When do Consumers accept imperfect products? Anthropomorphism & perfection/imperfection belief as cues

(1)

1

When do Consumers accept imperfect products?

Anthropomorphism &

perfection/imperfection belief as cues

The effect of anthropomorphism and

perfection/imperfection belief on the acceptance on

imperfect products

(2)

2

When do Consumers accept imperfect products?

Anthropomorphism & perfection/imperfection belief as cues

The effect of anthropomorphism and perfection/imperfection belief on the acceptance on

imperfect products

Department of Marketing

University of Groningen

MSc Marketing Master Thesis

January 2018

1st_{Supervisor: Dr. Jing Wan}

2nd_{Supervisor: Dr. Marijke Leliveld}

Welfred Cheng

S3070832

Welfred.cheng@gmail.com +31 6 25 24 89 67

(3)

3

Abstract

The present research aims to provide a science ground to reduce the waste of the imperfect yet fully functional products in retail sectors. Imperfect here refers to a superficial and minor imperfection on products which do not affect the functionality of the products at all. The effect of

anthropomorphism on the acceptance of the flawed products was investigated. In addition, the present research addressed the concepts of lay beliefs on perfection and imperfection as the possible influences on this effect. Data was collected through an online panel named Amazon Mechanical Turk from 203 respondents by displaying a candle as the study object for

anthropomorphism and acceptance and providing excerpts from prior research to convince the lay beliefs. A between subject study with 2 (anthropomorphism vs. non anthropomorphism) by 2 (perfection belief vs. imperfection belief), with four conditions was conducted. By quantitative analysis, results showed that the only marginally significant effect is the interaction effect of anthropomorphism and beliefs, while main effects were not significant, which may suggest that there were other concepts which might interfere with anthropomorphism on the acceptance on flawed products. Further analysis showed that the main effect of lay belief was significant in the male sample. Explanations of such results as well as the limitation of the present study and the recommendations to the future research was discussed in the Discussion section.

(4)

4

Preface

Completing a Master Thesis is never easy. Completing a Master Thesis in a foreign environment in which everything was unfamiliar at first makes it harder. I still remember that the piece was kept written, revised, deleted, and being started over again, repeatedly; not to say spending hours or even days for only thinking on that only one concept and getting the idea to proceed. I would be confident at this moment to understand why there is this Master Thesis to finalise the Master Programme. It truly demonstrates the student’s knowledge and skills in handling big academic projects as such; and most importantly, showcases the student’s effort and motivation put to complete the Master thesis. I would like to take this opportunity to thank Dr. Jing Wan, who advised and commented on my approach throughout the whole thesis journey. Her suggestions and insights have been valuable and motivated me during the thesis writing process, needless to say her

(5)

5

Individual and institutions are likely to avoid sharing mistaken information or errors with others (Edmondson, 1996; Michael, 1973; Stefaniak & Robertson, 2010; Uribe, Schweikhart, Pathak, & Marsh, 2002). Hence, a shopper may not buy a product consisting of errors. Any kind of mistake or error on a product, should it concern the initial functionality of the product or not, would be perceived as defective and often disregarded. It is simply because the products do not match consumers expectations of aesthetic perfection. In turn, sellers also tend to redirect customer attention from the product mistake to the other positive attributes of the product (Kotler & Mantrala, 1985).

(7)

7 & Macnaughton, 2010) and blemished, wrong sized or misshapen products (Stuart, 2009).

Consumers are unwilling to purchase a product which has a damaged packaging even if the product itself is fully functionable or edible, because they see the superficial packaging damage as a potential source of contamination to the product (White, Lin, Dahl, & Ritchie, 2016). Not only would the consumers tend to get rid of or forgo the purchase of imperfect food, the retailers are reluctant to sell imperfect food to fulfil the aesthetic standards of the food appearance of consumers so they will accept the food (Aschemann-Witzel, De Hooge, Amani, Bech-Larsen, & Oostindjer, 2015; Buzby & Hyman, 2012). Another example to look at is the clothing industry. The clothing industry, especially fast fashion, is based on extremely fast cycle of fashion and consumer unsustainable desires. Short term use and frequent clothing replacement could increase textile waste which severely increase the burden to the environment (Niinimäki & Hassi, 2011). In Britain, there has been already 2.3 million tonnes of textile waste annually (Department for Environment, Food & Rural Affairs, 2008). The clothing consumption is expected to increase by 23% (Nurmela, 2009), which consequently may lead to increase in textile waste. It is not uncommon to see consumers throw away or replace clothes simply because this clothing has some wrinkles or tiny threads sticking out.

With the United Nations (2015) recently establishing the Sustainable Development Goals, in which one of them is to ensure the sustainable consumption and production patterns, the

(8)

8 Prior research found that mistakes are undesirable and often lead to negative inferences about the entity which made the mistake (Chesney & Su, 2010; Michael, 1973; Palmer, Simmons, & De Kervenoael, 2010). However, Reich, Kupor, and Smith (2017), in a recent research found out that a product consisting of mistakes could be viewed more unique compared to the other identical products. In other words, a shopper would prefer the product with mistakes over the other perfectly made products.

(9)

9 time to feature a smiley face (Arnheim, 1969). Products are sometimes even perceived as having consciousness (Gilmore, 1919), personality (Aaker, 1997; Keller, Parameswaran, & Jacob, 2011), and relationship (Aggarwal, 2004; Fournier, 1998; Muniz & O'Guinn, 2001).

The tolerance that people hold on judging the extent a person could make mistakes may extend to judging the extent an anthropomorphised product could have been made with mistakes. It is because the anthropomorphised product is seen as a person, and people would apply the standards they judge a person to that anthropomorphised product. Lanier JR et al. (2013) revealed that interaction between human and anthropomorphised objects could be two way and an

anthropomorphised object could been seen as an acquaintance, in which judgements held only towards human could also be applied to that anthropomorphised object as an acquaintance. Lay belief has been shown to affect judgements and behaviour in a number of domains (Mukhopadhyay & Johar, 2005). For example, beliefs that people hold to judge things around them depend on the implicit theories they hold (Butler, 2000), in which the theories may have been learnt from everyday experiences (Ross & Nisbett, 2011), environments (Morris, Menon, & Ames, 2001), or simply by self observation (Bem, 1967). It is important to know that sometimes the belief that one person hold could be contrasting to another. One example could be the belief to pursue perfection on human. While Roedell (1984) suggested that pursuing perfection itself is needed and fundamental to

personal development, Pacht (1984) argued that perfectionism would make everyone the same as it would eliminate the mistakes that make a person unique. Among all the different beliefs people hold to make judgements, the perfection belief could be important to determine the consumer

(10)

10 In order to help companies find a way to better treat the products made with mistakes to reduce wastage to attain the Sustainable Development Goals, the present research is to answer whether anthropomorphising imperfect products will affect the consumer acceptance towards the imperfect products, with respect to the different beliefs that the consumer holds.

This thesis is divided into four parts, comprising of Literature Review, Methodology, Result, and Discussion. The first part, the Literature Review, will review the various theories of

anthropomorphism, the concept of imperfect, and the concepts about lay beliefs and beliefs people hold towards human beings’ perfection and imperfection. It is to review previous relevant studies and consequently come up with two research hypotheses to come up with a solution to the above challenge in a scientific way. The Methodology part of this study is to explain the data collection method and the study set up. The Result section will employ quantitative research methods to gain data insights and examine the hypotheses proposed in the Literature Review. The last part will discuss the research findings of this study, acknowledge the limitations of this research, and draw research conclusions.

1.1 Literature Review

The following section is to provide a theoretical background for concepts adapted for the present research, namely anthropomorphism, imperfect, and lay belief and belief about perfection and imperfection.

(11)

11 The word ‘anthropomorphism is derived from Greek words. Anthropo means human and morphe means shape or form. From its literal meaning, anthropomorphism could be understood as that nonhuman entities are given a human shape or form. Epley et al. (2007) defined anthropomorphism as the tendency to attribute the real or imagined behaviour of nonhuman agents with humanlike characteristics, motivations, intentions, or emotions. In other words, it means that

anthropomorphism does not only involve behavioural descriptions of imagined or observable actions imbuement to the non living, but also the representation of an agent’s mental or physical

characteristics using human like characters. To this understanding, anthropomorphism could be said as attributing humanlike properties, characteristics, or mental states to nonhuman agents and objects. From a person’s perception to a nonhuman object, it can involve the perception of humanlike mind, such as, conscious experience, metacognition, and intentions (Gray, Gray, & Wegner, 2007), and humanlike emotional states (Leyens et al., 2003).

It has been long noted that academicians from various disciplines found people tend to see nonhuman agents as humanlike (Darwin & Prodger, 1998; Feuerbach, 1873; Freud, 2012; Hume, 1957). Prior research found that people are able to provide anthropomorphic descriptions of

(12)

12 MacInnis and Folkes (2017) also suggested that by giving a human name and gender to an object, anthropomorphism could be more likely to occur. People could make judgements for humans, like in interpersonal relationships, to an anthropomorphised object since connections to a particular object or product will grow over time and lead people to form a sense of attachment that parallels

interpersonal attachment patterns (Ball & Tasaki, 1992; Bowlby, 1980; Wallendorf, & Arnould, 1988).

While people can humanise both human and nonhuman targets, however, the term

anthropomorphism only refers to humanising nonhuman agents (Waytz, Cacioppo, & Epley, 2010; Waytz, Epley, & Cacioppo, 2010). In other words, in this research, since anthropomorphism to products is discussed, the term anthropomorphism and humanisation are interchangeably used.

Hume (1957) suggested that anthropomorphism is a universal phenomenon that people

anthropomorphise objects and entities naturally and even unconsciously. The tendency of people to anthropomorphise could be explained by the need for making a better sense of the world around them (Aggarwal & McGill, 2007; Stewart, 1993). People use what they are familiar with – to ascribe human like characteristics which is their knowledge of themselves to entities – to account for better outcomes and things that they know less about. In addition, Guthrie and Guthrie (1995) suggested that anthropomorphism comforts people by providing relationships or companionship.

(13)

13 2011; Puzakova, Kwak, & Rocereto, 2013), it is then assumed that consumers would apply the way in which they comprehend a person to understand an anthropomorphised product.

The use of anthropomorphism has been widely used in marketing (Brown, 2010; Lanier JR, 2013). Mentioned earlier, both brands and products are and can be anthropomorphised for its marketing purposes. Marketers believe that anthropomorphising the brand or product can allow consumers to focus more on its personal characteristics, in other words – non utility relevant uniqueness, than its functionality; to avoid intense competition of the similar brands and products. Prior research has shown that specific anthropomorphic design of a product can influence consumers product

preference (Aggarwal & McGill, 2007; Landwehr, McGill, & Herrmann, 2011). For instance, Aggarwal and McGill (2007) found out that consumers like the anthropomorphised product more in the condition when its appearance resembles the expected human physical features. However, these findings about the effect of anthropomorphism on the product were limited to the situation where the study specified such as the extent of the anthropomorphism. Wen Wan, Peng Chen and Jin (2017), found that anthropomorphism increases consumers’ preference for products with superior appearance. In the opposite, the present research is to investigate whether anthropomorphism affects consumers acceptance for products with mistakes, in the situation where consumers see the product as a human.

1.1.2 Products with Mistakes

(14)

14 neutral or negative. For instance, if a baker mistakenly bakes a bread for five extra minutes than it is intended to be: the bread may be enhanced, which is a positive outcome; the bread may be

unaffected, which is a neutral outcome; or the bread may have worsened; all depending on consumer preferences (Reich et al., 2017). A product made with mistakes can be described as mismatching to design regulations or unfitting in the overarching mass production according to regulations that the product is carried into circulation (Campbell, 2007). In the case of a tangible product, a mistake can be a product packaging imperfection, which is more commonly called a flawed product, that does not deal will the product utility at all (White et al., 2016), although that mistake is likely to lead to a negative outcome where a consumer may have negative reactions, there are conditions where a consumer may have a positive or neutral reaction towards the product packaging imperfection. For the purpose of the present research, a product with mistakes, imperfect product or flawed product, which were termed by different academicians, refer to the products with minor yet recognisable imperfections which do not affect the overall functionality of the product. In addition, a flawed product should have inconsistency in terms of appearance compared to other identical products within the same group yet remain as usable as the other perfect products in terms of appearance. While product packaging imperfection can be an example for flawed products, others include the variation of colours on parts of a product or scratches on the surface of a products. Flawed products are in general undesirable since people have a tendency to strive for perfection with status (Schwartz, 2004). Schwartz (2004) further explained that the only way for people to be the best is to have the best. In this case, flawed product will not likely to be considered by

consumers as they would prefer a perfect product.

(15)

15 themselves to a certain extent, flawed product are therefore chosen as they are able to remind consumers of themselves (Gardner & Knowles, 2008); flawed products can cause feelings of empathy with the consumer, especially in handmade products (Snyder, 1992), and handmade products, with superficial flaws, receive a higher degree of liking by consumers, compared those without, because of the effort of maker and love that the consumer can feel behind the flawed product (Fuchs, Schreier, & Van Osselaer, 2015). While evaluation basis between products can change depending on different product features and consumer perceptions towards the product, people also use different sets of norms as a basis to evaluate humans compared to objects. Products are evaluated based on the information of quality and the facts about it (Troutman & Shanteau, 1976), on the other hand, people are usually evaluated based on their physical cues, such as smiles (Wilson & Eckel, 2006). In other words, anthropomorphism can enable such physical cues to be built by putting a smiley face on an object to allow the object to be evaluated as a person. This suggests that anthropomorphised flawed products, could be more accepted compared to non

anthropomorphised flawed products, since the association with humans with the product may let people focus more on the human evaluation part of the anthropomorphised product in which the evaluation of the facts about the anthropomorphised product would be less processed.

Numerous research found out different effects with anthropomorphism on products in various conditions, however, there is little about the effect of anthropomorphism with imperfect products. The present research is interested to fill this research gap in addition to potentially alleviate the issue introduced in the beginning of the thesis, which is to reduce the waste product which are still useful.

(16)

16 Previous sections explained that there is a possibility where people could evaluate an

anthropomorphised product in its human characteristics. It is thus important to understand the process how people evaluate things around them, and hence know how it might affect the people evaluation on a flawed product in an anthropomorphic form.

Rokeach (1973) defined a value as an enduring belief that a specific mode of conduct and end state is personally preferable to its opposite. Values guide individuals about what to do in different situations (Kahle, 1980). They are similar to attitudes in a way that both are emerged continuously from assimilation, accommodation, organisation, and integration of environmental information to promote with the environment favourable to the preservation of the optimal functioning (Kahle, 1983). Since values serve as basic prototypes from which attitudes and behaviours are

manufactured, Homer and Kahle (1988) defined that values lead to attitudes, then form behaviours. From the above, it could be concluded that different enduring beliefs, which form values, on the same topic could lead to different behaviours, in which the different behaviours could be translated as the decision of whether to purchase the anthropomorphised product of not. Moreover, a rather short term, named lay belief, as a result of persuasion for instance, can also influence consumer’s buying decisions (Friestad & Wright, 1994, 1995).

1.1.4 Lay Theories

(17)

17 lay theory held in another group could be different as they share different knowledge. They define the lay theory as an organised knowledge that directs behaviour, judgements and evaluations. Like scientific theories, lay theories also serve the epistemic function of better sense making (Hong, Levy, & Chiu, 2001). In case of unfamiliarity, people would tend to use lay theories to understand events and make inferences about social reality (Lickel, Hamilton, & Sherman, 2001). According to Dweck, Chiu and Hong (1995), a figurative way to understand lay theories is see it as a lens which people use to interpret events and make sense of their own and other people’s behaviour.

There has been numerous research on various domains: Zedelius, Müller and Schooler (2017) concluded explorations on lay theories about human psychological attributes, supernatural, mental and physical health. It is found that the most relevant lay theories to the present research is the lay theories about intrusive thoughts. Intrusive thoughts are spontaneous stimulus or task unrelated thoughts which occurrence is experienced as unwanted (Rachman, 1981). Intrusive thoughts can be negative, neutral or positive in content. In daily life, people can routinely experience intrusive thoughts (Bywaters, Andrade, & Turpin, 2004; Clark, 2005; Rachman & De Silva, 1978; Sarason, Pierce, & Sarason, 1996; Wegner & Pennebaker, 1993), sometimes without being consciously aware of them (Baird, Smallwood, Fishman, Mrazek, & Schooler, 2013). Consequently, as for the present research scope, the belief on perfection as intrusive thoughts in terms of lay theories could be tested with regards to the anthropomorphised flawed products.

1.1.5 Perfection Belief

(18)

18 1956; Hollender, 1965; Homey, 1950; Missildine, 1963; Pacht, 1984). Within, perfectionism was shown to be operationalised in some studies (Burns, 1980; Jones, 1968); some suggested that perfectionism might develop as a personality style (Hamachek, 1978; Hollender, 1965), which the perfectionistic personality style could direct and affect how a person behaves (Hewitt & Flett, 1991). Perfectionistic behaviour has been described as a positive factor in people judgement and

achievement (Hamachek, 1978), such as the motivation to strive for excellence and to realise the goals that individuals have. It has been viewed as in born and a basic human nature, that the perfectionistic behaviour is a pervasive neurotic style (Flett, Hewitt, & Dyck, 1989; Pacht, 1984; Weisinger & Lobsenz, 1983). Perfectionism can also lead to negative outcomes, such as the

characterological feelings of failure, guilt, shame, and low self-esteem (Hamachek, 1978; Hollender, 1965; Pacht, 1984; Solomon & Rothblum, 1984; Sorotzkin, 1985). Intense perfectionism can lead to serious forms of psychopathology such as depression and personality disorders (Burns & Beck, 1978), and even to suicidal (Blatt, 1995). One significant tendency regarding perfectionism is to engage in all in or none where only total success or failure are accepted as outcomes (Burns, 1980; Hamachek, 1978; Hollender, 1965; Pacht, 1984).

(19)

19 found out that attractiveness of a superior person is enhanced if he commits a clumsy blunder in various situations (Aronson, Willerman, & Floyd, 1966; Kiesler & Goldberg, 1968; Mettee & Wilkins, 1972). Product wise, negative information that follows the positive information as a whole regarding the product is preferred more than the positive information only containing product (Ein-Gar, Shiv, & Tormala, 2011). Flawed products and people making mistakes could somehow be more preferred in certain conditions. The effect of a mistake or error on an anthropomorphised product is to be tested in the present study.

Understanding the two types of perfectionism are useful in the present research for

anthropomorphism. Since consumers are to make a purchase decision based on their own beliefs, which include the belief towards how other people should be behave perfectly; the present research is to study the effect of the latter belief of perfectionism on the anthropomorphised flawed

products, as such the products could be seen as a person, by a consumer, by laying such people perfection beliefs.

1.2 Conceptual Model

(20)

20 H1: Anthropomorphising products with flaw will get higher acceptance rate, while non

anthropomorphising products with flaw will get lower acceptance rate

Products which are anthropomorphised will make people think they are human (Landwehr, McGill, & Herrmann, 2011). Products which are anthropomorphised would be more accepted depending on various factors, shown by previous studies, such as the ratio of its face and the consumption goals of the consumer (Maeng & Aggarwal, 2017), whether the consumer feels socially excluded (Chen, Wan, & Levy, 2016), and whether the brand is positioned as partner or servant (Kim & Kramer, 2015). This brings the current research to having the second question as whether people’s lay beliefs towards human perfectionism would affect people’s acceptance towards anthropomorphised flawed products, since lay beliefs have been shown to affect judgements and behaviour in a number of domains (Butler, 2000; Mukhopadhyay & Johar, 2005).

It is known that beliefs that individuals hold would also affect and direct judgement and behaviour. Particularly in this study, the belief on the extent of the perfectionism of how other people should behave would affect the effect of anthropomorphised flawed products as the anthropomorphised flawed products are likely to be seen as a human. Hence, the human perfection belief is introduced as a moderating effect.

(21)

21

Figure 1 - Conceptual model of whether anthropomorphism will increase flawed product acceptance

2. Methodology

This research focuses on the acceptance of flawed products, which a dented candle was used to represent the flawed product in either an anthropomorphised or non anthropomorphised form for the study. Such study is to answer the overall research question of whether people’s acceptance towards flawed products would be affected by anthropomorphism as the purpose of the present research.

2.1 Experimental Research Design

The predictions were tested in one study with four conditions. The study followed a 2

(anthropomorphised flawed product vs. non anthropomorphised flawed product) x 2 (human perfection lay belief vs. human imperfection lay belief) between subject design.

(22)

22 name as Candie Cozy, and with a smiley facial expression (Landwehr et al., 2011). Such

anthropomorphism application was in line with prior research on how objects could be more likely to be anthropomorphised by applying human like physical features (Aggarwal & McGill, 2007; Chandler & Schwarz, 2010; Haley & Fessler, 2005; Hur, Koo, & Hofmann, 2015; Jipson & Gelman, 2007; Kim, Chen, & Zhang, 2016; Kim & McGill, 2011; Touré-Tillery & McGill, 2015; Woodward, 1999), as mentioned in the previous part. For the non anthropomorphised flawed product, the image of the same candle with a dent was shown; this time, the third person point of view

description was used for the candle. However, the content of description for the two candles were about the same attributes, which were soft and warm light, pleasant scent; with a dent on it yet working just as well as other candles. The candle with dent was used as a study object because the flawed candle with dent has no functional difference to a perfect candle with no dent. A candle with dent could still full accomplish the goal of being burned as a candle without dent.

Figure 2 - anthropomorphised dented candle Figure 3 - non anthropomorphised dented candle

Hi, my name is Candie Cozy! I can provide soft and warm light for you throughout the cold winter months. I also smell nice! It’s a pity that there some small dents on my body, but I work just as well as my other candle friends.

This is a candle from the Cozy Collection. It can provide you with soft and warm light

throughout the cold winter months. It has a nice scent. There are some small dents on the candle, but it works just as well as other candles.

Manipulation of the lay beliefs was achieved by making reference from the study design of

(23)

23 belief manipulation) and answer the questions that followed. Corresponding to the manipulation of lay beliefs in the current study, participants were to read one of two possible paragraphs,

representing the two lay beliefs (human perfection vs. human imperfection), fully crossed. Each paragraph was of similar length (about 80 words). The human perfection manipulation was an excerpt from Roedell’s (1984) research highlighting the benefits of pursuing perfection in people’s lives; on the other hand, the human imperfection manipulation was an excerpt from Pacht’s (1984) research arguing the benefits of embracing individuals’ imperfections. Since such lay belief

manipulation required relatively stronger English comprehension in order to get trusty response, English speaking participants were employed for the study.

2.2 Data Collection

A Qualtrics survey was set up and distributed through Amazon Mechanical Turk, an online panel dedicated for work which requires human intelligence, which in this case is the completion of survey and data collection. Data was collected from 203 anonymous responses with 99% American

(24)

24

2.3 Questionnaire

Respondents were first asked to read a randomised excerpt for the manipulation of the lay belief (human perfection or human imperfection). A point to note is that respondents were told on the questionnaire to keep the content of the excerpt in mind for the later stage of the questionnaire as some comprehension questions were asked near the end of the questionnaire, in order to keep with the cover story. Respondents were then confronted with a randomised image of either

anthropomorphised or non anthropomorphised candle with dent.

Subsequently, the question set was shown. Firstly, questions regarding product acceptance were asked. Product acceptance can consist of different parameters. In the questionnaire, questions about degree of liking, willingness to purchase, degree of satisfaction, and the quality of the dented candle were asked. Degree of liking has been a prominent indicator to measure the product

(25)

25 scales, for attitude measures in terms of Likert scale in case of the present study, containing either five or seven response categories, are a common practice to acquire for research (Beardon,

Netemeyer, & Mobley, 1993; Peter, 1979; Shaw & Wright, 1967). Research has shown that there are various advantages with 7-point Likert scale compared to 5-point Likert scale. With seven response categories, reliability of the results from the data could be optimised (Colman, Norris, & Preston, 1997; Ghiselli, 1955; Symonds, 1924). Lewis (1993) found that 7-point scale could result in a stronger correlation in terms of results, which made the effects more easily seen. Research has also justified the popularity and use of 7-point scale rather than 5-point scale (Preston & Colman, 2000).

On the next page, manipulation check and control questions were asked for the dented candle image. For instance, whether the respondents thought the candle had humanised terms such as having intentions, being able to experience feelings, and having a personality. In addition, a question was asked to check whether the respondents were able to notice the flaw which was the dent on the candle. Control questions such as the frequency of respondents using candles and their general liking of candles were asked to ensure reliable experimental results. To have an unified scale of the data collected, a 7-point Likert scale was as well employed for these aforementioned questions.

Manipulation check for lay belief followed after. As mentioned, this section contained a few

(26)

26 Finally, demographic questions were asked. They included gender, level of education, age and nationality.

2.4 Validation

Internal validity was secured by collecting random anonymous data with the manipulation through random allocation of the four conditions to make the collected data solid and incorrupt. The same set of questions was distributed for every questionnaire. Manipulation of independent and

moderating variables was consistent across all conditions for all responses. Hence, it could be certain that the independent variable manipulated would be be the cause of the systematic variation in the dependent variable.

External validity was guaranteed by a moderately large sample size per academic standard and repetition of the same random one of the four conditions in order to assure that such research result is valuable for both academicians and marketing managers. The sample size of such study contains respondents from different age, sex and level of education to make the research result

generalisable. Accounting for confounding variables, same set of questions was used across the four conditions to keep other variables which were not relevant to the study as constant as possible.

2.5 Ethics

(27)

27 penalty per standard of ethics institution of the Faculty of Economic of Business, University of Groningen. The study was wholly anonymous. There was no loss privacy revelation at any time of the study.

3. Results

Before carrying out the analysis with the data collected, it is necessary to know whether the measured variables have the actual effect as represented at the study set up and not perceived as others. For instance, for the present study set up, it is important to ensure the flawed and

anthropomorphised products were actually perceived as they were intended to be by the respondents; and the given paragraphs did actually convey the desired lay belief for respective conditions. In order to achieve above, a manipulation check was performed with respect to the manipulation check questions raised in the questionnaire.

3.1 Manipulation Check

As the study set up was testing anthropomorphised flawed item, the first and foremost step is to test whether the anthropomorphism manipulation was effective. A dummy variable was created where 0 indicates no anthropomorphism and 1 indicates anthropomorphism applied on the product. In the questionnaire, there were several questions for the manipulation check for the

(28)

28 whether the three variables are reliable and consistent enough to be grouped together as one. The result showed that the internal consistency and reliability were excellent for the three variables which allowed them to be summed and averaged (α = 0.893). A new variable named

AnthroManCheck was created which summed and averaged the variables of intentions, feeling and personality. An One Way ANOVA was performed against the independent variable of

anthropomorphism and the dependent variable of AnthroManCheck to test whether the overall perceived humanisation effect is effective upon the anthropomorphised item. The result showed that the effect is significant (F(1, 200) = 11.99, p < 0.001) and there was a significant difference between the anthropomorphised condition and the non anthropomorphised condition (MAnthro =

2.03, SD = 1.53; MNon anthro = 1.43, SD = 0.85). It hence proves that the manipulation for

anthropomorphism on the flawed product was valid and successful.

(29)

29 = 1.54) agreeing to the statement of statements of ‘We as human beings should strive for perfection’ compared to the respondents given the condition of imperfection lay belief (MImperfection = 3.86, SD =

1.71). In a similar manner, to test the effectiveness of the paragraph of imperfection lay belief, the dummy variable was computed in the polar opposite way where 0 indicates the paragraph of perfection lay belief by Roedell (1984) and 1 indicates the paragraph of imperfection lay belief by Pacht (1984). An One Way ANOVA was then performed to confirm that the paragraph of

imperfection lay belief is effective. The result showed that the effect is significant (F(1, 200) = 14.73, p < 0.001). There was a significant difference of respondents given the condition of imperfection lay belief (MImperfection = 5.65, SD = 1.26) agreeing to the statement of statements of ‘We as human beings

should embrace imperfection’ compared to the respondents given the condition of perfection lay belief (MPerfection = 4.85, SD = 1.66). Given that all the results are significant for both conditions, the

manipulation for the lay belief was successful.

3.2 Hypothesis Test

It is fundamental to determine in advance the most appropriate method of analysis to test the hypothesis, which is largely driven by the type of data. Given that most of the data collected were with a 7-point Likert scale, the data were mostly interval. Firstly, as mentioned before, the

(30)

30 degree of satisfaction, and the perceived quality. The result showed that the reliability of these four variables was sufficient to allow them to be summed and averaged (α = 0.91). A new variable called OverallAcceptance was computed which summed and averaged the four variables of degree of liking, willingness to purchase, degree of satisfaction, and perceived quality.

Subsequently, since the independent variables for the present study are both nominal with two levels (Anthropomorphised vs. Non anthropomorphised & Perfection belief vs. imperfection belief) and dependent variable is interval, which is two by two, in order to test the separate main effects of the two independent variables, namely the anthropomorphism effect and the lay belief effect, and the interaction effect of them, a Two Way ANOVA was performed against the dependent variable of OverallAcceptance computed earlier. One point to note is that control factors were taken into account as covariates. For the present analysis, gender, education level and age were added as covariates. With these variables being controlled, it allowed the relationship between the

independent variable and dependent variable being tested to be better understood. A result from univariate test was generated. The result showed that the main effect of anthropomorphism was not significant (F(1, 195) = 0.03, p = 0.86). There was no significant difference between

(31)

31 is marginally significant on the acceptance of the flawed products. The relationship stated in H2 is marginally confirmed. The findings, especially regarding the main effect of anthropomorphism, is somehow counterintuitive to a degree with respect to the findings from the Literature Review which leads to the current Conceptual Model, since the study showed that the respondents tend to accept non anthropomorphised flawed product than the anthropomorphised flawed product. The

explanation is to be discussed later in the Discussion section.

Although the main effects were not significant, it would be interesting to check the size of each effect, so that it would be possible to expect which variable, either anthropomorphism or lay belief, has a stronger power within the effect of the interaction term. Looking further into the Pairwise Comparisons, the mean difference of belief (I-J = 0.217, p = 0.32) was bigger than the mean difference of anthropomorphism (I-J = 0.037, p = 0.86) on the dependent variable of

OverallAcceptance. The statistics suggest that the lay belief would have a stronger power compared to anthropomorphism in the interaction effect. In addition, it may suggest the below mean

differences are not significant.

With the marginal significant effect of the interaction between anthropomorphism and belief, the conditions with respect to the different levels in the independent variables showed slight difference on the acceptance of flawed products. Given the perfection belief which conveys ‘We as human beings should strive for perfection’, the acceptance of anthropomorphised flawed product (MPerfectAnthro = 4.18, SD = 1.71) was lower than the non anthropomorphised flawed product

(MPerfectNonAnthro = 4.65, SD = 1.51). Given the imperfection belief which conveys ‘We as human beings

should embrace imperfection’, the acceptance of anthropomorphised flawed product (MImperfectAnthro

= 4.85, SD = 1.35) was higher than the non anthropomorphised flawed product (MImperfectNonAnthro =

(32)

32

Conditions M SD

Perfect Anthro 4.18 1.71 Perfect Non Anthro 4.65 1.51 Imperfect Anthro 4.85 1.35 Imperfect Non Anthro 4.51 1.50

Table 1 - Overall flawed product acceptance in different conditions

Figure 4 - Overall acceptance of flawed products bar chart

In sum, as shown in the results, both anthropomorphism and belief do not have a significant main effect on the acceptance on flawed products. However, the interaction between anthropomorphism and belief showed a marginal significant effect on the acceptance on flawed products. Explanation and comparison of the above scores along both the anthropomorphism and belief dimensions will be further discussed in the Discussion section.

3.3 Additional Factors 3.8 4 4.2 4.4 4.6 4.8 5

Perfect Anthro Perfect Non Anthro Imperfect Anthro Imperfect Non Anthro

(33)

(34)

34 of the flawed product. Therefore, usage frequency and general liking could be an extraneous factor which affect the dependent variable.

Significant results of the control variables together with the results generated earlier may provide insights on what aspect researchers would need to be careful about while conducting similar research. The results of the analysis performed by a univariate analysis with the aforementioned covariates are shown below.

Dependent variable: OverallAcceptance

Control variable Mean square F p value

Age 0.80 0.45 0.50

Gender 3.88 2.19 0.14

Education level 0.06 0.32 0.86

Usage frequency 4.86 2.74 0.10

General liking 83.1 46.9 0.00

Table 2 - Univariate covariate analysis

The results show that there is a significant positive relationship between the general liking of the product and overall acceptance of the flawed product. This may suggest that the general liking of the product will also increase the acceptance of the product when it has flaws. Such finding is consistent to the explanation mentioned just before regarding the general liking of candles being an extraneous factor and needing to be controlled. Since general liking was significant as a covariate effect towards the overall acceptance of flawed products, a further test was performed to see how the size of the covariate effect of general liking differed on different levels. Sub samples were created by splitting the sample into those who generally like candles and those who generally do not like candles. Median split was performed on variable ‘candle_liking’ (MedianCandle_liking = 4), in which those rated

(35)

35 rated above 4 were grouped into sample which generally like candles (N = 151). Separate Univariate covariate analyses were performed with these two samples. The result showed that, for the do not like candles sample, the covariate effect of general liking became insignificant (p = 0.18). However, it made the interaction effect of anthropomorphism and the lay belief significant (p < 0.05). Looking at the Partial Eta Squared which indicates the effect size, its value for general liking was 0.04, which is between small (0.01) and medium (0.06) according to Cohen (1988). For the like candles sample, the covariate effect of general liking was significant (p < 0.001), in which the Partial Eta Squared value was 0.09, which is more than medium. However, the interaction effect of anthropomorphism and the lay belief insignificant (p = 0.37). This result suggested that the interaction effect is less likely to happen in respondents in general liking candles, compared to the respondents in general not liking candles, which is probably because the effect of general liking was so strong to interfere the dependent variable.

Respondents did not quite notice the flaw on the product as represented by the dent on the candle in the study (M = 2.94, SD = 1.80). Most of the respondents stated that they did not notice the flaw at all, while the least respondents answered that they did fully notice the flaw. The number of respondents who did not notice the flaw at all to the number of respondents who did fully notice the flaw went in a descending order.

Conditions M SD

Table 3 - Flaw notice in difference conditions

Flaw notice Descriptive Statistics

Frequency Percentage Cumulative Percentage

(36)

36 2 – Unnoticeable 44 21.8 50.0 3 – Slightly unnoticeable 32 15.8 65.8 4 – Neutral 23 11.4 77.2 5 – Slightly noticeable 22 10.9 88.1 6 – Noticeable 17 8.4 96.5 7 – Absolutely noticeable 7 3.5 100 Total 202 100

Table 4 – Flaw notice Descriptive Statistics

The spread of the answer to notice the flaw (dent) on the candle lied more below 4, meaning that respondents were not quite able to notice the flaw on the candle shown on the picture in the questionnaire. Since, explained above, several variables would be the confounding variables which might influence the independent variable or dependent variable, these variables were therefore run with ANOVA against the flaw notice to check whether these variable would have an effect on the flaw notice. Result showed that there was interaction effect of gender and education level towards the awareness to notice the flaw. An explanation of this finding could be that, consistent with the previous research (Davis, 1983; Hoffman, 1977; Mestre, Samper, Frías, & Tur, 2009), women being more empathetic than men might have increased the tolerance towards imperfections in a way that they do not even consider it as imperfections, especially to a tiny imperfection which was

represented by the dent on the anthropomorphised candle.

Furthermore, during the hypothesis testing, it was found that gender had a significant relationship with the tendency to agree on the perfection belief (F(1, 200) = 8.62, p < 0.005). The results showed that males (MMale = 4.61, SD = 1.45) agreed more on the perfection belief of ‘We as human beings

should strive for perfection’ than female (MFemale = 3.92, SD = 1.86). Correspondingly, gender also

had a significant relationship with the tendency to agree on the imperfection belief (F(1, 200) = 9.62, p < 0.005). The results showed that females (MFemale = 5.61, SD = 1.38) tend to agree more on the

(37)

37 SD = 1.58). With the significant relationship between gender and lay belief, the sample was divided into two sub samples of male (N = 112) and female (N = 90) to see whether the main effects were present within these sub groups. Results showed that, for male group, the main effect of lay beliefs was significant (p < 0.05); however, for female group, there was no significant effect of

anthropomorphism, lay belief or interaction of anthropomorphism and lay belief. It means that the lay belief would significantly influence the acceptance of flawed products within males. The overall flawed product acceptance in difference conditions for males is shown below.

Conditions M SD

Table 5 - Overall flawed product acceptance in different conditions for males

It is seen that with the imperfection belief, the overall acceptance of flawed product did not differ a lot between whether the flawed product was anthropomorphised or not for males. However, with the perfection belief, the difference of overall flawed product acceptance with respect to the anthropomorphism and non anthropomorphism towards the product was large, compared to the results generated from the sample with both male and female respondents. Explanation of such findings is to be discussed in the discussion session.

4. Discussion

(38)

38 anthropomorphism and lay belief while other main effects were found insignificant. An interesting result was found that the main effect of lay belief was significant within male respondents. In addition, the findings suggest that there is room for a deeper understanding of how

anthropomorphised product is perceived by people and how imperfect, mistakes or flaws on products are interpreted by consumers, since the results showed that the flaw on the product are not found noticeable, in which the flaw was most noticed under imperfection belief non

anthropomorphism condition (MImperfectNonAnthro = 3.21, SD = 1.94), while the flaw was least noticed

under the perfection belief anthropomorphism condition (MPerfectAnthro = 2.69, SD = 1.83). In sum,

none of the hypothesis is significantly confirmed.

(39)

39 An explanation of anthropomorphism effect being not significant is that there could be other

concepts which might interfere the acceptance of flawed products for the result, as such that other concept was strong enough to wipe out or influence the effect of the tested variables, for instance, the hedonic or utilitarian product nature. Research suggested that products are usually bought with hedonic or utilitarian goals, hence making the products a hedonic or utilitarian product (Batra & Ahtola, 1991; Mano & Oliver, 1993). A hedonic product is valued by its intrinsically pleasing properties, in which it provides experiential consumption, fun and pleasure, for example, it is not uncommon to see a price premium for hedonic products, given that they are generally more valued by consumers; while a utilitarian product is valued by only its useful function, where it is seen for its primarily and instrumental purposes to satisfy the practical and basic needs, and prices for utilitarian products tend to be more rigid (Hirschman & Holbrook, 1982; Park & Mowen, 2007; Strahilevitz & Myers, 1998). Research also showed that a product could be bought with both hedonic and utilitarian goals (Dhar & Wertenbroch, 2000; Voss, Spangenberg, & Grohmann, 2003). Hedonic and utilitarian being the goals for consumption, part of consumers attitudes towards the products are based on these (Voss et al., 2003). In the present study, the candle could have been seen as either. On one hand, a candle could function as an utilitarian product as it fulfils function tasks only to provide the light in the room when it is dark. On the other hand, a candle could also be served as a hedonic product to provide warm feeling in the room with the soft light and scent, even when there is light in the room. Whether the product is seen as hedonic or utilitarian shall result in the

(40)

40 flawlessly. The utilitarian product with flaws would be doubted by consumers whether that product could be able to fulfil the needs and tasks. A flaw on hedonic products might not bother consumers that much as they are usually bought for pleasure but not on its functionality. This could be the reason why the respondents were not aware of the flaw on the candle as shown in the result. Past research has shown that consumer anthropomorphism tendencies varied across different types of product (Hart, Jones, & Royne, 2013). Based on this finding, the consumer anthropomorphism tendencies were expected to differ between the products perceived as utilitarian and hedonic. Hart (2013) found that when a product was hedonic, anthropomorphism appeals would have an

additional influence on consumer attitude. This may suggest the hedonic or utilitarian product nature might moderate the anthropomorphism effect in the present study. Furthermore, it could be a future research direction to see the interaction between the effect of anthropomorphism and utilitarian or hedonic product nature towards the acceptance of flaw product since utilitarian or hedonic product nature might have an influence towards the toleration of flaws. In case of the utilitarian product with flaws, the acceptance of the flawed product might depend on whether the strength of the anthropomorphism could convince the consumer that the respective tasks could be carried out expectedly, in other words, outweigh the negativity of the flaw. The anthropomorphism effect might have been altered by the product nature.

(41)

41 and equal partner (Aggarwal & McGill, 2011). A servant is expected to create benefits to consumers and work for them; while a partner is to co create the benefits with consumers and work together with them. If the performance does not meet the expectations for servant products, they would be preferred less favourably since servants are seen to be competent in a task given and able to finish it (Dong & Aggarwal, 2016). In the present research, the flaw studied would have influence on the perceived competence which affect the evaluation of the product. In the present study set up, the flawed candle was neither particularly illustrated as a servant nor partner product. If the flawed candle was perceived as a servant product, it could be assumed that the flaw would not fulfil the perceived competence and the candle could be evaluated less favourably compared to if it was perceived a partner product.

Although the manipulation of anthropomorphism on the product in the present study was successful where the respondents found there were more human characteristics in the anthropomorphised object than the non anthropomorphised object, respondents still in general found the humanised characteristics not recognisable enough as shown in the result that the mean score for

anthropomorphised condition lied on not noticeable. This may suggest that anthropomorphism on products could be more sophisticated than merely applying human characteristics to a product to let a person perceive a nonhuman object to human. Few research suggested that anthropomorphism was thought to be strictly a categorical mistake (Fisher, 1991). It argued that anthropomorphism was only the mistaken categorisation of the nonhuman world as the human world. Animism, the

tendency to perceive life in non-living entities which is a form of anthropomorphism has also been seen as a categorical mistake, although such mistakes usually occurred in children who lack enough experience to understand better the differences between self and the world (Piaget, 1929).

(42)

42 emotions (Epley et al., 2007). For the present study set up, further manipulation for

anthropomorphism such as making the anthropomorphised product visuals animated shown on pictures, with human voice over for its description (Guthrie & Guthrie, 1995; Morewedge et al., 2007), would amplify the anthropomorphism manipulation as mentioned as well in the theoretical section. In the future research of similar interest, more cautions are needed for the better way of anthropomorphising a product.

The results showed that there was a marginal significant interaction effect when anthropomorphism was crossed with the lay beliefs of perfection or imperfection on the acceptance on the flawed product for the second hypothesis. Based on the significant result, while under the imperfection lay belief, the acceptance of anthropomorphised flawed products was higher than that of non

(43)

43 of the product character. People are not perfect, yet they learn to accept the imperfections they face in others (Grant, 2008). In this case, respondents accepted the imperfections on the

anthropomorphised products as they are seen as humans, to result in a higher acceptance on the anthropomorphised flawed product. It is interesting to see that the acceptance of non

anthropomorphised flawed products with the person having the imperfection belief was shown higher than the acceptance of anthropomorphised flawed products with the person having the perfection belief, from the study. When a consumer encounters a non anthropomorphised imperfect product with the imperfection belief which expresses imperfections need to be embraced, such imperfection belief which initially projects to human beings was believed to also project to the imperfections on non human objects, the product. Meanwhile, when a consumer encounters an anthropomorphised imperfect product with the perfection belief, which expresses human being should be prefect, the imperfection on the anthropomorphised product which has been seen as human would bother the consumer as the product contradicts the belief he holds. It may suggest why the acceptance of anthropomorphised imperfect products under the perfection belief was lowest among the four conditions.

(44)

44 addition, it was found that gender has a significant relationship with the agreement to the beliefs where males tend to agree more on the belief of ‘We as human beings should strive for perfection’ and females tend to agree more on the belief of ‘We as human beings should embrace

imperfection’. It may be explained by the stereotyped social status of men and women. Men are often seen more masculine in society. They are in control of more power and authority compared to women, such as in work place and household (Hagan, Simpson, & Gillis, 1987; Lester, 2008). In this context, this stereotype may drive men to excel in order to gain or consolidate power. Women are more empathetic than men (Davis, 1983; Hoffman, 1977; Mestre, Samper, Frías, & Tur, 2009). This difference in empathy in women may result in more sensitive to others feeling and hence accept others imperfections which echoes the belief. Also, the main effect of the belief was found

significant within the male group but not the female group. For men having imperfection belief, the acceptance of flawed products was almost the same between anthropomorphism and non

anthropomorphism; however, for men having the perfection belief, the acceptance of flawed products differed a lot between anthropomorphism and non anthropomorphism, in which the acceptance of non anthropomorphised flawed product was higher than the acceptance of

anthropomorphised flawed product, which is consistent to the findings with the sample with both males and females. This may mean that the imperfection belief was not quite convincing to men while men comparatively believe in striving for perfection more, and this translated to the acceptance to flawed products.

4.1 Limitations & Suggestions for Future Research

(45)

45 laboratory study. The disadvantage of such data collection is that that extraneous factors could not be controlled, such as how much time was given to each of the questionnaire given that there was a reading comprehension in the questionnaire and how seriously the questionnaire was completed. Also, the unique characteristic of the online panel Amazon Mechanical Turk used for the present study is that it provides a monetary incentive for the respondents and the respondents are believed to be familiar with filling in research questionnaires, some of them could even be well trained to earn money. One risk is that the respondents would answer questions in a cost effective manner which uses the least time for the questionnaire. Another risk is that respondents might have filled in similar research questionnaires before which they may tend to put favourable answers to optimise the results from data collection. Given that the result of the collected data is inconclusive in the sense that only the interaction effect was proved with marginal significance, it is advised to repeat the same study in a controlled laboratory setting where variables could be controlled to gain more accurate data. In this case, the manipulation for the independent variables could be more flexible as in not only limited to showing only on a computer screen but also showing a real product,

accompanied by visuals and voices, since the laboratory setting could allow face to face interaction and researchers could be able to record any extraneous factor during the study.

(46)

46 differences between culture. These dimensions have been useful in business field to predict

consumer behaviour (De Mooij & Hofstede, 2002). One key dimension to define different culture is the individualism and collectivism, in which whether the members within the culture would like to be independent or interdependent. Since people in collectivism, which interdependent culture value social relation more (Markus & Kitayama, 1991), they would be more likely to anthropomorphised products to fulfil the need of social relations and establish a human and personal relationship with the products. Because of the need for dependency, collectivism members may feel lonely in the absence of social interaction and people were shown to anthropomorphise more when they feel lonely (Epley et al., 2008; Epley et al., 2008; Eyssel & Reich, 2013), since anthropomorphised objects could been perceived as a human and allow people to form and build relationships with them as a way to substitute the social connections with humans in real life (Epley et al., 2008). In the opposite, people from the individualistic culture may not be anthropomorphising as much since their need for social relation or interacting with social agents could have been fulfilled in real life with real humans since they do not need as much social interactions compared to members from collectivism culture and there would be less needs for anthropomorphising. Hence, difference in culture would result in variation in anthropomorphism effect, which would influence the result of the present study.

(47)

47 which was not imperfect as in the study setup. A vast majority of studies used cars as a study object for the anthropomorphism effect (Aggarwal & McGill, 2007; Chandler & Schwarz, 2010; Keaveney, Herrmann, Befurt, & Landwehr, 2012; Landwehr et al., 2011). The appropriate type of product for the present study to find out the relationship of the particular variables is uncertain and needs to be revealed by trial and error.

Concerning what was mentioned in the Discussion section, future research may be interested to see whether the hedonic or utilitarian product nature would influence on anthropomorphism effect, or the product nature of hedonic or utilitarian has to be clearly differentiated to avoid inaccurate results. In addition, the consumer product relationship would be good to look in to as an interaction with anthropomorphism effect on flawed products. The future research is hopefully to provide more insights of anthropomorphism on flawed products in order to tackle the issue of unreasonable wastage in the retail scenario.

4.2 Conclusion

To conclude, the present research was used to analyse whether anthropomorphism and the lay belief towards human beings would have an effect on the acceptance on flawed products compared with no anthropomorphism. Based on the prior research that under certain conditions

anthropomorphism would have a positive influence on certain products, as mentioned in the earlier section, such as a anthropomorphised car will be liked more or less depending on the ratio of its face and the consumption goals of the consumer (Maeng & Aggarwal, 2017), in a similar framework, the present research found out that there was a marginal significant effect of the interaction of

(48)

48 research also aims to provide insights to the marketing of the flawed products which has received little research interest. For instance, based on the present research finding, marketers could try to promote and advertise a flawed product in a way that the product is anthropomorphised it with hinting the message of imperfection should be accepted with human being in the advertisement to encourage consumers to accept the flaw product. Academicians could also further look into the anthropomorphism effects with different kinds of flaws and different kinds of beliefs for the future research. Furthermore, it helps provide a solution to reduce waste in modern society as a support for the Sustainable Development Goals stated by the United Nations.

(49)

49

Reference

Aaker, J. L. (1997). Dimensions of brand personality. Journal of marketing research, 347-356.

Aaker, J., Fournier, S., & Brasel, S. A. (2004). When good brands do bad. Journal of Consumer

research, 31(1), 1-16.

Adler, A. (1956). The neurotic disposition. The individual psychology of Alfred Adler, 239-262.

Aggarwal, P. (2004). The effects of brand relationship norms on consumer attitudes and behavior. Journal of consumer research, 31(1), 87-101.

Aggarwal, P., & McGill, A. L. (2007). Is that car smiling at me? Schema congruity as a basis for evaluating anthropomorphized products. Journal of Consumer Research, 34(4), 468-479.

Aggarwal, P., & McGill, A. L. (2011). When brands seem human, do humans act like brands? Automatic behavioral priming effects of brand anthropomorphism. Journal of Consumer

Research, 39(2), 307-323.

Ahn, H. K., Kim, H. J., & Aggarwal, P. (2014). Helping fellow beings: Anthropomorphized social causes and the role of anticipatory guilt. Psychological Science, 25(1), 224-229.

Ahn, H. I., & Picard, R. W. (2014). Measuring affective-cognitive experience and predicting market success. IEEE Transactions on Affective Computing, 5(2), 173-186.

Arnheim, R. (1969). Visual thinking. Univ of California Press.

Aronson, E., Willerman, B., & Floyd, J. (1966). The effect of a pratfall on increasing interpersonal attractiveness. Psychonomic Science, 4(6), 227-228.

(50)

50 Baird, B., Smallwood, J., Fishman, D. J., Mrazek, M. D., & Schooler, J. W. (2013). Unnoticed

intrusions: Dissociations of meta-consciousness in thought suppression. Consciousness and

cognition, 22(3), 1003-1012.

Ball, A. D., & Tasaki, L. H. (1992). The role and measurement of attachment in consumer behavior. Journal of consumer psychology, 1(2), 155-172.

Barrett, J. L., & Keil, F. C. (1996). Conceptualizing a nonnatural entity: Anthropomorphism in God concepts. Cognitive psychology, 31(3), 219-247.

Batra, R., & Ahtola, O. T. (1991). Measuring the hedonic and utilitarian sources of consumer attitudes. Marketing letters, 2(2), 159-170.

Beardon, W. O., Netemeyer, R. G., & Mobley, M. (1993). Handbook of marketing scales.

Bem, D. J. (1967). Self-perception: An alternative interpretation of cognitive dissonance phenomena. Psychological review, 74(3), 183.

Berger, J., & Heath, C. (2007). Where consumers diverge from others: Identity signaling and product domains. Journal of Consumer Research, 34(2), 121-134.

Bering, J. M. (2006). The cognitive science of souls: Clarifications and extensions of the evolutionary model. Behavioral and brain sciences, 29(5), 486-493.

Blatt, S. J. (1995). The destructiveness of perfectionism: Implications for the treatment of depression. American psychologist, 50(12), 1003.

Bowlby, J. (1980). Attachment and loss: Loss, sadness and depression (Vol. 3).

Brown, S. (2010). Where the wild brands are: Some thoughts on anthropomorphic marketing. The Marketing Review, 10(3), 209-224.

(51)

51 Burns, D. D., & Beck, A. T. (1978). Cognitive behavior modification of mood disorders. Cognitive

behavior therapy: Research and application, 109-134.

Butler, R. (2000). Making judgments about ability: the role of implicit theories of ability in moderating inferences from temporal and social comparison information.

Buzby, J. C., & Hyman, J. (2012). Total and per capita value of food loss in the United States. Food

Policy, 37(5), 561-570.

Bywaters, M., Andrade, J., & Turpin, G. (2004). Intrusive and non‐intrusive memories in a non‐clinical sample: The effects of mood and affect on imagery vividness. Memory, 12(4), 467-478.

Cacioppo, J. T., & Petty, R. E. (1979). Effects of message repetition and position on cognitive response, recall, and persuasion. Journal of personality and Social Psychology, 37(1), 97-109.

Cameron, J. A., Alvarez, J. M., Ruble, D. N., & Fuligni, A. J. (2001). Children's lay theories about ingroups and outgroups: Reconceptualizing research on prejudice. Personality and Social Psychology

Review, 5(2), 118-128.

Campbell, C. (Ed.). (2007). International Product Liability [2007]. Lulu. com.

Castelli, F., Happé, F., Frith, U., & Frith, C. (2000). Movement and mind: a functional imaging study of perception and interpretation of complex intentional movement patterns. Neuroimage, 12(3), 314-325.

Chandler, J., & Schwarz, N. (2010). Use does not wear ragged the fabric of friendship: Thinking of objects as alive makes people less willing to replace them. Journal of Consumer Psychology, 20(2), 138-145.

(52)

52 Chesney, T., & Su, D. K. (2010). The impact of anonymity on weblog credibility. International journal

of human-computer studies, 68(10), 710-718.

Clark, D. A. (Ed.). (2005). Intrusive thoughts in clinical disorders: Theory, research, and treatment. Guilford Press.

Cicatiello, C., Franco, S., Pancino, B., & Blasi, E. (2016). The value of food waste: An exploratory study on retailing. Journal of Retailing and Consumer Services, 30, 96-104.

Cohen, J. (1988). Statistical power analysis for the behavioral sciences . Hilsdale. NJ: Lawrence

Earlbaum Associates, 2.

Colman, A. M., Norris, C. E., & Preston, C. C. (1997). Comparing rating scales of different lengths: Equivalence of scores from 5-point and 7-point scales. Psychological Reports, 80(2), 355-362.

Crisosto, C. H., Crisosto, G. M., & Metheney, P. (2003). Consumer acceptance of ‘Brooks’ and ‘Bing’cherries is mainly dependent on fruit SSC and visual skin color. Postharvest Biology and

Technology, 28(1), 159-167.

Darwin, C., & Prodger, P. (1998). The expression of the emotions in man and animals. Oxford University Press, USA.

Davis, M. H. (1983). Measuring individual differences in empathy: Evidence for a multidimensional approach. Journal of personality and social psychology, 44(1), 113-126.

De Mooij, M., & Hofstede, G. (2002). Convergence and divergence in consumer behavior: implications for international retailing. Journal of retailing, 78(1), 61-69.

Department for Environment, Food & Rural Affairs. (2008). Sustainable Clothing Roadmap.

Dhar, R., & Wertenbroch, K. (2000). Consumer choice between hedonic and utilitarian goods. Journal

When do Consumers accept imperfect products? Anthropomorphism & perfection/imperfection belief as cues

When do Consumers accept imperfect products?

Anthropomorphism &

perfection/imperfection belief as cues

The effect of anthropomorphism and

perfection/imperfection belief on the acceptance on

imperfect products

When do Consumers accept imperfect products?

Anthropomorphism & perfection/imperfection belief as cues

The effect of anthropomorphism and perfection/imperfection belief on the acceptance on

imperfect products

Table of Contents