Interrupting the fast fashion cycle
Can a visible green label attached to apparel made from organic cotton
influence people’s purchase intentions towards organic cotton?
Thesis for obtaining the Master of Science in Marketing
By:
Lize Adriaanse
MSc. Marketing Management
Interrupting the fast fashion cycle
Can a visible green label attached to apparel made from organic cotton
influence people’s purchase intentions towards organic cotton?
-MASTER THESIS-
Author Lize Adriaanse
Address Parkweg 6a, 9725EH Groningen
Phone number 0633873232
E-mail l.m.adriaanse@student.rug.nl
Student number S3195090
Department Faculty of Economics and Businesses
Study Marketing Management (M.Sc.)
First supervisor Dr. Wander Jager
Second supervisor Dr. Laurens M. Sloot
Abstract The fast fashion industry has harmful consequences for the environment. The fibre most
often used in the clothing industry is cotton, which is harvested by using a lot of pesticides and water. In contrast, organic cotton is harvested without the use of pesticides and requires less water. Unfortunately, organic cotton is not the standard yet. Therefore, the purpose of this study is to find a way to increase people’s purchase intention towards clothing made out of organic cotton. This is done by introducing a visible logo attached to apparel by which other people can identify whether a person is wearing organic cotton. Making organic cotton visible provokes the influence of social norms. Moreover, the influence of the different variables of the theory of planned behaviour on people’s purchase intentions towards clothing made out of organic cotton is tested in this study. The results revealed that a recognizable logo attached to clothing made out of organic cotton did not affect people’s purchase intentions. Moreover, only descriptive norms seemed to have an effect on people’s purchase intentions concerning clothing made from organic cotton.
Keywords: sustainability, organic cotton, fashion, theory of planned behaviour. Research theme: sustainability in the fashion industry
Management summary
Fashion is a way to express oneself and therefore it plays an important role in many people’s lives. As a result, people started buying more and more clothing. It is a fact that the fashion industry has harmful consequences for the environment. Not only the amount of clothing which is wasted but also the way fashion is produced. The fibre most often used in the clothing industry is cotton, for what many pesticides are used and needs a lot of water to grow. In contrast, organic cotton is harvested without the use of pesticides and requires less water. Unfortunately, organic cotton is not the standard yet. Therefore, the purpose of this study was to find a way to increase people’s purchase intention towards clothing made out of organic cotton. This was done by introducing a visible green logo attached to apparel by which other people could identify whether a person was wearing organic cotton. Making organic cotton visible provoke the influence of social norms. Moreover, the influence of the different variables of the theory of planned behaviour on people’s purchase intentions towards clothing made out of organic cotton were tested during this study.
An online survey with 158 participants was conducted to investigate participant’s purchase intentions towards clothing made out of organic cotton. First, information about participant’s demographics, clothing shopping behaviour and knowledge about organic cotton was gathered. Second, participants were assigned to one of the two conditions: green label condition versus control group. In the green label condition participants were exposed to a fictitious web page showing a Tommy Hilfiger t-shirt with a green logo instead of the original red logo. Moreover, a textbox informed the participants about the fact that the t-shirt was made out of organic cotton. In the control condition, participants saw a Tommy Hilfiger t-shirt with the regular (red) logo. People’s purchase intentions towards the t-shirts in both conditions were measured. Besides, the influence of descriptive norms and the feeling of perceived behaviour control towards purchasing clothing made out of organic cotton were examined. It is worthwhile to know that not that the majority of the participants were students with a high education level and a low income on a yearly base.
get inspired. Thus, information seekers could be the first group trying out clothing made out of organic cotton with a green label attached to it. Conspicuous consumers form an interesting follower group, as they are sensitive to descriptive norms. If this group sees other people with whom they want to identify with (innovators) wearing clothing with a green logo, conspicuous consumers’ probability to buy organic cotton rises. Overall it is important that people perceive a feeling of behavioural control when buying clothing made out of organic cotton. Hence, the availability of clothing made out of organic cotton with a green logo should be sufficient and the price should not be that high.
Preface
The fashion industry has a huge influence on the environment. The influence is greater than one would have ever thought. Therefore, I got inspired with the topic of fashion, and especially how organic cotton can become the new fashion standard. This master thesis was written as part of the Master Marketing Management at the University of Groningen. I conducted research and wrote my thesis in the period from February to June 2018.
The present thesis deals with an important and complex issue, namely the impact of the fashion industry on the environment. This issue is difficult to visualize, because so many parties are involved with the fashion industry. The goal of this thesis is to, first, examine whether a recognisable logo which is visible when wearing the apparel can increase people’s purchase intentions towards clothing made out of organic cotton. Secondly, I want to find out which drivers of the Theory of Planned Behaviour have an influence on people’s purchase intentions towards clothing made out of organic cotton.
My motivation for conducting this study comes from my trip to Nepal, where I saw the impact of agriculture on the environment. Also, during my trip I noticed that people in Nepal are happy with less belongings compared to the Netherlands. I would love to find a way to make people in the Netherlands less materialistic by reducing their fashion consumption.
I would like to thank my supervisor Dr. Wander Jager for his guidance and constructive feedback during working on my master thesis. I would also like to thank my family and friends for their support and off course all participants who took part in my questionnaire. Without them my research would not have any value.
Table of content
List of figures and tables ... 1
1. Introduction ... 2
2. Literature review ... 6
2.1 Theory of planned behaviour ... 6
2.2 Attitudes toward organic cotton ... 7
2.3 Subjective norms towards organic cotton ... 8
2.3.1 Fashion dynamics ... 9
2.3.2 Dimensions of purchasing fashion ... 9
2.3.3 Customer segments ... 11
2.4 Perceived behavioural control ... 12
2.5 Hypotheses ... 13
2.5.1 Attitudes ... 13
2.5.2 Subjective norms ... 14
2.5.3 Perceived behavioural control ... 15
3. Research methodology ... 17
3.1 Research method ... 17
3.2 Experimental design ... 17
3.2.1 Questionnaire ... 18
3.2.2 Measures ... 19
3.3 Sampling and data collection ... 21
4. Empirical results ... 22
4.1 Sample characteristics ... 22
4.1.1 Fashion factors ... 24
4.2 Effect of control variables ... 25
4.3 Reliability and validity of the measures ... 27
4.3.1 Factor analysis ... 27
4.3.2 Reliability analysis ... 28
4.4 Manipulation test ... 29
4.5 Analysis of variance ... 29
4.6 Regression analysis ... 30
4.6.1 Regression analysis including the different fashion factors ... 31
4.7 Discussion ... 33
4.7.2 The effectiveness of the different drivers of the TPB ... 36
5. Conclusion and Recommendations ... 41
5.1 Conclusion ... 41
5.2 Limitations and Future Research ... 42
5.3 Managerial implications ... 44
References ... 45
Appendices ... 52
Appendix A – Questionnaire ... 52
Appendix B – Fictitious websites ... 57
Appendix C – Demographics Fashion Factors ... 59
Appendix D – Factor analysis ... 61
1
List of figures and tables
Table Title Page
Table 3.2.1 Fashion lifestyle factors 20
Table 3.2.2 Measure items 20
Table 4.1.1 Percentage of different age categories at CBS (2017) and the sample of this study
23
Table 4.1.2 Percentage of different income categories at CBS (2017) and the sample of this study
23
Table 4.3.1 Cronbach’s Alpha scores variables 28
Table 4.3.2 Cronbach’s Alpha scores fashion factors 29 Table 4.6.1 Estimation of results for the multiple regression model 31
Table 4.6.2 Cluster analysis using Ward’s method 32
Figure Title Page
Figure 1 Theory of Planned Behaviour 6
2
1. Introduction
Fashion plays an important role to many people, as it is a way to express oneself (Young et al. 2013). On average people bought 60 per cent more clothes in 2014 in comparison to 2000 (Remy Speelman and Swartz 2016). In 2015, Dutch men spent approximately 73 euro on clothing each month whereas woman had a monthly budget of 66 euro for clothing (Kortingscode.nl 2015). With consumers’ increasing demand for clothing, the pressure on the fashion industry to produce as much clothing as possible has never been this high (Pensupa et al. 2017). In recent years, the textile industry is expanding which increases the concern about the negative impact of the textile industry on the environment. This growth of textile markets is the result of a growing population worldwide and economic and fashion cycles. The fast fashion industry is a culprit in this story, which is responsible for a high level of consumption and waste of textile products. It is a fact that the manufacturing of textile requires a high volume of water (NRDC 2011; Pensupa et al. 2017). Moreover, it is a chemical-intensive process which is harmful for the environment. Subsequently, the production process wastes a lot of water and fibre which cannot be reused (Pensupa et al. 2017).
The harmful consequences of the clothing industry on the environment raise consumers’ interest for sustainable apparel (Carrigan and de Pelsmacker 2009). Sustainability is defined as the ability of humans to make developments more sustainable and by doing so ensure the meets of the present without negatively influencing future generation’s ability to meet their needs in the future (UNWCED 1987). Consumers raising interest for sustainable clothing can be explained by the growing awareness of consumers’ individual impact on the environment. Due to consumer’s interest in sustainable apparel, companies felt the need to insert sustainable practises in their (marketing) strategies. As a consequence, the number of sustainable products in the market place rises (Gam 2011). This trend is confirmed by a visible shift in mind-set of big fashion brands. For example, even H&M who has been haunted by scandals for years is now changing direction (e.g. The Guardian 2017). By 2020, the company aims to get all their cotton from sustainable resources (H&M 2017a). H&M argued that in 2016 already 43 per cent of their total cotton used came from sustainable sources (H&M 2017b). This could mean clothing made out of recycled cotton but also organic cotton plays a big role here.
3 conventional cotton. In 2011, sixteen per cent of all insecticides used for the fibre and global food production were used for the production of conventional cotton (NRDC 2011). Moreover, cotton is a challenging crop to grow, as it is sensitive to changes in temperature and attacked by various insects. As a result of these issues that occur during growing conventional cotton, farmers use genetic modified organisms to increase their harvest. Also, the production of conventional cotton is highly water intensive, which is a problem since most cotton grows in countries with water shortages (NRDC 2011).
In contrast, organic cotton is free from genetically modified organisms’ material, and only methods and materials that have a low impact on the environment are used (Flachs 2014; NRDC 2011). Hence, there are no synthetic agricultural chemicals such as pesticides or fertilizers used during the process of growing organic cotton. This reduces the spread of toxic chemicals that can end up in the air, ground water and food. Moreover, it guarantees the health of the farmers, as they are not exposed to the toxic chemicals which can cause cancer or asthma. Therefore, the production of organic cotton supports healthy ecosystems and biodiversity in regions were the fibre is produced (GOTS 2013; Textile exchange 2014; Textile Exchange 2017). Also, organic cotton often uses less blue water than its conventional brother. Here, blue water can be defined as surface water reservoirs (e.g. via irrigation systems) or water obtained from groundwater. The term basically embraces all freshwater options except from rainwater. Blue water is also used for other purposes, such as laundry or, when filtered, consumption (Textile Exchange 2014). Because cotton often grows in areas where there is a scarcity of fresh water, it is important to reduce the amount of water used (NRDC 2011). In 2012, 2,120m³ blue water was used to produce 1,000 kilogram of conventional cotton fibres. In contrast, 182m³ water from blue water occasions was used to produce 1,000 kilogram of organic cotton fibres. Hence, organic cotton saved 91 per cent blue water over conventional cotton (Textile Exchange 2014). So, when looking at the consequences of conventional cotton on the environment, organic cotton would be a good alternative. This raises the question why organic cotton is not the status quo yet.
4 Previous research showed that consumers have a positive attitude towards organic cotton (Hustvedt and Dickson 2009). The most important determinants to buy organic cotton are people’s belief that organic products are of better quality, health concerns (Hustvedt and Dickson 2009), and environmental concerns (Gam 2011; Hustvedt and Dickson 2009). Besides, social approval affects people’s purchase intentions towards clothing. In order to gain social approval, consumers might buy organic cotton to signal ethical consumer behaviour (Cass 2001). Also, fashion is a domain were social influences are present, which means that consumers are influenced by their peers when buying clothing (Banister and Hogg 2004; Kim, Lee and Hur 2012; Simmel 1957). Important reasons why people do not buy sustainable clothes include a higher price (Gam 2011; Joergens 2006; Meyer 2001), and limited availability (Joergens 2006). Moreover, consumers lack of knowledge about sustainable clothing (Niinimaki 2010), and are therefore not sure about the benefits of it (Meyer 2001). Improvements in availability, consumer knowledge and affordable prices of sustainable clothing can still be made (Hustvedt and Dickson 2009; Joergens 2006; Meyer 2001). Besides, it is important that people recognise clothing made out of organic cotton. Nowadays, consumers can already recognise organic clothing by certified labels like GOTS and the Organic Content Standard (GOTS 2013; Textile exchange 2014). Hence, this is only at the point of purchase, but not visible when wearing the apparel. Making organic cotton more visible when wearing can be an interesting determinant, as here social influences come into play.
This paper has two goals. First it aims to examine if the use of visible labels attached to clothing made out of organic cotton when wearing it can generate an increase in purchase intentions towards clothing made out of organic cotton. Hence, there are two possibilities in which consumers can recognise clothing made out of organic cotton. One possibility is an informational label attached to the product in the store (e.g. GOTS label). Here, consumers have the possibility to look at the informational label and learn about the environmental benefit of clothing made from organic cotton to the environment. Consumers remove this type of label as soon as they start wearing the apparel. Another possibility is a visible label attached to apparel made from organic cotton when wearing it. For example, the normal label of a brand coloured in green so that people are able to recognise that this product is made out of organic cotton. In this way, social influences may play a role as other people in the environment of the consumer can now recognise if someone is wearing organic cotton apparel. For this study, the latter option to make organic cotton more recognisable is used.
6
2. Literature review
In this chapter the different variables of the theoretical framework will be explained. For the purpose of this study we based the theoretical framework on the Theory of Planned Behaviour (TPB) (Ajzen 1991). First, we will explain the TPB. Second, we will deliberate on consumers’ attitudes towards organic cotton. Next, the importance of social influence on consumers purchase intentions towards clothing made out of organic cotton will be discussed. Third, perceived behavioural control will be explained. Finally, based on the literature study different hypothesis will be derived.
2.1 Theory of planned behaviour
Ajzen (1991) developed the TPB which is a cognitive model of human behaviour. TPB is an extended version of the Theory of Reasoned Action, whereby Ajzen (1991) added perceived behaviour control. According to TPB, a person’s behavioural intention is shaped by attitudes towards behaviour, subjective norms and the perceived behavioural control (Ajzen 1991). The model is shown in figure 1.
Figure 1: Theory of Planned Behaviour (Ajzen 1991, p. 182).
7 varies across situations and behaviours. So, it can happen that e.g. attitudes are more of impact on behavioural intentions in one situation, whereas the subjective norm combined with perceived behavioural control might have more impact in another situation. Also, all three aspects can be a predictor of behavioural intentions in some situations (Ajzen 1991).
2.2 Attitudes toward organic cotton
Different aspects of people’s attitudes towards clothing made out of organic cotton can have an influence on people’s purchase intentions. Those aspects include attitudes towards clothing made out of organic cotton, knowledge about apparel made from organic cotton and beliefs about clothing made out of organic cotton. Not much research is done about the attitudes towards organic cotton. Therefore we will sometimes refer to sustainable clothing as this is a broader field of research. Because of the fact that organic cotton is a fibre of which sustainable clothing can be made from, we assume that people hold approximately the same attitudes towards both concepts.
People’s attitudes towards clothing made out of organic cotton are influenced by people’s concern about the environment. Here, people’s environmental concern is defined as a general attitude toward protecting the environment (Ajzen 1989; Minton and Rose 1997). Concerns about the environment arise when people become aware of the harmful consequences of problems in the environment from a certain value orientation (Schultz 2001). So, attitudes toward environmental issues are based on the importance that people place on the biosphere, other people or on the self (Schultz 2001). Previous research on this topic showed that people’s concern about the environment is a determining factor of eco-friendly behaviours (e.g. recycling and buying products based on their environmentally friendly attributes) (Ellen, Wiener and Cobb-Walgren 1991; Minton and Rose 1997; Ohtomo and Hirose 2007). Hence, when people are highly concerned about the environment they might have more positive attitudes towards clothing made out of organic cotton in comparison to people who are less concerned about the environment. However, Kim, Lee and Hur (2012) showed that social norms have a greater influence on purchase intentions towards sustainable clothing than environmental concern.
8 fashionable clothing then apparel produced in a sustainable way. Moreover, Birthwistle and Moore (2007) found that people dispose clothing due to lack of knowledge about the impact of disposal on the environment. People are not aware of the negative effects of the fashion industry on the environment. Meyer (2001) identified the problem that consumers are not sure about the actual benefit of sustainable clothing to the environment. As a consequence, there is a gap in knowledge about sustainable clothing (Meyer 2001). Likewise, Yan, Hyllegard and Blaesi (2012) argued that sustainable clothing is not really understood in terms of materials and production methods used, therefore consumers’ perceptions about the complexity of sustainable clothing is high. Hence, people should be made aware of the impact of sustainable clothing on the environment to increase purchase intentions towards apparel made from organic cotton (Meyer 2001; Yan, Hyllegard and Blaesi 2012).
2.3 Subjective norms towards organic cotton
Many studies in social psychology focused on the effects of social influences on sustainable consumption. Though, only little research has been executed in the context of sustainable fashion, let alone organic cotton. A reason could be that the market for sustainable clothing is rather small and the textile industry has been seen as a cause for environmental damage (Yan et al. 2012). From the academic literature, it does not become clear why there is so little research done about (organic) sustainable clothing. A reason we came up with is the fact that the impact of clothing made out of organic cotton and sustainable clothing in general, is less visible to the environment. For example, there is not much data available about the water or the millilitres of pesticides used to produce one t-shirt. Therefore, the environmental benefit is harder to detect. As a result, the impact of organic cotton has only been studied by a limited amount of researchers (e.g. Hustvedt and Dickson 2009). Due to the fact that not much research has been done about social influences that are at play when buying apparel made from organic cotton, we will sometimes refer to sustainable clothing as this is a broader field of research. Since organic cotton is a fibre of which sustainable clothing can be made, we assume that for these categories the same social forces are present.
9
2.3.1 Fashion dynamics
Clothing is an important aspect of nonverbal behaviour (Rosenfeld and Timothy 1977). Simmel (1957) wrote one of the earliest papers about fashion. According to his study ‘fashion is merely a product of social demands, even though the individual object which it creates or recreates may represent a more or less individual need’ (Simmel 1957, p. 544). This is explained by the fact that often no functional reason can be found why fashion designers designed one piece of clothing, even when it is inconvenient to wear. Normally, the clothing people wear is adapted to their needs (e.g. warm piece of clothing in the winter), but there is no functionally logic in the method by which fashion dominates. Moreover, Simmel states that ‘the absolute indifference of fashion to the material standards of life is well illustrated by the way in which it recommends something appropriate in one instance, something abstruse in another, and something materially and aesthetically quite indifferent in a third. The only motivations with which fashion is concerned are formal social ones’ (Simmel 1957, p. 544). In sum, people wear different types of clothes in different situations as a matter of social conformity. For example, this includes wearing formal clothes when in a business setting whereas one will wear casual clothes among friends and family.
An early definition of fashion is given by Sproles (1979) who defined fashion as temporal behaviour which is adopted by most members of a social group. This behaviour is based on the relevance for time and situation (Sproles 1979). Solomon and Rabolt (2004) define fashion as ‘a style that is accepted by a large group of people at a given time’ (Solomon and Rabolt 2004, p. 9). Although a style does not change over time, it can become fashion at a certain moment in time. Moreover, fashion refers to ‘a form of collective behaviour, or a wave of social conformity’ (Solomon and Rabolt 2004, p. 19).
2.3.2 Dimensions of purchasing fashion
There are different reasons for people to buy fashion. The essential foundations for fashion are individuality and conformity. On the one hand, people use clothing to identify themselves with others (or a group). On the other hand, one uses clothing to distinguish oneself from others (Simmel 1957). Banister and Hogg (2004) agree with this view and add that consumers want to manage their appearance by means of clothes. Moreover, Simmel (1957) states that ‘fashion is a form of imitation and so of social equalization’ (Simmel 1957, p. 541). It distinguishes different time periods from each other and units’ social groups and differentiates them from others. Simmel (1957) also argues that fashion signs reduces freedom, as people do not want to be excluded from their social group.
10 clothing. Hence, consumers reject or accept brands/products based on their symbolic meaning. According to (Levy 1959) an object can be seen as symbolic when individuals focus on meanings away from their tangible and physical characteristics. Thus, clothing can be seen as a social tool, but in order to function as a communication symbol, the symbolic meaning must be socially shared. Also, this meaning should frequently be produced and replicated during interactions with other individuals (Dittmar 1992). So, people often buy clothing for their symbolic meanings rather than its functional purpose (Solomon and Rabolt 2004). For some consumers, branding is an important consideration for buying certain type of clothing. The reason is a brands’ function to show the symbolic value of a piece of clothing (Banister and Hogg 2004). In sum, it becomes clear that fashion plays a big role in many people’s lives and is affected by social influences.
People hold different motives to buy clothing made out of organic cotton. For instance, individuals buy organic cotton because they are concerned about the environment (Gam 2011; Hustvedt and Dickson 2009), health issues, and people believe that organic cotton is of better quality in comparison to conventional cotton (Hustvedt and Dickson 2009). Moreover, people are influenced by the pressure of social forces when buying sustainable clothing (Cass 2001). Niinimaki (2010) found that a person’s values and ethical commitment are strong drivers to purchase sustainable clothes (e.g. recyclable clothes, ecological material, etc.). So, people who hold strong values towards saving the environment and want to behave according to these values are more likely to buy sustainable clothing. Moreover, consumers’ interest in buying sustainable clothing results from fun and wanting to try-out new things (Gam, 2011).
11
2.3.3 Customer segments
In the market, different groups of fashion consumers can be distinguished. McNeil and Moore (2015) conducted an explanatory study with in-depth interviews, and categorized fashion consumers into three different groups: ‘ ‘Self’ consumers, concerned with hedonistic needs, ‘Social’ consumers, concerned with social image and ‘Sacrifice’ consumers who strive to reduce their impact on the world’ (McNeil and Moore 2015, p. 212). These groups have a different view on fashion. The ‘Self’ consumer is focused on fast fashion cycles due to social pressure, and therefore prioritizes speed of new collections and price. They care less about sustainable fashion. On the other hand, the ‘Sacrifice’ consumer really cares about the environment and has a positive attitude towards sustainable clothing. Moreover, this consumer is consistent with its attitude so actually buys sustainable clothes. Nevertheless, this consumer is in conflict between the wish to be fashionable and the reduction of consumption overall. The most interesting segment to target for sustainable clothing is the ‘Social’ consumer. This segment shows an increasing concern for the environment and well-being of people who produced their clothes. They hold positive attitudes towards sustainable clothing, but their attitudes are inconsistent with their behaviour. This is due to the fact that they are aware of social norms, so care about the opinion of others which drives their consumption behaviour (McNeil and Moore 2015). McNeil and Moore (2015) argued that the ‘Social’ consumer segment perceives a number of barriers to buy sustainable fashion. Those barriers include high prices of sustainable fashion; a lack of awareness about sustainable fashion; and social acceptance for sustainable clothing is lacking. To target this segment, McNeil and Moore (2015) suggest making use of mass and social media to create awareness in this market. The study did not clarify the size of the different segments.
Another way to segment the market is based on a distinction between fashion leaders and followers. Shen, Qian and Choi (2017) made this distinction between fashion leaders and fashion followers in the luxury fashion industry based on a database of a large retailer. They argue that both groups have an influence on each other and cause social influences in the clothing market (Shen, Qian and Choi 2017). Their study showed that retailers do provide a better service to fashion leaders when an increase in social influence appears in the market (e.g. a trend). Unfortunately their study does not indicate how many consumers were recorded in the database. Consequently, we cannot identify what percentage of the market consists of fashion leaders or fashion followers.
12 female consumers originating from Korea, Europe and the US. We will give an overview of the different groups based on the study of Ko et al. (2007):
- Segment 1: The ‘information seekers’ are characterized by an information orientation towards shopping. Consumers belonging to this segment search for information about fashion trends even without having an immediate intention to buy something. This group is really innovative and is among the first wearing new fashion trends. Moreover, they care less about the product performance of the apparel they buy. 26.6 per cent of the participants belonged to this group.
- Segment 2: The ‘sensation seekers’ value aesthetic features in clothing and take these aesthetic aspects into account when buying clothing. This segment puts high importance on colour coordination and also believes to have a good taste in selecting apparel. This segment existed out of 29.5 per cent of the sample.
- Segment 3: The ‘utilitarian consumers’ are concerned about the functionality and comfort of clothing they buy. They see clothing as a necessity and do not buy clothing on an impulse. This sample accounted for 25.2 per cent of the sample.
- Segment 4: The ‘conspicuous consumers’ value high quality clothing which are associated with (high) social status. This segment believes in the value of high-priced and/or prestige clothing. 18.5 per cent of the sample belongs to this group.
Due to the fact that this study made a clear distinction between different segments in the market, based on percentages, we will emphasize this segmentation base in our further study.
2.4 Perceived behavioural control
13 find that a pair of jeans made out of organic cotton costs approximately 1.14 euro in comparison to 0.74 euro for a pair of jeans made out of conventional cotton. So the price difference in raw material for one pair of jeans is approximately 40 eurocents. The average price of a pair of jeans in the Netherlands is 82.87 euro, based on Levi’s or a similar brand (Numbeo 2018). On the other hand a pair of jeans made out organic cotton cost approximately 98 euro (Mud Jeans 2018). Hence, the price difference is more than just the costs of raw material.
Moreover, Ellis et al. (2012) studied people’s willingness to pay for clothing made out of organic cotton. The study was conducted in a sample of people between 18 and 21 years old (mainly female) and the result indicated that on average people are willing to pay 25 per cent extra for organic cotton (Ellis et al. 2012). In contrast, other studies showed that although people have a positive attitude towards sustainable clothing, they are not willing to pay a higher price. The reason is that price is often an important factor in buying decisions for clothing (Iwanow, McEachern, and Jeffrey 2005; Meyer 2001). Note that this research was based on sustainable clothing and not on clothing made out of organic cotton in specific. As illustrated, there are some contradictions in the literature about the importance of price when a consumer buys sustainable clothing. Meyer (2001) stated that although a higher price can withhold people from buying sustainable clothing, highlighting other attributes of sustainable clothing can convince people to buy sustainable clothing.
Not only price, but also availability plays a role in people’s purchase intentions towards organic cotton. As mentioned before, an important reason why people do not buy organic cotton is the limited availability (Joergens 2006). When a lot of effort is needed to search for and buy organic cotton, this might decrease people’s purchase intentions towards organic cotton.
2.5 Hypotheses
2.5.1 Attitudes
14 predictors of consumer’s purchase intentions towards clothing made out of organic cotton. So, people’s knowledge about organic cotton will have an effect on their purchase intentions toward apparel made from organic cotton. From this standpoint the following hypothesis is derived:
Hypothesis 1: The more knowledge a consumer has about organic cotton, the stronger the purchase intention towards clothing made out of organic cotton.
2.5.2 Subjective norms
Since fashion consumption behaviour is driven by social forces, we will link the concept of organic cotton to the normative conduct theory of Cialdini, Reno and Kallgren (1990). According to the theory of normative conduct, human behaviour is motivated by two types of social norms: injunctive norms and descriptive norms (Deutsch and Gerard 1955). Cialdini, Reno and Kallgren (1990) define injunctive norms as the appropriate thing to do which help people in guiding their behaviour in certain cultures (e.g. what type of behaviour is accepted / rejected). So, injunctive norms are based on what other people think a person should do/how a person should behave. On the other hand, descriptive norms show people behaviour what others do (e.g. is visible in the environment). The latter is based on ones observations of how people act in a particular situation (Cialdini, Reno and Kallgren 1990).
Kim, Lee and Hur (2012) studied the impact of normative social influence on sustainable clothing. Their results show that both injunctive and descriptive norms have a greater impact on purchase intentions than a person’s concerns about the environment. Moreover, the authors argued that since buying clothes often takes place in a social context, observing other people’s behaviour is of great importance in decision making about sustainable consumption (Kim, Lee and Hur 2012).
Another study showed the effect of injunctive norms on recycling behaviour. The authors found that recycling behaviour increases when the injunctive norm supports particular kind of behaviour and people become aware of it (Ohtomo and Hirose 2007). On the other hand, research about descriptive norms showed that they are a driving force behind people’s behaviour. This is due to the fact that people use information about others actions as a guide regarding proper behaviour (Rivis and Sheeran 2003). In accordance, Cialdini (2003) suggests that descriptive norms are often more effective then injunctive norms when it comes to information processing. As descriptive norms tend to be a better predictor of behaviour (Cialdini 2003; Rivis and Sheeran 2003), it is proposed that:
Hypothesis 2a: The more people observe other people wearing organic cotton (descriptive norm), the stronger their intention to purchase clothing made out of organic cotton.
15 certified label is only visible at the point of purchase, but not while wearing the clothing. Moreover, people use information about others actions as a guide for their own behaviour (Rivis and Sheeran 2003). Making organic cotton more visible when wearing can be an interesting determinant, as here social influences come into play. When a visible label attached to organic cotton can increase the visibility, this might affect consumer responses as consumers want to behave according to the social norm. If consumers see other people wearing organic cotton a new social norm might be formed. Hence:
Hypothesis 2b: The more recognisable as organic, the more people will have the intention to buy clothing made out of organic cotton.
2.5.3 Perceived behavioural control
This determinant refers to whether it is difficult for people to buy organic cotton or not. We will focus on the price and availability of organic cotton. As mentioned before, price is an important determinant in purchase decision of organic cotton (Kim, Lee and Hur 2012). Although one research showed that people are willing to pay 25 per cent more for organic cotton (Ellis et al. 2012), other studies find that consumers are not willing to pay a higher price for organic cotton (Iwanow, McEachern, and Jeffrey 2005; Meyer 2001). As demonstrated before, a pair of jeans made out of organic cotton costs approximately 40 eurocents more in comparison to a pair of jeans made out of conventional cotton. However, when one looks on the website of large fashion retailers, the price differences between organic and conventional cotton are higher than our calculation (e.g. H&M). For that reason, although people can afford organic cotton, it is still questionable if people are willing to pay a premium price (Joergens 2006), because price is an important factor in deciding whether to buy a product made out of organic cotton or not (Kim, Lee and Hur, 2012). Due to the influence of price the following hypothesis arises:
Hypothesis 3a: When the perceived price of organic cotton is high, this will have a negative influence on people’s purchase intentions towards clothing made out of organic cotton.
Another factor of perceived behavioural control is availability. As discussed before, a reason why people do not buy organic cotton is their lack of availability (Joergens 2006). When people have to put a lot of effort in searching for apparel made from organic cotton, this might decrease their purchase intentions. Therefore, it is proposed that:
16 Figure 2 gives an overview of the conceptual model used.
17
3. Research methodology
3.1 Research method
This study was based on primary data, so the data were collected by the researchers itself (Malhotra 2009). The research method used is conclusive, as the purpose of this study was to test specific hypotheses and examine relationships between the different drivers of the TPB. Quantitative data was used to make a statistical analysis of the gathered data and by doing this testing the various hypotheses. Here, descriptive data was used to give insight into people’s drivers to buy clothing made out of organic cotton (attitudes, subjective norm and perceived behavioural control), and causal data was used to conduct a small experiment (green label vs. normal label condition).
3.2 Experimental design
An online questionnaire was developed to test the various hypotheses, as this was the most efficient means to gather relevant information for this study. The advantage of an online questionnaire is that a relatively large amount of data can be gathered in a short period of time. Due to the fact that this study was conducted with a limited amount of time this was the most appropriate way to gather data. Furthermore, this setup offered convenience for participants as they did not have to fill in the questionnaire within a certain timeframe. Appendix A gives an overview of the questionnaire used.
First a pre-study was conducted to find out what clothing brands with a visible label when wearing the apparel were in people’s top of mind. A total number of 15 respondents participated in the pre-study. Respondents were asked to list the names of three different brands with a visible label attached to the apparel when wearing it. The brand most often mentioned was Tommy Hilfiger. Therefore, this brand was used in the further study.
People buy more and more of their clothing online. Moreover, many companies started using the internet as a platform to communicate with their customers, as websites are more than only sales channels. Hence, websites can also be seen as a marketing tool to provide product information and offers possibilities for advertising messages (Balabanis and Reynolds 2001). Therefore, we developed an image of a fictitious website which was used in this study. Two different fictitious product web pages which showed a t-shirt of Tommy Hilfiger were developed. One of the fictitious product web pages showed the brand logo attached to the apparel in its normal condition whereas the other fictitious product web page transformed the logo into a recognisable green logo. So, the online questionnaire existed out of two experimental conditions. Moreover, we created a female and a male version, as the shape of a t-shirt differs across gender.
18 participants were confronted with a Tommy Hilfiger t-shirt made out of organic cotton. Here, the label of the brand was coloured green, so that it became visible for other people in the environment. Because of the fact that the green coloured logo was visible when wearing the apparel, we expected that social influences would have an influence on people’s purchase intentions towards clothing made out of organic cotton. In the control group, participants were confronted with the same t-shirt as in the test condition but here the original colour of the brands’ label was used. Moreover, both conditions contained a small textbox with some information about the product. In the test condition, this textbox informed participants about the fact that the green logo made the product recognisable as made out of organic cotton. On the other hand, the control group saw a textbox with information of the same length as in the test condition. In short, the two fictitious web pages were identical to each other except from the colour of the label and the sentence in the textbox. Besides, the fictitious web pages included a list of product attributes, and a choice menu for colour and style. The content was designed with reference to existing retail websites which sell clothing so that the fictitious web pages looked realistic. The fictitious web pages can be found in appendix B.
People’s indication of their purchase intentions towards the two different conditions served as dependent variable. In addition, the participants were asked to answer questions concerning individual characteristics (attitudes, subjective norms and perceived behavioural control) and demographical backgrounds (gender, age, education level and income). These questions served as independent variables.
3.2.1 Questionnaire
19 offer for either organic cotton or conventional cotton of the same brand, participants purchase intentions towards the apparel were measured. Next, participants were asked to rate their perception on social norms concerning purchasing organic cotton, and perceived behavioural control to buy clothing made out of organic cotton.
Most of the scales used have been validated in previous research and were adjusted to the context of our conceptual model. All items were measured on a 5-point Likert-scale to prevent a number-of-levels effect. The questionnaire was distributed in Dutch.
3.2.2 Measures
20
Fashion lifestyle factors Questions
Conspicuous consumers Accessories must be expensive to have class
I believe that wearing famous brands helps me to be acknowledged by others I believe that expensive clothes are good
Sensation seekers I have good taste in coordinating colours and designs of clothes I have a good eye for selecting clothes
I tend to consider overall coordination of clothes
Utilitarian consumers When selecting clothes I look for their function rather than their colour or design I usually wear clothes that are comfortable
I make purchases only when there is a need
Information seekers I go shopping to get ideas even though I have no intention to buy I like to shop in many different stores
I am the first to try new fashions
I often read fashion-related books and magazines
Table 3.2.1: Fashion lifestyle factors.
Variable in conceptual model Hypothesis Questions
Attitude H1 (Knowledge) - I know where to buy apparel made from organic cotton. - I know pretty much about apparel made from organic cotton. - I do not feel very knowledgeable about apparel made from
organic cotton.
- Among my circle of friends, I’m one of the experts on apparel made from organic clothing.
- Compared to most other people, I really don’t know a lot about apparel made from organic cotton.
- When it comes to apparel made from organic cotton, I really don’t know a lot.
Subjective norms H2a (Descriptive norms)
- Most people at the website have an interest in apparel made from organic cotton.
- I think that many other shoppers are purchasing apparel made from organic cotton.
- I think that many other shoppers are viewing this item on the website.
H2b (Visibility) - This product is recognisable as apparel made from organic cotton.
Perceived Behavioural Control H3a (Price) - I believe that apparel made from organic cotton is more expensive than apparel made from conventional cotton. (strongly agree – strongly disagree)
H3b (Perceived availability)
- For me to buy apparel made from organic cotton will be … (very easy – very difficult)
- If I want to I will easily be able to buy apparel made from organic cotton. (strongly agree – strongly disagree)
- The number of external influences that may prevent me from buying apparel made from organic cotton are … (numerous – non et all)
- How much control do you think you have over your ability to buy apparel made from organic cotton? (complete control – absolutely no control)
Behavioural intention - I would be willing to buy this product for myself.
- I would be willing to recommend this product to other people. - I would be willing to defend this product when others speak
poorly about it.
- I would be willing to buy this product for family or friends.
21
3.3 Sampling and data collection
This research employed individual-level survey data regarding characteristics of consumers and socio-demographic information. Before the start of the data collection we executed a pre-test to get objective feedback about the questionnaire regarding the visibility of the products, the orthography, intelligibility and the estimated time for the survey.
22
4. Empirical results
The purpose of this study is to find out whether a visible label attached to clothing made out of organic cotton can influence people’s purchase intentions towards clothing made out of organic cotton. Moreover, this study intent to find out which drivers of the TPB affect people’s intention to purchase clothing made out of organic cotton. The analysis of the empirical data and a discussion are described in this chapter. To measure the results, SPSS was used. First the dataset was cleaned by deleting incomplete surveys. Moreover, some variables where recoded to better serve the analysis.
4.1 Sample characteristics
23 people lived together with their family (33 %) or partner (30 %). Besides, a smaller proportion of the respondents indicated to live alone (21 %) or with a roommate (17 %).
Age category CBS (2017a) Sample
10 – 20 15.34 % 3.8 % 20 – 30 16.41 % 67.1 % 30 – 40 15.61 % 6.3 % 40 – 50 17.94 % 8.2 % 50 – 60 18.83 % 12.7 % 60 – 70 15.86 % 1.9%
Table 4.1.1: Percentage of different age categories at CBS (2017) and the sample of this study.
Yearly income (on average) CBS (2017b) Sample
€ 0 - €10,000 16.05 % 39.9 % €10,000 - €20,000 26.28 % 21.5 % €20,000 - €30,000 17.99 % 13.9 % €30,000 - €40,000 14.29 % 10.8 % €40,000 - €50,000 9.5 % 5.1% €50,000 - €100,000 13.45 % 7.6% €100,000 - €200,000 2.03 % 1.3 % More than €200.000 0.37 % -
Table 4.1.2: Percentage of different income categories at CBS (2017) and the sample of this study.
On average people buy new clothing 14 times a year and they spend about 81 euro on clothing each month. Most people buy their clothing online as well as in a store (36 %). An interesting fact is that many participants did not know if they had ever bought apparel made from sustainable resources before (37 %). However, almost half of the participants indicated that they sometimes buy apparel made from sustainable resources (49 %). Participants, who indicated that they had ever bought sustainable clothing before, mentioned the following fibres: bamboo, organic cotton, recycled cotton, and second hand clothing. The average knowledge level about clothing made out of organic cotton was quite low (M = 2.45, SD = 0.8524). Likewise, people’s descriptive norms were low (M = 2.31, SD = 0.6718). Further, people’s feeling of perceived behavioural control were moderate (M = 2.89, SD = 0.9183).
24 the normal condition. To check if there existed any differences in socio-demographic properties between the two conditions, we conducted multiple t-tests on the mean of the socio-demographic variables (age, gender, education, type of household and yearly income). None of the t-tests showed a significant result, so the samples in both conditions did not significantly differ from each other (p < 0.05).
4.1.1 Fashion factors
To be able to identify the various fashion factors in practice, the demographics of the four different fashion factors will be described. An overview with figures about the demographics of the different fashion factors can be found in appendix C.
Information seekers
Only a small percentage (8.9 %) of the participants in the sample used can be categorized as information seekers. From this group 29 per cent are men and 71 per cent are female. The majority of this group belongs to the age category of people between 20 and 30 years old (71.43 %). Information seekers are well educated, because the majority of them went to university of applied sciences (35.72 %) or university (42.86 %). Their average yearly income is slightly spread among the different income categories. The majority has a yearly income between €0 till €10,000 (28.57 %), but also €10,000 till €20,000 (21.43 %) and €20,000 till €30,000 (21.43 %) score high. Lastly, most of the information seekers live with their family (35.72 %), with a partner (28.57 %) or alone (28.57 %). Information seekers have the largest monthly budget spend on clothing, namely €119.64. On average, they go shopping about 18 times a year. They buy their clothing online as well as in the store. The majority of them indicate that they sometimes buy sustainable apparel (64.3 %).
Utilitarian consumers
25 sometimes buy sustainable clothing (48 %). Still, another group indicated that they did not know whether they had ever bought sustainable apparel before (40 %).
Sensation seekers
The sensation seekers form a fairly large group (27.8 %), of which 13 per cent are men and 87 per cent are women. A large proportion of this group belongs to the age category of 20 till 30 years old (79.55 %). Most of the sensation seekers have a yearly income between €0 and €10,000 (61.36 %). Because the income level is quite low, we can suggest that this group exists out of students. This group is highly educated; the majority went to university (40.91 %) and university of applied sciences (50 %). They live with their family (34.2 %) or a partner (25 %). This group buys new clothing 13 times a year, and they have an average monthly budget of €83.41 for clothing. They buy their clothing in a psychical store as well as online. Most of them indicated (54.5 %) that they sometimes buy sustainable clothing. So this group is actually aware of sustainable clothing.
Conspicuous consumers
The conspicuous consumers exist out of a large group (31.6 %). Most of them are women (74 %) and a smaller proportion exists out of men (24 %). Their age is also ranging between 20 and 30 years old (66 %). However, there is also quite a large group which is in the age of 40 till 50 years old (12 %), and in the age group of 50 till 60 years old (14 %). Though, the largest group receives a yearly income between €0 and €10,000 (42 %). Another group receives an income between €10,000 and €20,000 on a yearly base (25 %). Moreover, there is a small proportion which has a yearly income between €50,000 and €100,000 (10 %). The conspicuous consumers are well educated, 38 per cent went to university of applied sciences and 42 per cent did university. Most of them live with their family (42 %), but also quite a large group lives with a partner (22 %) or a roommate (20 %). Conspicuous consumers buy new clothing nine times a year and they spend approximately €80.80 on apparel each month. They prefer to go to a real store to buy their clothing. Many of them did not know if they had ever bought sustainable clothing before (44 %). However, some of them indicated that they sometimes buy sustainable clothing (40 %).
4.2 Effect of control variables
26 the monthly budget for clothing on people´s purchase intentions towards clothing made out of organic cotton.
An independent sample t-test with gender and people’s purchase intention towards clothing made out of organic cotton revealed a significant result, t(157) = 2.569, p = 0.011. Hence, there is a statistically significant difference between gender and the mean purchase intention towards clothing made out of organic cotton. Namely, men have a slightly higher purchase intention towards clothing made out of organic cotton (M = 2.90, SD = 0.880) compared to women (M = 2.51, SD = 0.805).
We performed a regression analysis to check whether age (interval variable) has an influence on people’s purchase intentions towards clothing made out of organic cotton. The regression analysis for age was not significant, R2 = 0.008, F(1,156) = 0.009, p = 0.925. Hence, age does not have an influence on people´s purchase intentions towards clothing made out of organic cotton.
The effect of people’s yearly income on their purchase intentions towards clothing made out of organic cotton was measured by means of a one-way ANOVA. The test showed insignificant results,
F(6, 151) = 0.524, p = 0.790. Hence, people’s yearly income does not have an influence on people’s
purchase intentions towards clothing made out of organic cotton.
To analyse whether level of education has an influence on people’s purchase intentions towards clothing made out of organic cotton we performed a one-way ANOVA test as education is a categorical variable. The one-way ANOVA showed no significant difference in people’s purchase intention towards clothing made out of organic cotton, F(3, 154) = 0.368, p = 0.776. Hence, people’s education level does not have an influence on people’s purchase intentions towards organic cotton.
Next we conducted a one-way ANOVA test to examine whether purchase intentions towards clothing made out of organic cotton differ across type of household. There exists no significance in the relationship between type of household and purchase intentions towards clothing made out of organic cotton, F(3, 154) = 0.540, p = 0.656. So, type of household does not affect people’s purchase intentions towards clothing made out of organic cotton.
A regression analysis with frequency of buying new clothes as the independent variable and purchase intention towards clothing made out of organic cotton as the dependent variable showed that there is no significant relationship between those variables, R2 = 0.006, F(1,156) = 0.943, p = 0.333. Hence, frequency of buying new clothing does not influence people’s purchase intentions towards clothing made out of organic cotton.
27
4.3 Reliability and validity of the measures
4.3.1 Factor analysis
The distributed survey was based on validated existing scales. Though, as only a selection of questions from those existing scales were used it is wise to test whether the same factors would emerge when conducting a factor analysis. When checking the correlation coefficients, multicollinearity did not seem to occur. Moreover, the correlation matrix showed no values >0.8, so there was no need to eliminate any questions at this stage. The Kaiser-Meyer-Olkin Measure of Sampling Adequacy indicated a value of 0.778 which exceeded the required value of >0.5. Furthermore, Barlett’s Test of Sphericity was significant (p < 0.000), so correlations could be established. Appendix D includes an overview of the different results of the factor analysis.
Next, we checked the communalities which should be larger than 0.4. Three items in our analysis (“I think that many other shoppers are viewing this item on the website” - “The number of
external influences that may prevent me from buying apparel made from organic cotton are …” - “I believe that apparel made from organic cotton are more expensive than apparel made from conventional cotton”) scored below this threshold (0.289, 0.236 and 0.349), which meant that the
variance in those variables was barely explained by all the extracted factors. Therefore, these items were removed from the dataset, and the factor analysis was performed again.
Now the communalities had acceptable scores, we checked the total variance explained which should exceed 60 per cent. As we had four different variables in our analysis (knowledge, descriptive norm, perceived behavioural control and purchase intention), we checked the total variance explained for four components. The analysis indicated that 71.70 per cent of the variance was explained by four components, which exceeded the threshold of 65 per cent. Also, those components had an initial eigenvalue greater than 1, which is required when forming factors. Likewise, they explained more than 5 per cent (8.969 %) of the variance in the analysis.
To find out whether the variables loaded on the right factor, we paid attention to the different rotated factor loadings. The item loading on one factor should go beyond 0.5. One variable (“Compared to most other people, I really don’t know a lot about apparel made from organic cotton”) had a loading of 0.375 and was therefore removed from the dataset.
In conclusion, the results from the factor analysis did not deviate that much from the scales used in the distributed questionnaire. All scales loaded on the right factors, and the ones that did not were removed from the dataset. Hence, we could continue with the analysis.
Fashion factors
28 Kaiser-Meyer-Olkin Measure of Sampling Adequacy had a value of 0.812 which is larger than the required value of 0.5. Besides, Barlett’s Test of Sphericity showed a significant result (p < 0.000). Furthermore, the communalities seemed good (>0.4), so no items needed to be removed from the dataset. In total, there were four different fashion factors. Those factors explained 64 per cent of the variance, which is acceptable. Also, those four components had an initial eigenvalue greater than 1, and they explained more than 5 per cent (8.37 %) of the variance in the analysis. The rotated factor loadings (>0.5) indicated that every scale loaded on the right factor. Hence, we could continue with the further analysis. An overview of the outcomes of the factor analysis can be found in appendix D.
4.3.2 Reliability analysis
The reliability was checked by using Cronbach’s Alpha. This gave insight in the internal consistency of the different items in a group and showed if the variables used were measuring the same underlying concept. To be accepted, the reliability coefficient should be at least higher than 0.6 (Churchill 1979), and preferably pass 0.70 (Nunnally 1978). Items with scores below 0.60 on the Cronbach’s Alpha test were removed to sustain the validity of the study.
Table 4.3.1 indicates that almost all variables exceeded the threshold of 0.6. Only the score of subjective norms (α = 0.577) is questionable. Because of the fact that the Cronbach’s Alpha score could not be increased by removing an item of the variable we continued the analysis with the low score. However, we have to take this low alpha score in mind when analysing further results.
Note that to reach these scores, one item on the Perceived Behavioural Control scale (“How
much control do you think you have over your ability to buy apparel made from organic cotton?”) was
deleted because this induced to an immense increase of the internal consistency of the scale from 0.681 to 0.834.
Variable in conceptual model Hypothesis Cronbach’s Alpha Mean SD
Attitude H1 (Knowledge) α = 0.861 2.45 0.852 Subjective norms H2a (Descriptive
norms)
α = 0.577 2.31 0.672 Perceived Behavioural Control H3b (Perceived
availability)
α = 0.834 2.90 0.918 Behavioural intention α = 0.884 2.61 0.837
Table 4.3.1: Cronbach’s Alpha scores variables.
Subsequently, the reliability for the different fashion factors was tested by means of Cronbach’s Alpha. Table 4.3.2 gives an overview of the outcomes. As can be seen, all factors exceeded the threshold of 0.6. To increase the Cronbach’s alpha score for factor 2 (sensation seekers) one item (“I
tend to consider overall coordination of clothes”) was deleted to make an increase of internal
29
Factors Cronbach’s Alpha Mean SD
Information seekers α = 0.819 2.45 0.922 Sensation seekers α = 0.730 2.31 0.706 Utilitarian consumers α = 0.608 3.01 0.748 Conspicuous consumers α = 0.620 3.50 0.741
Table 4.3.2: Cronbach’s Alpha scores fashion factors.
4.4 Manipulation test
To check whether the manipulation of the colour of the Tommy Hilfiger logo did work an independent sample t-test was conducted. By doing this we could evaluate whether participants varied in their responses across treatment groups. If the manipulation worked, we could analyse whether or not there existed a difference in people’s purchase intentions towards clothing made out of organic cotton when exposed to the green logo in comparison to the normal (red) logo. After being exposed to the fictitious web page during the questionnaire, respondents were asked to state their agreement on a single Likert-scale item addressing the extent to which the fictitious website showed a t-shirt made out of organic cotton. The results of the t-test indicated that participants’ perceptions of the colour of the logo varied across the two treatment groups, t(158) = 6.32, p > 0.000. Hence, respondents who were exposed to the green logo (M = 2.93, SD = 0.972) reported a higher level of agreement with the item about visibility than those who were exposed to the red logo (M = 2.01, SD = 0.835). However, the question was measured on a Likert-scale with 5 options (1 = strongly disagree, 5 = strongly agree). Participants in the green logo condition had a mean score of 2.93 which is in the middle of the Likert-scale. Therefore, we have to be careful with drawing conclusions about the fact that a green logo attached to the t-shirt made the apparel more recognisable as organic cotton.
4.5 Analysis of variance
30 label attached to apparel made from organic cotton did not lead to an increase in people’s purchase intentions towards the apparel in comparison to conventional cotton. Hence, H2b is not supported.
Moreover, the test of homogeneity of variance was carried out to check if the variances across samples were equally spread. Levene’s test indicated unequal variances (p = 0.418). However, as Malhotra (2010) stated, the two sample sizes do not have to be equal to carry out the analysis.
4.6 Regression analysis
31 A multiple regression was used to test the different effects of the possible drivers of the TPB (attitude, subjective norm, perceived behavioural control) on participants purchase intentions towards clothing made out of organic cotton. Here, people’s purchase intention was the dependent variable and people’s attitude, subjective norm and perceived behavioural control were the independent variables. The results of this regression revealed a significant effect, R2 = 0.070, F(4,153) = 2.88, p = 0.025. Hence, there is a positive relationship between purchase intention and descriptive norms B = 0.323, t(157) = 3.33, p < 0.001. So H2a is supported, the more people observe other people wearing clothing made out of organic cotton, the stronger their intention to purchase clothing made out of organic cotton. Nevertheless, the other variables (knowledge and perceived behavioural control) did not have a significant effect on people’s purchase intentions. Therefore, it can be concluded that H1 and H3 are not supported. Table 4.6.1 presents the hypotheses and the fact whether they are supported or not.
Note that the R-squared (0.070) of the regression is quite close to 0 which is bad, because it indicates that the model explains none of the variability of the response data around its mean. However, in predicting human behaviour it is quite common to have a low R-squared value, because there are many intervenient variables. Although the R-squared value is low, we still have a statistically significant predictor, so we are still able to draw conclusions about how changes in the descriptive norms are associated with changes in people’s purchase intentions towards clothing made out of organic cotton.
Variable Hypothesis Main effect
(significance)
Conclusion
Attitude H1 (+) 0.990 Not supported
Subjective norm (descriptive norms) H2a (+) 0.001 Supported
Subjective norm (visibility) H2b (+) 0.233 Not supported
Perceived behavioural control (price) H3a (-) 0.842 Not supported Perceived behavioural control (availability) H3b (+) 0.901 Not supported
Table 4.6.1: Estimation of results for the multiple regression model.
4.6.1 Regression analysis including the different fashion factors
32 were we indicated one single solution, namely four segments. Wards method for clustering was used, because this method uses squared Euclidean distances to optimize an objective function (Ward 1963). Table 4.6.2 gives an overview of the different clusters. With all respondents appointed to a cluster, we could continue with a multiple regression analysis and check if differences in result were present between clusters.
Factors Frequency Percentage
Sensation seekers 44 27.8
Conspicuous consumers 50 31.6 Utilitarian consumers 50 31.6 Information seekers 14 8.9
Table 4.6.2: Cluster analysis using Ward’s method.
For the regression analysis twelve new variables were created. Specifically, every single cluster was multiplied with all of the independent variables (attitude, subjective norm, perceived behavioural control), which resulted in three new variables for each cluster. These new variables were used in the regression analysis.
