption of hypothesis 2 can potentially exist.

concern about little imperfections can be removed by using anthropomorphism as communication tool.

5.1 Managerial Implications

This research supports the assumption that anthropomorphism has a positive effect on flawed vegetables. It means people accept imperfect vegetables more when they are anthropomorphized compared to when they are not. Retailers can make use of this groundbreaking finding. Anthropomorphism can be used in a marketing campaign to increase the acceptance of imperfect vegetables. It might help people to see that flaws on vegetables are something likeable, as it is for humans (Breines, 2015). Moreover, the finding supports the statement that it does not count what is on the outside but rather what is inside which in case of the vegetables are nutrients and vitamins. This supports the acceptance of imperfect vegetables. Retailers can utilize this finding in their communication strategy by using anthropomorphism to advertise imperfect vegetables. They can, for example, put a happy face on the imperfect shaped vegetable within their advertisement or information brochure. This way they can sell imperfect foods which will contribute to the decrease of food loss since the vegetables do not have to be thrown away by the farmer. This means in the future imperfect vegetable do not have to be wasted but can be sold by retailers. Farmers can also use this finding to make retailers clear that it is not the imperfect shape why people do not buy foods but rather a way of communication to promote the acceptance of imperfect vegetables.

vegetables especially in combination with anthropomorphism. Retailers do not have to be afraid of the wrong idea that customers do not value the quality of imperfect vegetables and hence, avoid their stores as found by Gunders (2010). Not only retailers can benefit from this finding also farmers can make use of this insight. The fact that imperfect vegetables are still perceived as high quality can be used by farmers in their communication with the retailers. This way farmer can stress the fact that it is not the quality of the imperfect vegetable that leads to its rejection but that it is rather a way of communication such as using anthropomorphism to sell those kinds of foods to the end-customer. Both implications can easily be progressed and will support the overall goal of this research, reduction of food loss.

5.2 Limitations

This research provides valuable insights on imperfect vegetables combined with anthropomorphism. Despite these useful results, the research has some limitations which can potentially influence the results.

The first important limitation that needs to be mentioned is the fact that involvement did not work as a moderator. Since involvement was not manipulated and only asked after the vegetables are displayed, 70% of the participants stated to be medium involved (answering between 3 and 5; see Appendix 11). More precisely, 137 participants out of the 193 participants were medium involved. When splitting the involvement variable into low and high (by using a median split), the involvement means were not equally distributed across the involvement scale. This might have led to the fact that hypothesis 2 was not supported since low involved people, following the peripheral route, more likely use anthropomorphism as a peripheral cue.

hypothesis 1. Bushak (2014) found out that most people start eating and living healthy within their thirties. Hence it can be said that a healthy lifestyle including eating a lot of vegetables is a manner of life that comes with a higher age. The study was only filled in by 22 participants who were 25 years or older. This might have affected the results since only about 11 % of the participants were close to the age were a health-consciousness lifestyle starts. Results show that in general older participants (>25 years) have higher acceptance of vegetables, except one outlier who never buys vegetables on his own and hence, got excluded. This supports the assumption that older people eat more health-conscious and hence might be more willing to accept vegetables in general (see appendix 15).

The third limitation is the fact that while conducting the study, this research was shown to participants as last study out of three. The whole study (all three together) took about 50 minutes. Since this study was the last, it is possible that participants lost their concentration. Bunce et al. (2010) suggest that students only have a concentration span of about 20 minutes. This means that after 20 minutes the concentration of the students declines and they might not have read the questions thoroughly and hence, did not answer the questions correctly. This inattentiveness could lead to inconsistent answers by not reading carefully which might have influenced the results.

of participants is not less than 50. Since within this research the number undershot 50 participants per condition, sample size is mentioned as limitation.

5.3 Further Research

The results and limitations lead to several ideas for future research. Imperfect vegetables were a good way to check if anthropomorphism can lead to a higher acceptance. Interesting insights were gained especially because these findings can contribute to the reduction of food loss. Nevertheless, it is advised to test other product categories to get a broader perspective of this important topic. It is important to explore the acceptance of flawed products more broadly detached from foods in order to find out if anthropomorphism is a good strategy to support the acceptance of flawed products in general. This is why it is advised to extend this study by further product categories. To cover a broader range and to get deeper insights, it is advised to combine hedonic and utilitarian products. Hedonic products are goods that are consumed affective and utilitarian products are consumed goal-oriented (Dhar, 1999). These two different types of products will lead to different insight bringing broader knowledge into the acceptance of anthropomorphism on flawed products in general.

division between low and high involvement. For people being low involved, it is predicted that they accept flawed products more when they are anthropomorphized because low involved people, following the peripheral route (Petty et al. 1983), react to easy accessible cues like anthropomorphism (Basfirinci & Çilingir, 2010). This would probably lead to the support of hypothesis 2. When people are low involved anthropomorphism serves as peripheral cue (Lee, 2010), leading to an attitude chance towards the anthropomorphized product. This attitude change would lead to a higher acceptance of the anthropomorphized, flawed product compared to people with high involvement.

In general it is suggested to distribute the study for example via Amazon Mechanical Turk (MTurk). This will help to get a broader age-range and people from different social classes which will probably help to cover a wider range of participants compared to only young students as in this research. With a higher average age it is expected that participants like the vegetables more since older people rather eat more healthy food compared to young people (Bushak, 2014). This will probably increase the acceptance of anthropomorphized, flawed vegetables even more, leading to a higher significance for hypothesis 1. To make sure that concentration span can be excluded from the limitations, further research should distribute the study in changing order with other studies or progress detached from other studies.

6. CONCLUSION

REFERENCES

Aggarwal, Pankaj and McGill, Ann L. (2007), “Is that car smiling at Me? Schema Congruity as a Basis for Evaluating Anthropomorphized products”, Journal of Customer Research, 34, 468-479

———, ——— (2012), “When Brands Seem Human, Do Humans Act Like Brands? Automatic Behavioral Priming Effects of Brand Anthropomorphism”, Journal of Consumer Research, 39, No. 2, 307-323

Baker, Sara M. and Richard E. Petty (1994), “Majority and Minority Influence: Source Position Imbalance as a Determinant of Message Scrutiny,” Journal of Personality and Social Psychology, 67 (1), 5–19.

Basfirinci, Cigdem and Cilingir, Zuhal (2015), “Anthropomorphism and Advertising Effectiveness: Moderating Roles of Product Involvement and the Type of Consumer Need”, Journal of Social and Administrative Sciences, 2, 108 – 131

Bian, Xuemei and Moutinho Luiz (2009), “The role of brand image, product involvement, and knowledge in explaining consumer purchase behaviour of counterfeits - Direct and indirect effects”, European Journal of Marketing, 45 (1/2), 191-216

Breines, Juliana (2015), “5 'Flaws' That Just Make You More lovable”, Psychology today, Online: June 30, 2015, (accessed November 28, 2016), [available at: https://www.psychologytoday.com/blog/in-love-and-war/201506/5-flaws-just-make-you-more-lovable]

Bushak, Lecia (2014), “Age 36 Brings A Health 'Wake Up Call' To Many, Thanks To Doctor Warnings And Unflattering Pictures”, The Medical Daily, Online: July 23, 2014, (accessed: January 3, 2017), [available at http://www.medicaldaily.com/age-36-brings-health-wake-call-many-thanks-doctor-warnings-and-unflattering-pictures-294614]

Campbell, Christian (2007), “International Product Liability”, Austria: Yorkhill Law Publishing

Chandler, Jesse and Schwarz, Norbert (2010), “Use does not ragged the fabric of friendship: Thinking of objects as alive makes people less willing to replace them” Journal of Consumer Psychology, 20, 138-145

Chassan, Jacob B. (1979). Research Design in Clinical Psychology and Psychiatry. New York: Irvington Publishers Inc.

Chhabria, Rachna (2012), “Why do we like imperfect characters?”, Online: April 24, 2012,

(accessed December 4, 2016), [available at

http://rachnachhabria.blogspot.nl/2012/04/why-do-we-like-imperfect-characters.html]

Cicatiello, Clara, Franco, Silvio, Pancino, Barbara, and Blasi, Emanuele (2016), “The value of food waste: An exploratory study on retailing”, Journal of Retailing and Consumer Services, 30, 96–104

Dhar, Ravi (2000), “Consumer Choice between Hedonic and Utilitarian Goods”, Journal of Marketing Research, XXXVll, 60-71

Epley, Nicholas, Waytz, Adam, and Cacioppo John T. (2007), “On seeing human: A three-factor theory anthropomorphism”, Psychological Review, 114(4), 864-886

———, Akalis, Scott, and Cacioppo John T. (2008), “When We Need A Human: Motivational Determinants of Anthropomorphism”, Social Cognition, 26 (2), 143-155

Fennis, Bob M. and Stroebe, Wolfgang (2016), “The psychology of advertising” (2nd Ed.). New York, NY: Routledge

Fuchs, Christoph, Schreier, Martin, and van Osselaer, Stijn M. J. (2015),”The Handmade Effect: What’s Love Got to Do with It?”, Journal of Marketing, 79, 98-110

Gardner, Wendi L. and Knowles, Megan L. (2008), “Love makes you real: favorite television characters are perceived as “real” in a social facilitation paradigm”, Social Cognition, 26 (2), 156-168

Garrone, Paola, Melacini, Marco, Perego, Alessandro, and Sert, Sedef (2016), “Reducing food waste in food manufacturing companies”, Journal of Cleaner Production, 137, 1076-1085

Glaser, Tina et al. (2015), “Lateral Attitude Change”, Personality and Social Psychology Review 2015, 19(3), 257–276

Goldenberg, Suzanne (2016), “Half of all US food produce is thrown away, new research suggests”, The Guardian, Online, July 13, 2916, (accessed December 1, 2016), [accessed at https://www.theguardian.com/environment/2016/jul/13/us-food-waste-ugly-fruit-vegetables-perfect]

November 30, 2016) [available at https://www.nrdc.org/sites/default/files/wasted-food-IP.pdf ]

Hart, Phillip M., Jones, Shawn R., and Royne, Marla B. (2013), “The human lens: How anthropomorphic reasoning varies by product complexity and enhances personal value”, Journal of Marketing Management, 29 (2), 105-121

Howard, Jennifer (2011), “Are We Addicted to the Idea of Perfection?”, The Huffington Post, Online: September 16, 2011, (accessed January 7, 2017), [available at

http://www.huffingtonpost.com/jennifer-howard-phd/beauty-and-wisdom_b_954404.html]

Intermarché (2014), “Intermarché - Inglorious fruits & vegetable”, Itm.marcelww.com, (accessed: October 11, 2016), [available at http://itm.marcelww.com/inglorious/]

Katigbak, Karryl (2014), “Selling Imperfection: Analysis on the Inglorious Fruits and Vegetables Campaign”, Broadcast Critism – Worldpress, Online: November 18, 2014,

(accessed October 23, 2016), [available at

https://broadcastcriticism.wordpress.com/2014/11/18/selling-imperfection-analysis-on-the-inglorious-fruits-and-vegetables-campaign/]

Kim, Jin-Sun, Kaye, Judy, and Wright, Lore K. (2001), “Moderating and Mediating Effects in Causal Models”, Issues in Mental Health Nursing, 22, 63–75

Landwehr, Jan R., McGill, Ann L., and Herrmann, Andreas (2011), “It’s got the look: The effect of friendly and aggressive “facial” expressions on product liking and sales”, Journal of Marketing, 75(3), 132-146.

Lee, Eun-Ju (2010), “The more humanlike, the better? How speech type and users’ cognitive style affect social responses to computers”, Computers in Human Behavior, 26, 665-672

Lipinski, Brian, Hanson, Craig, Lomax, James, Kitinoja, Lisa, Waite, Richard, and Searchinger, Tim (2013), “Reducing Food Loss And Waste”, Working Paper, World Resource Institute, Online: June 1, 2013, (accessed November 30, 2016), [available at http://www.wri.org/publication/reducing-food-loss-and-waste]

Malhotra, Naresh K. (2011), ‘Marketing Research – An applied orientation’, 6th edition, New Jersey, United Stated of America, Pearson Education Inc.

Petty, Richard E. and Cacioppo, John T. (1981), “Issue Involvement As a Moderator of the Effects on Attitude of Advertising Content and Context”, Advances in Consumer Research, 8, 20-24

———, ——— (1986), “The Elaboration Likelihood Model of Persuasion”, Advances in Experimental Social Psychology, 19, 123-205

———, ———, and Schumann, David (1983), “Central and Peripheral Routes to Advertising Effectiveness: The Moderating Role of Involvement”, Journal of Consumer Research, 10, 135-146

Radwan, Farouk M. (20XX), “Striving for perfection psychology”, (accessed December 2,

2016), [accessed at

Schwartz, Barry. (2004), “The paradox of choice: why more is less”, New York, Ecco

Segal-Caspi, Lihi, Roccas, Sonia, and Sagiv, Lilach (2012), “Don’t Judge a Book by Its Cover, Revisited: Perceived and Reported Traits and Values of Attractive Women”, Psychological Science, 23(10), 1112–1116

Slavin, Joanne L. and Lloyd, Beate (2012), “Health Benefits of Fruits and Vegetables”, Society for Nutrition, Adv. Nutr. 3, 506–516

Smithers, Rebecca (2013), “Almost half of the world's food thrown away, report finds”, The Guardian, Online: January 10, 2013, (accessed November 30, 2016), [available at https://www.theguardian.com/environment/2013/jan/10/half-world-food-waste]

Stevenson, Angus (2016): „flaw - definition of flaw in English - Oxford Dictionaries“, Oxford Dictionaries - English, (accessed October 19, 2016), [available at https://en.oxforddictionaries.com/definition/flaw]

Triantos, Alexandros, Plakoyiannaki, Emmanuella, Outra, Evaggelia, and Petridis, Nikilaos (2016) “Antropomophic Packaging: Is ther Life on Mars?” European Journal of Marketing, 50 (1/2), 260 – 275

Troutman, C. Michael and Shanteau, James (1976), “Do consumers evaluate products by adding or averaging attribute information?”, Journal of Consumer Research, 3, 101-106

VonRiesen, R. Dale and Herndon, Neil C. (2011), “Consumer Involvement with the Product and the Nature of Brand Loyalty”, Journal of Marketing Channels, 327-352

Appendix 1: Survey

The survey was carried out with four different conditions. In each condition participants could choose 3 out of 6 vegetables. The vegetables were:

• Potato • Carrot • Eggplant • Tomato • Sweet pepper • Mushroom

Depending on the condition and the selected vegetables, different pictures w the screen. As example, the pictures of the tomato are shown for each condition: Condition 1: Non-anthropomorphized / Non

Condition 2: Anthropomorphized / Flawed

APPENDIX

d out with four different conditions. In each condition participants could The vegetables were:

Depending on the condition and the selected vegetables, different pictures w the pictures of the tomato are shown for each condition: anthropomorphized / Non-flawed (using the example of tomatoes)

Condition 2: Anthropomorphized / Flawed (using the example of tomatoes)

d out with four different conditions. In each condition participants could

Depending on the condition and the selected vegetables, different pictures were displayed on the pictures of the tomato are shown for each condition:

(using the example of tomatoes)

Condition 3: Anthropomorphized / Non

Condition 4: Non-anthropomorphized / Flawed

For each displayed vegetables, • Overall, how much do y

• How much do you like this vegetable in general?

• If you saw this vegetable in the supermarket, how likely would you purchase it? • How much would you enjoy eating this vegetable?

After seeing all vegetables, the f General Liking:

• How much do you like eating vegetables in general?

Condition 3: Anthropomorphized / Non-flawed (using the example of tomatoes)

anthropomorphized / Flawed (using the example of tomatoes)

, the following questions needed to be answered: Overall, how much do you like the displayed vegetable?

How much do you like this vegetable in general?

If you saw this vegetable in the supermarket, how likely would you purchase it? How much would you enjoy eating this vegetable?

After seeing all vegetables, the following questions were asked:

How much do you like eating vegetables in general?

(using the example of tomatoes)

(using the example of tomatoes)

the following questions needed to be answered:

Perceived Quality:

• I think the vegetable would taste good

• I think the vegetable would taste the same compared to regular looking ones • I think the vegetables are of good quality

• I think the vegetables are fresh • I think the vegetables are healthy • I think the vegetables are organic • I think the vegetables are natural Involvement:

• Vegetables interest me

• I am not at all familiar with vegetables

• Vegetables help me attain the type of lifestyle I strive for

• I rate vegetables as being of highest importance to me personally • I could talk about vegetables a long time

Empathy:

• I know everything I need to know about the vegetables to make my buying decision • The vegetables in the marketing campaign from earlier were well-explained

• I felt happy when I saw the vegetables in the marketing campaign from earlier • I felt sad when I saw the vegetables in the marketing campaign from earlier Feeling:

Imagine you see one of the vegetables of the previously shown campaigns in a supermarket and they were left behind or not chosen at all compared to other vegetables. How would you feel?

• I would feel sad

• I would feel empathy (towards the vegetable) • I would feel angry

• I would not care Mood Measure:

• How do you feel at the moment: o Bad - Good o Sad - Happy o Negative - Positive o Bored - Excited Campaign:

• Were there any vegetables you did not like that you saw earlier? If so, which one(s)? • How familiar are you with other campaigns about ugly fruits and vegetables (e.g.,

Buitenbeentjes, Kromkommer, Inglorious Vegetables)? • How do you feel at this moment?

Demographics:

• Questions about demographics were asked

Appendix 2: Hypothesis 1 – ANOVA - 2 (humanization: anthropomorphized vs. non-anthropomorphized product) x 2 (vegetable: flawed vs. non-flawed vegetable) between subject

Dependent Variable: Acceptance of vegetables Source

Type III Sum of

Squares df Mean Square F Sig.

Corrected Model 83,599a 3 27,866 35,098 ,000

Intercept 5070,635 1 5070,635 6386,541 ,000

Flaw / No Flaw 74,757 1 74,757 94,157 ,000

Anthro /No Anthro 6,356 1 6,356 8,006 ,005

Flaw/No Flaw * Anthro/No Anthro 2,297 1 2,297 2,893 ,091 Error 150,058 189 ,794 Total 5382,815 193 Corrected Total 233,657 192

Appendix 3: POST-HOC TEST (LSD) - Hypothesis 1

Appendix 4: ANCOVA - 2 (humanization: anthropomorphized vs.

non-anthropomorphized product) x 2 (vegetable: flawed vs. non-flawed vegetable) between subject with CUTENESS as Covariate

Dependent Variable: Acceptance of vegetables Source

Type III Sum

of Squares df Mean Square F Sig.

Corrected Model 91,546a 4 22,886 30,277 ,000 Intercept 299,027 1 299,027 395,585 ,000 Cuteness High/Low 7,946 1 7,946 10,512 ,001 Flaw 75,338 1 75,338 99,665 ,000 Anthro 3,749 1 3,749 4,960 ,027 Flaw * Anthro 2,398 1 2,398 3,172 ,077 Error 142,111 188 ,756 Total 5382,815 193 Corrected Total 233,657 192 a. R Squared = ,392 (Adjusted R Squared = ,379)

Dependent Variable: Acceptance of vegetables

(I) Condition (J) Condition

Appendix 5: ANCOVA - 2 (humanization: anthropomorphized vs.

non-anthropomorphized product) x 2 (vegetable: flawed vs. non-flawed vegetable) between subject with MOOD as Covariate

Dependent Variable: Acceptance of vegetables Source

Type III Sum of

Squares df Mean Square F Sig.

Corrected Model 84,818a 4 21,204 26,783 ,000

Intercept 505,164 1 505,164 638,076 ,000

Mood_High/Low 1,218 1 1,218 1,539 ,216

Flaw / No Flaw 74,711 1 74,711 94,368 ,000

Anthro /No Anthro 6,249 1 6,249 7,893 ,005

Flaw/No Flaw * Anthro/No Anthro 2,433 1 2,433 3,073 ,081 Error 148,839 188 ,792 Total 5382,815 193 Corrected Total 233,657 192 a. R Squared = ,363 (Adjusted R Squared = ,349)

Appendix 6: ANCOVA - 2 (humanization: anthropomorphized vs.

non-anthropomorphized product) x 2 (vegetable: flawed vs. non-flawed vegetable) between subject with CUTENESS and MOOD as Covariate

Dependent Variable: Acceptance of vegetables Source

Type III Sum of

Squares df Mean Square F Sig.

Appendix 7: Hypothesis 2 – ANOVA - 2 (humanization: anthropomorphized vs. non-anthropomorphized product) x 2 (vegetable: flaw vs. non-flawed vegetable) +

Involvement between subject

Dependent Variable: Acceptance of vegetables Source

Type III Sum

of Squares df Mean Square F Sig. Corrected Model 86,469a 7 12,353 15,526 ,000 Intercept 4993,016 1 4993,016 6275,719 ,000 Flaw / No Flaw 72,672 1 72,672 91,341 ,000 Anthro / No Anthro 6,902 1 6,902 8,675 ,004 Involvement_High/Low 1,457 1 1,457 1,831 ,178 Flaw/No Flaw * Anthro/No Anthro 1,992 1 1,992 2,504 ,115 Flaw/No Flaw * Involvement_High/Low ,638 1 ,638 ,802 ,372 Anthro/No Anthro * Involvement_High/Low ,020 1 ,020 ,025 ,873 Flaw/No Flaw * Anthro/No Anthro * Involvement_High/Low ,750 1 ,750 ,942 ,333 Error 147,188 185 ,796 Total 5382,815 193 Corrected Total 233,657 192 a. R Squared = ,367 (Adjusted R Squared = ,343)

Appendix 8: Table - Median Split Involvement

Involvement

Appendix 9: POST-HOC TEST (LSD) – Hypothesis 2 – Involvement Low

Dependent Variable: Acceptance of vegetables

(I) Condition (J) Condition

Mean Difference (I-J) Std. Error Sig. 95% Confidence Interval Lower Bound Upper Bound No Anthro/No Flaw Anthro/Flaw ,7193* ,22540 ,002 ,2719 1,1668 Anthro/No Flaw -,3228 ,21558 ,138 -,7507 ,1052 No Anthro/Flaw 1,2005* ,21729 ,000 ,7691 1,6318 Anthro/Flaw No Anthro/No Flaw -,7193* ,22540 ,002 -1,1668 -,2719 Anthro/No Flaw -1,0421* ,21016 ,000 -1,4593 -,6249 No Anthro/Flaw ,4811* ,21190 ,025 ,0604 ,9018 Anthro/No Flaw No Anthro/No Flaw ,3228 ,21558 ,138 -,1052 ,7507 Anthro/Flaw 1,0421* ,21016 ,000 ,6249 1,4593 No Anthro/Flaw 1,5232* ,20143 ,000 1,1233 1,9231 No Anthro/Flaw No Anthro/No Flaw -1,2005* ,21729 ,000 -1,6318 -,7691 Anthro/Flaw -,4811* ,21190 ,025 -,9018 -,0604 Anthro/No Flaw -1,5232* ,20143 ,000 -1,9231 -1,1233 Based on observed means.

Appendix 10: POST-HOC TEST (LSD) – Hypothesis 2 – Involvement High

Dependent Variable: Acceptance of vegetables

(I) Condition (J) Condition

Mean Difference (I-J) Std. Error Sig. 95% Confidence Interval Lower Bound Upper Bound No Anthro/No Flaw Anthro/Flaw ,9925* ,29031 ,001 ,4157 1,5692 Anthro/No Flaw -,0301 ,28397 ,916 -,5943 ,5340 No Anthro/Flaw 1,6836* ,30636 ,000 1,0749 2,2922 Anthro/Flaw No Anthro/No Flaw -,9925* ,29031 ,001 -1,5692 -,4157 Anthro/No Flaw -1,0226* ,29560 ,001 -1,6099 -,4353 No Anthro/Flaw ,6911* ,31718 ,032 ,0609 1,3212 Anthro/No Flaw No Anthro/No Flaw ,0301 ,28397 ,916 -,5340 ,5943 Anthro/Flaw 1,0226* ,29560 ,001 ,4353 1,6099 No Anthro/Flaw 1,7137* ,31138 ,000 1,0951 2,3323 No Anthro/Flaw No Anthro/No Flaw -1,6836* ,30636 ,000 -2,2922 -1,0749 Anthro/Flaw -,6911* ,31718 ,032 -1,3212 -,0609 Anthro/No Flaw -1,7137* ,31138 ,000 -2,3323 -1,0951 Based on observed means.

Appendix 11: Frequencies Involvement

Means Frequency Percent

Appendix 12: Quality – ANOVA

Dependent Variable: Perceived Quality Source

Type III Sum of

Squares df Mean Square F Sig.

Corrected Model 5,356a 3 1,785 2,337 ,075 Intercept 6136,614 1 6136,614 8032,833 ,000 Flaw 1,829 1 1,829 2,394 ,123 Anthro 1,892 1 1,892 2,476 ,117 Flaw * Anthro 1,791 1 1,791 2,344 ,127 Error 144,385 189 ,764 Total 6317,000 193 Corrected Total 149,741 192 a. R Squared = ,036 (Adjusted R Squared = ,020)

Appendix 13: Means Quality by condition

Dependent Variable: Perceived Quality

Flaw Anthro Mean N

No Flaw No Anthro 5,7431 48

Anthro 5,7484 53

Flaw No Anthro 5,3551 46

Appendix 14: POST-HOC TEST (LSD) – Quality

Dependent Variable: Perceived Quality

(I) Condition (J) Condition

Mean Difference (I-J) Std. Error Sig. 95% Confidence Interval Lower Bound Upper Bound No Anthro/No Flaw Anthro/Flaw -,00332 ,18034 ,985 -,3591 ,3524 Anthro/No Flaw -,00537 ,17415 ,975 -,3489 ,3382 No Anthro/Flaw ,38798* ,18034 ,033 ,0322 ,7437

Anthro/Flaw No Anthro/No Flaw ,00332 ,18034 ,985 -,3524 ,3591

Anthro/No Flaw -,00205 ,17613 ,991 -,3495 ,3454

No Anthro/Flaw ,39130* ,18225 ,033 ,0318 ,7508

Anthro/No Flaw No Anthro/No Flaw ,00537 ,17415 ,975 -,3382 ,3489

Anthro/Flaw ,00205 ,17613 ,991 -,3454 ,3495

No Anthro/Flaw ,39336* ,17613 ,027 ,0459 ,7408

No Anthro/Flaw No Anthro/No Flaw -,38798* ,18034 ,033 -,7437 -,0322

Anthro/Flaw -,39130* ,18225 ,033 -,7508 -,0318

Anthro/No Flaw -,39336* ,17613 ,027 -,7408 -,0459

*. The mean difference is significant at the 0.05 level.

Appendix 15: Means Acceptance divided by age (excluding 1 participant age 31)

Dependent Variable: Acceptance of vegetables

How old are you? Mean: Acceptance Overall Mean N