There is a big controversy concerning the definition of luxury since previous studies provide several contradicting definitions. Many researchers argue that the definition of luxury resides in its unattainability, uniqueness, and exclusivity (Dubois and Paternault, 1995). On the other hand, empirical observations of the market detected that many luxury brands are expanding into offering increasing accessibility to luxury by providing a range of products in several price strata (Catry, 2003; Okonkwo, 2009). The word ” luxury” signifies products or services of upscale standing; nevertheless, this provides no clear understanding because of consumers’ strong involvement, contextual factors and the different values perceived by consumers (Shukla, 2011; Wiedmann et al., 2009). Nueno and Quelch (1998) define luxury brands as expensive purchases that provide consumers with a sense of indulgence. Moreover, luxury brands hold the following features; superior quality, uniqueness, identifiable style, a craftsmanship legacy, high price and worldwide reputation (Nueno and Quelch, 1998). However, consumers purchase luxury products for symbolic reasons to reach their individual or social objectives (Wilcox et al., 2009). Phau and Prendergast (2000) argue that luxury is a subjective concept; therefore consumers’ perceive luxury differently across market segments (Bian and Forsythe, 2011). There has been limited research on the scale measurement of consumers’ perceptions of a brand’s luxury. Vigneron and Johnson (2004) propose the following framework for luxury to determine the factors that make a brand luxurious for consumers (figure x): Figure : Proposed framework for luxurySource: Vigneron and Johnson (2004)This framework (figure x) suggests two dimensions for luxury; personal and non-personal perceptions. Personal perceptions include hedonic and extended self while non-personal perceptions include conspicuousness, uniqueness and quality. This framework provides a better understanding of brand luxury and the factors included in the decision making process of consumers (Vigneron and Johnson, 2004).
Luxury buyer behavior
” Bandwagon” consumption behaviour
The leading utilitarian perspective in economics considers consumption to be bound by the laws of supply and demand ignoring the subject of symbolism (Kastanakis and Balabanis, 2012). From this perspective, consumers’ buying decisions are independent from other factors in the market. Veblen (1899) was the first to note that as wealth prevails within the society, buyer behaviour becomes less about survival needs and more about gaining esteem and the envy of others. This leads to other groups in society engaging with the same consumption behaviour in order to be associated with higher social classes (Veblen, 1899). Bandwagon is defined as the tendency of people to buy a product just because the majority is buying it which consequently keeps increasing the demand (Chaudhuri and Majumdar, 2006; Vigneron and Johnson, 1999). Kastanakis and Balabanis (2012) examined the psychological factors associated with this type of consumption behaviour in luxury. Kastanakis and Balabanis (2012) argue that people’s engagement in this type of consumption behaviour comes from their desire to conform to the groups they aspire to belong to. They developed and empirically confirmed the following conceptual model for bandwagon luxury consumption behaviour. Figure : ” Bandwagon” consumption conceptual modelH: LUXURY-DissertationandwagonConceptualModul. jpgSource: Kastanakis and Balabanis (2012)Their results demonstrated that interdependent self-concept triggers bandwagon consumption. This relationship is mediated by the tendency of a consumer to seek status and their predisposition to social normative influence (Kastanakis and Balabanis, 2012). Therefore, this conceptual model provides insights into the underlying factors that trigger this type of consumption behaviour.
‘ Trading-up’ phenomenon
Luxury brands are starting to take notice of the general population’s interest in luxury and are starting to expand their range by providing lower price forms of their products in order to reach a wider consumer base (Meyers, 2004). Some marketers are sceptical about this approach since they believe it diminishes from the value of luxury products and possibly harms brand image. Others believe that this strategy is viable as long as the brand can preserve its quality and prestige (Meyers, 2004). The luxury market is becoming more and more accessible to the masses, not only through luxury brands expending to a wider range of affordable goods, but through the fact that the middle class is willing to pay, at least occasionally, premium prices for their desired products (Meyers, 2004). The latter trend is known as ” trading up”, people trade up when they buy a big number of their products for low prices in order to save up to afford luxurious ones in other categories (Meyers, 2004). The phenomenon can be described as a trade-off; people are economizing in some areas to have the possibility to indulge in others. Silverstein et al. (2003) identified demand side and supply side factors to the trading-up phenomenon in the USA. Among the demand side factors are the evolution of women roles and family structure, the increase of divorce rates, the growing materialism of customers, and the rising social acceptance of the spending culture. Among the identified supply side factors are; changes in the retailing industry and the growing access to international resources and sophisticated supply chain systems (Silverstein et al., 2003). Silverstein et al. (2003) also identified demographic factors of the phenomenon; high education levels, rise in disposable income, and the shifts in the role of working women and their increasing influence on purchase decisions. This phenomenon indicates that people from all social classes and levels of income are starting to engage in luxury consumption which is providing luxury brands with new lines of opportunity (Silverstein et al., 2003).
Motivations for luxury consumption
Social identity and social comparison theories
Social identity and social comparison are two consumer psychology theories that provide theoretical explanations for luxury consumption (Lertwannawit and Mandhachitara, 2012). Social identity theory refers to consumption behaviour and the acquisition of goods that are aimed at fulfilling identity-centred goals (Kleine et al., 1993). Song and Hattie (1984) state that people across cultures engage in these behaviours that are related to self-concept. Social identity theory perceives humans as complex and multifaceted creatures in relation to their mundane lives (Kleine et al., 1993). According to Hawkes (1994), the development of social identity is influenced by two factors; social forces to conform and internal individualistic forces. Social identity theory explains that individuals strive to acquire products or brands in order to help them express certain facets of their identities, initially directed by social expectations and consumption stereotypes (Lertwannawit and Mandhachitara, 2012). Social comparison theory states that people evaluate themselves through comparing and competing with others (Lertwannawit and Mandhachitara, 2012). This leads to a type of consumption that is intended to satisfy social needs; people buy products in order to impress their peers and gain social status which is referred to as status consumption. In this age of global media, the competition is less about personal reputation, achievements or family but rather about the ability to possess, consume and acquire expensive goods (Phau and Woo, 2008; Phau and Yip, 2008). Comparison between members of society is done through the demonstration of social status and power through the ability to acquire status goods (Roberts & Jones, 2001).
Purchase intention predictors
Limited research has been done to unveil the motivations for luxury consumption by investigating the elements that impact consumers’ purchase intention. The framework (Figure x) proposed by Vigneron and Johnson (2004) suggest two main dimensions for luxury that provide insights into consumption predictors; personal perceptions (which include hedonic and extended self) and interpersonal perceptions (conspicuousness, uniqueness and quality). In addition to measuring the dimensions of a brand’s luxuriousness, this framework provides understanding of consumers’ motivations for luxury consumption (Shukla and Purani, 2011). Wiedmann et al., (2007) extend this framework (suggested by Vigneron and Johnson (2004)) of the luxury value perceived by consumers proposing four dimensions; financial value (price), functional value (uniqueness, quality and usability), individual value (self-identity, materialistic and hedonic) and social value (conspicuousness, prestige). Wiedmann et al., (2007) argue that the underlying motives for luxury consumption surpass national boundaries and originate from the consumer’s situation and the luxury value dimensions. Although fashion luxury is one of the most lucrative and fast growing markets, researchers agree that value dimensions that influence luxury consumption is a very under-researched and misunderstood area ( Tynan et al., 2010; Vigneron and Johnson, 2004; Wiedmann et al., 2007). Actually, customers’ perceptions of the value of luxury brands constitute an imperative aspect influencing luxury buyer behaviour (Tynan et al., 2010; Wiedmann et al., 2007). Luxury brands have more pressure to justify their premium prices since consumers must identify enough value to balance between the product and the price (Shukla and Purani, 2011). Shukla and Purani (2011) conducted a comparative study of luxury value perceptions and its impact on the purchase intention between British (individualist society) and Indian (collectivist society) consumers. The conceptual model constructed and empirically tested involves five dimensions as illustrated in figure x. Figure : Conceptual model for luxury consumption motives #1Full-size image (19 K)Source: Shukla and Purani (2011). The results of the study demonstrated that the dimensions of luxury value perceptions are significant in all cultures; however their degree of influence vary considerably (Shukla and Purani, 2011). According to Shukla and Purani (2011), consumers in Britain use more complex criteria in evaluating the value of a luxury brand than in India (where they use simpler evaluation criteria). The study demonstrated that the perception of Utilitarian/functional value is highly significant for the British customer’s purchase intention. Utilitarian value is related to the function that a product serves as well as its overall characteristics and usefulness (Shukla and Purani, 2011). Zhan and He (2011) conducted a study that investigates luxury consumption motives for the Chinese middle class. Their study was based on three psychological parameters; value consciousness, need for uniqueness, and susceptibility to normative influence (figure x). Figure : Conceptual model for luxury consumption motives #2Full-size image (14 K)Source: Zhan and He (2011)The elements studied in the above study provide insight into the dimensions involved in the prediction of consumers’ purchase intention. Wong and Ahuvia (1998) argue that luxury consumption of consumers in collectivist cultures are influenced by social needs while consumers in individualistic cultures are more motivated by self-expression needs. Bian and Forsythe (2011) conducted a cross cultural study of luxury purchase intention for American and Chinese consumers. As illustrated in figure x, this study is different to the studies presented above (that include only individual characteristics) because it includes both individual characteristics and brand-associated elements (Bian and Forsythe, 2011). Individual characteristics include need for uniqueness and self-monitoring while brand-associated elements include social function attitudes and affective attitude. Figure : conceptual model for luxury consumption motives #3Full-size image (18 K)Source: Bian and Forsythe (2011)Contradicting the view of Wiedmann et al. (2007), Bian and Forsythe (2011) argue that luxury is a subjective concept that varies across cultures. It is worth noting that other studies focused exclusively on brand associated variables in predicting the purchase intention (e. g. Miller and Mills, 2010)
Summary
The literature review presented buyer behaviour… and the different conceptual models for the underlying motivation for luxury consumption. There are individual characteristics and brand associated motives… individual characteristics models have been focused on in the literature since this study is concerned abt that…. Chapter 3Conceptual model and hypotheses developmentCONCEPTUAL MODEL AND HYPOTHESES DEVELOPMENTThis chapter presents the conceptual model developed by this study and the rationale behind it. It also provides definitions for all the constructs of the conceptual model and develops the hypotheses to be tested by the study. Based on the motivations for luxury consumption identified in the literature review, this study develops and empirically tests a conceptual model including four elements; need for uniqueness, susceptibility to normative influence, value consciousness and experiential/hedonic needs (figure x). The relationships between these elements and consumers’ purchase intention are investigated directly without a mediating factor. Given the scale of the study and the time limitations, other factors involved in predicting the purchase intention could not have been included. Figure : Conceptual modelThe study focuses only on individual characteristics ignoring brand associated variables that previous studies explored (e. g. Bian and Forsythe, 2011; Miller and Mills, 2010). In the light of the literature review, some key individual characteristics have been identified. First, there is a strong social dimension that includes self (need for uniqueness) and others (susceptibility to normative influence). Second, it seems important to include an element relating to the personal dimension of luxury that relates to indulgence and experiencing pleasure (experiential/hedonic needs). Thirdly, luxury products are desired for their premium quality and utility which is an element that seems non-negligible (value consciousness).
Need for Uniqueness
The need of uniqueness is defined as ” an individual’s pursuit of differentness relative to others that is achieved through the acquisition, utilization, and disposition of consumer goods for the purpose of developing and enhancing one’s personal and social identity” (Tian et al., 2001, p. 50). It is the desire of being different from others and expressing one’s unique identity. Since one’s material belongings can be considered as extensions of the self (Belk, 1988), acquiring unique items can fulfil this need (Snyder, 1992). According to Snyder and Fromkin (1977), individuals may feel threatened regarding their identities and feel the need to differentiate themselves from others. Unique products are generally hard to obtain, original, scarce, potentially unpopular, or exclusive to a number of people (Tian et al., 2001). Luxury products can be seen as scarce and unique given their exclusivity and premium price (Zhan and He, 2011). Many studies suggested a relationship between need for uniqueness and purchase intention, usually mediated by one or more elements (e. g. Zhan and He, 2011, figure x). In fact, consumers may purchase luxury goods in order to stand out from their peers and express their unique identities.
H1: Need for uniqueness has a positive impact on purchase intention.
Susceptibility to normative influence
Consumer susceptibility to normative influence is defined as ” the need to identify or enhance one’s image with significant others through the acquisition and use of products and brands, and the willingness to conform to the expectations of others regarding purchase decisions” (Bearden et al., 1989, p. 474). This trait indicates individual differences in conforming to societal influences ( Zhan and He, 2011). The impact of susceptibility to normative influence is apparent in different consumer behaviours including the decision process of choosing products (Bearden et al., 1989). Previous studies suggest that susceptibility to normative influence may have an impact on luxury consumption (Zhan and He, 2011). Consumers high on susceptibility to normative influence are more concerned about avoiding negative impressions in public situations (Wooten & Reed, 2004) therefore seek brands that are widely identifiable and approved by others (Zhan and He, 2011). Moreover, the influence of susceptibility to normative influence is greater for behaviours that are visible publicly (Batra et al., 2001). Therefore, luxury brands are attractive to consumers with high susceptibility to normative influence because it helps them establish their status in society, especially using goods that are visible to the public (Zhan and He, 2011).
H2: Susceptibility to normative influence has a positive impact on purchase intention.
Value consciousness
Zeithaml (1988) defined a product’s value as ” the ratio of overall benefit to total cost” (p. 14). Normally, consumers seek products that fulfil all the required needs with the lowest cost (Zhan and He, 2011). However, a value conscious customer is more cautious about the benefit to cost ratio and looks for items that provide the best possible value (Lichtenstein et al., 1990). Luxury products are considered to have premium quality especially that brand name usually works as an indicator and guarantee for quality (Rao & Monroe, 1989). Therefore, luxury goods provide consumers with functional benefits that can be important for value conscious consumers. Although luxury items are considered to have the highest ratios of cost to quality in the market (McKinsey Corporation, 1990), the psychological benefits of consuming these items can still make them extremely wanted (Belk, 1988). Psychological benefits involve social benefits and experiential benefits. First, luxury goods can communicate status and social rank because of their symbolic associations and high cost (Batra et al., 2000; Bearden and Etzel, 1982; Steenkamp et al., 2003). Second, luxury items can provide customers with emotional benefits such as experiencing pleasure, excitement and satisfaction (Sheth et al., 1991; Westbrook and Oliver, 1991). Therefore, all these functional and psychological benefits constitute the overall perceived value of a product which can motivate consumers to purchase luxury goods (Zhan and He, 2011).
H3: Value consciousness has a positive impact on purchase intention.
Experiential/Hedonic Need
Hedonic consumption relates to the area of consumer behaviour associated with the emotional, fictional and sensory facets of a consumer’s experience with products (Hirschman and Holbrook, 1982). Actually, luxury goods can offer these intangible and experiential advantages to their consumers (Dubois and Duquesne, 1993). McCracken (1986) states that in developed societies with high levels of prosperity, people seek products beyond their functional utilities and pursue the symbolic characteristics of items. This type of consumption has first appeared in the west as a societal norm for seeking happiness (Campbell, 2005 and McCracken, 1986). According to Berg (2007), people have been purchasing luxury products for personal pleasure since the Middle Ages in Europe. Therefore, Shukla and Purani (2011) argue that consumers in the UK are likely to show high predisposition toward hedonism when purchasing luxury products.
H4: Experiential/Hedonic need has a positive impact on purchase intention.
Low and high income groups’ motivations for purchasing luxury
As discussed in section 2. 2 (” bandwagon” consumption behaviour), groups that started acquiring luxury goods wished initially to establish their status in society and gain the envy of others, while lower social groups started to engage in the same consumption behaviour in the purpose of associating with those higher social classes (Veblen, 1899). This suggests that different groups in society have different motives for purchasing luxury products. Moreover, the ‘ trading-up’ phenomenon (section 2. 2) suggests that low income groups engage in luxury consumption by buying a big number of their products for low prices to save up in order to indulge in other product categories (Meyers, 2004). In fact, luxury brands took notice of the general population’s interest in luxury and started offering lower price versions of their products (Meyers, 2004). However, at the present there is no research that specifically investigates the differences of motivations between groups of different levels of income. According to Schor (1991), upper-class groups acquire luxury products to establish their high status within a prosperous society, while lower-class groups purchase luxury goods in order to enhance their self-esteem. Therefore, there might be differences in the motivations of low and high income groups.
H5: Low and high income groups have different motivations for purchasing luxury products.
Summary
Chapter 4MethodologyMETHODOLOGYThis study relied on both primary and secondary research. First, the secondary research helped in understanding the conceptual framework of the luxury buyer behaviour. Moreover, it was useful in determining some of the factors that underlie the luxury mass consumption buyer behaviour. The previous research helped in providing conceptual models that have already been tested through empirical studies. This was useful in choosing and forming a conceptual model to be tested by the study. Remove? Add chap intro.
Research design
Research philosophy
According to Saunders, Lewis and Thornhill (2009), it is important that the researcher is aware of his/her overall research philosophy as this holds essential assumptions about the way in which the world is viewed. These assumptions influence the research strategy and the choice of the methods used throughout the research. Moreover, it also affects the way we understand and interpret what is investigated (Saunders, Lewis and Thornhill, 2009). This research reflects the philosophy of positivism since it used existing theories to come up with hypothesis, and tried to test them in order to develop further theory and generalizations that can be then tested by further research (Saunders, Lewis and Thornhill, 2009). Positivism is defined as ‘ working with an observable social reality and that the end product of such research can be law-like generalizations similar to those produced by the physical and natural scientists (Remenyi et al., 1998, cited by Saunders, Lewis and Thornhill, 2009, p. 113)’. Furthermore, in a positivist research, the researcher takes an objective stand and does not affect or gets affected by the subject which resembles how this research was approached (Saunders, Lewis and Thornhill, 2009). Other alternative philosophies could have been adopted such as the interpretivism philosophy, which emphasizes on the way we as human interpret the world around us (Saunders, Lewis and Thornhill, 2009). However, it can be argued that this would have been more appropriate if there were not enough theoretical grounds to build hypotheses and test them. Therefore, since the factors influencing consumers’ purchase intention towards luxury goods have already been identified, tested and retested in different contexts by previous literature, this research was more inclined towards a positivism philosophy.
Research approach
There are two main approaches to research; deductive; where hypotheses are developed from existing theory and a strategy is adopted to test them, and inductive; where data is collected and analyzed and a resulting theory is then developed (Saunders, Lewis and Thornhill, 2009). According to Robson (2002), there are five stages involved in deductive research: Generating a hypothesis from existing theory (such as a relationship between two or more variables). Indicating how the variables are to be measured. Testing the hypothesis. Analyzing the outcome of the test (confirming or refuting the hypothesis). Modifying the theory based on the findings if necessary. The deductive approach is criticized for having a rigid methodology that does not allow for alternative explanations of the subject. On the opposite side, the inductive approach is less structured and allows room for alternative explanations, but it is less concerned about the generalization of the results (Saunders, Lewis and Thornhill, 2009). Thus, both approaches have advantages and disadvantages and the choice of approach is entirely up to the researcher and the nature of the research subject. One of the characteristics of deductive research is that it tries to clarify causal relationships between variables (Saunders, Lewis and Thornhill, 2009). This research is aimed at investigating the relationship between the factors (need for uniqueness, value consciousness, susceptibility to normative influence, and experiential/hedonic needs, income?) and their influence on consumers’ purchase intention using hypotheses drawn from the literature. Therefore, a deductive research approach was adopted.
Research purpose
The research question and objectives tend to determine the research purpose. Saunders, Lewis and Thornhill (2009) suggest three types of research purposes; descriptive, exploratory and explanatory. Exploratory studies are concerned with clarifying ‘ what is happening; to seek new insights; to ask questions and to assess phenomena in a new light’ (Robson, 2002, p. 59). This type of study aims to explore a given environment which makes the research very flexible (Saunders, Lewis and Thornhill, 2009). The objective of descriptive studies is ‘ to portray an accurate profile of persons, events or situations’ (Robson, 2002, p. 59). Hence, the main objective in this type of research is to give a clear picture of a phenomenon. As for explanatory research, it aims to establish causal relationships between two or more variables (Saunders, Lewis and Thornhill, 2009). Given the conceptual model proposed by this research which aims at investigating the relationship between consumers’ personal motivations? and their effect on the purchase intention, the choice of adopting an explanatory research purpose is justifiable.
Quantitative vs. Qualitative
Data collection methods and analysis procedures can be either quantitative, qualitative, or a mixture of both (multiple methods[1]) (Saunders, Lewis and Thornhill, 2009). Qualitative research is usually about in-depth investigations that are case-centered, as each case is treated individually. It usually uses small samples and data is typically gathered through interviews or focus groups (Saunders, Lewis and Thornhill, 2009). In contrast, quantitative research is highly structured, usually uses bigger samples and tests a hypothesis that can be than generalized and compared to other studies (Saunders, Lewis and Thornhill, 2009). Given the deductive nature of the research, its positivist philosophy and explanatory purpose, a quantitative study can be considered more appropriate for this research.
Research strategy
There are many strategies for research, to list a few; experiment, survey, case study and action research. The strategy is dependent on the research objectives, time available for the researcher as well as the philosophical position of the research (Saunders, Lewis and Thornhill, 2009). This research takes a survey strategy which can be justified given its deductive, positivist and quantitative nature and also given the time constraints of the study.
Sampling design
Target population
Setting the right background population is of crucial importance since it should reflect the aims and objectives of the research (Saunders, Lewis and Thornhill, 2009). The target population of this research is luxury goods consumers in the UK, including both females and males with no specification to age or other categorizations. To minimize access issues, the study focused on luxury shoppers in luxury stores and malls and excluded internet luxury shoppers. Therefore, the main characteristics of the target population are: Females and malesLuxury goods shoppersStores shoppers rather than internet shoppersLuxury shoppers in the UK. Ideally, all of the people in the target population could have been reached, but given the time and financial limitations of the study, this was not possible. Therefore, a sample was drawn from the background population.
Sampling frame
According to Saunders, Lewis and Thornhill (2009), a sampling frame is a complete list of all the people in the background population from which the sample is drawn. Given that the sampling frame for this research is unknown as it is extremely hard to determine all luxury buyers in the UK, this does not apply to this study.
Sampling technique
Not having a known sampling frame dictated a non-probability sampling technique. According to Saunders, Lewis and Thornhill (2009), probability sampling in business research using a sample chosen statistically at random may sometimes not be possible because of the lack of a sampling frame, or appropriate for answering the research question. This choice was also supported by the easy administration of this technique and the time limitations of the study. Non-probability sampling provides an alternative technique to select an appropriate sample based on the researcher’s subjective judgement (Saunders, Lewis and Thornhill, 2009). It is important to note that this type of sampling technique produces results that can be limited in their generalizability. However, Saunders, Lewis and Thornhill (2009) argue that results from this technique can still be generalized to the population but not on statistical grounds. A self-selection technique (which is a type of volunteer sampling technique) was used in this research. This technique allows people to decide whether they wish to participate in a study or not (Saunders, Lewis and Thornhill, 2009). The researcher approached shoppers, explained the need for volunteers for the research and collected data from the ones who approved to take part in the study. The questionnaire was distributed in some luxury high streets where consumers of luxury can be directly reached. The choice of location allows reaching consumers of luxury products as these cases fit the profile of the target population characteristics. This provided the research with the advantage of reaching real consumers of luxury rather than the general population.
Sample size
Deciding on an appropriate sample size can be problematic when the sampling frame is unknown. Unlike probability sampling, there are no set rules about the sample size required for a non-probability sampling (Saunders, Lewis and Thornhill, 2009). However, the methods of analysis used in a research can give an indication of the required sample size as some methods require a minimum size in order for the results to be valid (Quinlan, 2011). This research uses multiple regression analysis to determine the causal relationships between the variables. This method requires a minimum sample size that should be used. Tabachnick and Fidell (2007) provided the following formula for calculating the required sample size for multiple regression: N > 50 + 8m (m = number of independent variables) (Pallant, 2010, p. 150). Since the study involves four independent variables (need for uniqueness, value consciousness, susceptibility to normative influence, and experiential/hedonic needs), the required sample size is more than 82 respondents. A total of 200 responses have been collected for the questionnaire which exceeds the requirement and can be considered as a reasonable size of a study at this scale.
Data collection
This research is in the context of the UK fashion luxury consumption, ideally, the sample could have been drawn from different regions of the UK. However, given the limitations of time and resources, the sample population of this study was restricted to shoppers in some luxury high streets in London only. The data was collected in luxury high streets in central London; Oxford Street, Bond Street, and Knightsbridge area. These locations were chosen as they enabled the researcher to directly reach real consumers of luxury products. This approach was fast and effective especially compared to online questionnaires that can take time and may not reach the target population. The primary data collection was done on a span of two weeks; from Monday the 4th of February to Sunday the 17th of February 2013. Once a luxury shopper was identified by the researcher, they were asked whether they wished to participate in the study. The respondent then was explained the purpose of the research and given an estimation of the time needed to complete the questionnaire. When consent was obtained from a respondent to participate in the study, a drop-and-collect method was used to administer the questionnaire. The first paragraph of the survey clearly stated the purpose of the research and explained the commitment to the participant’s privacy protection (APPENDIX 1). Quinlan (2011) argues that administering a questionnaire using a drop-and-collect method can boost the response rate. In fact, Out of 285 people asked to do the survey, 200 approved to take part and the rest refused to participate resulting in a 70% response rate.
Questionnaire and scale development
In order to test the hypotheses, a structured questionnaire was used based on the conceptual model proposed. The questionnaire used scale items that are well established in the literature to measure the constructs of the conceptual model; need for uniqueness, value consciousness, susceptibility to normative influence, and experiential/hedonic needs. The study derives items relating to need for uniqueness from Tian et al. (2001), value consciousness from Lichtenstein et al. (1990), susceptibility to normative influence from Bearden et al. (1989), experiential/hedonic need from Spangenberg, Voss, and Crowley (1997), and purchase intention from Dodds, Monroe, and Grewal’s (1991). Table : Measurement itemsConstructsScale itemsNeed for uniqueness (NFU)When a product I own becomes popular among the general population, I begin to use it less. I often try to avoid products or brands that I know are bought by the general population. Products don’t seem to hold much value for me when they are purchased regularly by everyone. Value consciousness (VC)I am very concerned about low prices, but I am equally concerned about overall product utility. When purchasing a product, I always try to maximize the utility I get for the money I spend. When I buy products, I like to be sure that I am getting my money’s worth. Susceptibility to normative influence (SNI)When buying products, I generally purchase those brands that I think others will approve of. If I want to be like someone, I often try to buy the same brands that they buy. I often identify with other people by purchasing the same products and brands they purchase. Experiential/hedonic Need (EN)It is important to me to own really nice things. Buying luxury products gives me a lot of pleasure. Purchase intention (PI)If I were going to purchase a luxury product, I would consider buying this brand. If I were shopping for a luxury brand, the likelihood I would purchase this luxury brand is high. Source: Tian et al. (2001), Lichtenstein et al. (1990), Bearden et al. (1989), Spangenberg, Voss, and Crowley (1997), Dodds, Monroe, and Grewal’s (1991). The value consciousness scale developed by Lichtenstein et al. (1990) focuses on quality while this study focuses on overall product utility. Therefore, adjustments have been made accordingly to the scale following the practice of Zhan and He (2011). Although the researcher would have preferred to use the complete scale for each of the constructs, the study used only two to three items from each of the established scales in order to keep the questionnaire at a reasonable length. Other studies also used two to three scale items per construct for the same length issue (e. g. Batra et al. 2000; Zhan and He, 2011). The items chosen have been used, tested and retested by several studies (e. g. Zhan and He, 2011; Bian and Forsythe, 2011; Batra et al. 2000). Moreover, these items have obtained support for their psychometric characteristics, reliability and validity (Netemeyer et al. 1991; Batra et al. 2000; Bian and Forsythe 2011). Following the practice of Zhan and He (2011), respondents were asked to indicate their favourite brand in order to use it as a point of reference when answering the purchase intention questions to avoid framing answers with one brand chosen by the researcher (APPENDIX 1). The questionnaire indicated that the questions referred to leather goods (e. g. handbags, wallets, belts. suitcases…). This was done in order to minimize the effects ascribable to specific product categories (Zhan and He, 2011). As illustrated in APPENDIX 1, the questionnaire also included other questions to determine respondents’ characteristics such as the average monthly amount spent on luxury products as well as demographic information (gender, age, occupation and household income). All the measures used Five-point Likert rating scale which is commonly used for collecting data related to opinions and beliefs (Saunders, Lewis and Thornhill, 2009). Respondents were asked about the extent to which they agree or disagree with statements (from ” strongly disagree” to ” strongly agree”). Table : Five-point Likert rating scaleStrongly disagreeDisagreeNeutralAgreeStrongly AgreeSource: Saunders, Lewis and Thornhill, 2009.
Questionnaire pilot
Prior to the questionnaire distribution in London luxury high streets, the questionnaire was assessed in corporation with the research supervisor for feasibility. A pilot study was then conducted to evaluate the clarity of the questions and questionnaire length. The pilot study was done in Kingston University where the questionnaire was distributed to a small number of eight students who buy luxury products. Some changes were made to the questionnaire according to the feedback received.
Analytical approach
This study used SPSS to analyse the results of the questionnaire conducted. After the data was imputed on the software, the analysis was done in three main parts. First, a preliminary analysis was done in order to scan the data for errors, assess the reliability of scales, and the normality of the data. Second, a descriptive statistics analysis was done in order to determine the profile of the respondents. Finally, two statistical tests were used to address the research questions. A multiple regression analysis was used in order to investigate the relationship between the constructs of the conceptual model (need for uniqueness, value consciousness, susceptibility to normative influence, and experiential/hedonic needs) and purchase intention. This intended to test the hypotheses H1, H2, H3 and H4 of the research. An independent-samples t-test followed in order to examine whether there is a difference in high and low income groups’ motivations for purchasing luxury products. This test aims to address H5 of the research.
Multiple regression analysis
This analysis test is used to explore the relationship between a dependent variable and a number of independents variables (Pallant, 2010). Given its complexity, it needs a number of assumptions to be checked in order for the results to be valid (Pallant, 2010). Since H1, H2, H3 and H4 are concerned about testing the relationships between consumers’ motivations and their impact on their purchase intention, the use of this test can be justified given the underlying test assumptions.
Independent samples t-test
Independent samples t-test is used to compare two different groups by comparing the mean score on some continuous variable (Pallant, 2010). This test is one of the parametric tests which assume the data is normally distributed. Since H5 is concerned about comparing two groups (low income and high income), this test is applicable given the assumption that data is normally distributed.
Research ethics
Research ethics refer to questions about whether all the stages included in the process of doing a research are done in a responsible and morale way (Saunders, Lewis and Thornhill, 2009). To ensure the study was conducted in an ethical way, the research stayed objective in all the stages of the process. Moreover, the respondents’ data was treated with great confidentiality to ensure the respect of privacy and anonymity.
Summary
Chapter 5Results and AnalysisRESULTS AND ANALYSISThis chapter…….
Data Profile
Data description
A total of 200 responses to the survey were collected, 67. 5% respondents are female and 32. 5% are males as figure 1 shows. The majority of the respondents were young adults; 43% of the respondents are in the 18-24 age bracket and 35. 5% are in the 25-34 age bracket, figure 2 describes the respondents’ age distribution in more detail. Figure : Gender DistributionSource: APPENDIX 2Figure : Respondents’ Age DistributionSource: APPENDIX 2In terms of Household income, 59. 5% of the respondents (119 respondents) provided information about their income, while 40. 5% refrained from answering this question (APPENDIX 2). As figure 3 shows, 16. 8% have a yearly household income of less than a £10, 000, 23. 5% of the respondents between £10, 000 and £30, 000, 21% between £30, 000 and £50, 000 and 15. 1% have a household income between £100, 000 and £150, 000 (more details are presented in figure 3). Figure : Respondents’ Household IncomeSource: APPENDIX 2As for the average amount spent monthly on luxury products, 34. 8% of the respondents spend less than £200 on luxury products, 28. 3% spend between £200 and £500, 21. 7% spend between £500 and £1000 while only 15. 1% spend more than £1000[2](Figure x). Figure : Average Monthly Spent on Luxury ProductsSource: APPENDIX 2Figure 5 shows that respondents in lower income brackets also purchase luxury products on a monthly basis. Surprisingly, 20% of the respondents who earn less than £10, 000 a year spend on average between £500 and £1000 monthly on luxury goods (APPENDIX 2). Figure : Household Income and Average Monthly Spent on Luxury ProductsSource: APPENDIX 2As figure 6 illustrates, most of the respondents are either in full-time employment (44. 79%) or students (41. 15%). The employment statuses for the rest of the respondents are part-time, housewife, self-employed, retired or unemployed. Figure : Respondents’ Employment StatusSource: APPENDIX 2The most popular luxury leather goods brands among the respondents are; Louis Vuitton, Prada, Mulberry and Chanel.
Data Normality
According to Pallant (2010), it is important to check the normality of the data as that sets the direction of the analysis and the choice of statistical tests. In general, there are two types of statistical tests; parametric and non-parametric tests. Parametric tests are usually used when the data is normally distributed, otherwise non-parametric tests are used instead (Pallant, 2010). There are many ways to check for data normality in SPSS. Several techniques were used but in order to keep the results concise, the researcher chose to only include normal probability plots in the results. Normal probability plots are frequently used since they are good indicators of data normality (Pallant, 2010). APPENDIX 3 shows the normality plots for each of the study constructs; they all show a reasonably straight line which suggests normally distributed data. Therefore, parametric tests were chosen to carry on the analysis.
Data reliability
To ensure data reliability it is essential to measure the internal consistency[3]of scales. Cronbach’s alpha coefficient is one of the most used indicators of internal consistency of scales. Nunnally (1978) recommends a minimum Cronbach’s alpha coefficient of 0. 7 (Pallant, 2010). Table 1 shows that the Cronbach’s alpha coefficients of the constructs lies fairly below 0. 7. Pallant (2010) states that Cronbach’s alpha values are sensitive to the number of items in a scale, with less than 10 items the value of the coefficient tend to be quite low. This is the case in this study as only 3 or 2 items are used to measure the different constructs. In this situation, Pallant (2010) suggests that it is more appropriate to calculate the mean inter-item correlation for items. Briggs and Cheek (1986) recommend values from 0. 2 to 0. 4 for this mean (Pallant, 2010). Table 2 shows that all the inter-item correlation means lie within this range which indicates that the scales used are reliable. To carry on the analysis, items in the different scales where summed and averaged and used to represent the underlying construct, which is how other similar studies measured their constructs (e. g. Zhang and Kim, 2013). Table : Reliability TestConstruct itemsN of itemsCronbach’s AlphaNeed for UniquenessValue ConsciousnessSusceptibility to Normative InfluenceExperiential NeedPurchase Intention333220. 6840. 5930. 6260. 5950. 628Source: APPENDIX 4Table : Mean Inter-item CorrelationConstruct itemsMeanNeed for UniquenessValue ConsciousnessSusceptibility to Normative InfluenceExperiential NeedPurchase Intention0. 4190. 3380. 3590. 4250. 457Source: APPENDIX 4
Consumer motivations affecting the purchase intention of luxury goods
In order to answer the first question of the research and test the hypotheses H1, H2, H3 and H4, multiple regression analysis is used in the study.
Multiple Regression Analysis – Assumptions
In order to investigate if a change in more than one (independent) variable causes a change in a (dependent) variable, multiple regression analysis is used (Saunders, Lewis and Thornhill, 2009). This analysis is complex and requires the data to conform to a number of assumptions in order for the results to be valid (Pallant, 2010). Therefore, these assumptions are checked in the following section.
Sample size
In order for the results to be generalizable, Pallant (2010) suggests that there is a specific requirement for the sample size. Tabachnick and Fidell (2007) provided the following formula for calculating the required sample size for multiple regression: N > 50 + 8m (m = number of independent variables) (Pallant, 2010, p. 150). Since the study involves four independent variables, the required sample size is more than 82 respondents. Given that the sample size used in this research is N= 200, this requirement has been fulfilled.
Multicollinearity check
This assumption refers to the correlation[4]between the variables that make up a model. The correlation between the independent variables should not be high, preferably less than 0. 7 (Pallant, 2010). Table 4 shows that the correlation between the variables is low, meaning that there is no strong correlation between the variables. Table : Correlation checkConstruct ItemsNFUVCSNIENPIPurchase IntentionNeed for UniquenessValue ConsciousnessSusceptibility to Normative InfluenceExperiential Need. 2031. 000. 065. 091. 141-. 005. 0651. 000. 039. 068. 043. 091. 0391. 000. 322. 169. 141. 068. 3221. 0001. 000. 203-. 005. 043. 169Source: APPENDIX 6In addition, multicolleniarity should be checked through the values of Tolerance[5]and VIF[6](Pallant, 2010). The Tolerance value should be more than 0. 10 and the VIF should be less than 10. Table 5 illustrates that the conditions for these indicators are met. Therefore, there is no multicolleniarity between the variables and the assumption holds. Table : Multicolleniarity checkConstruct ItemsToleranceVIFNeed for UniquenessValue ConsciousnessSusceptibility to Normative InfluenceExperiential Need. 975. 992. 894. 8811. 0261. 0081. 1191. 135Source: APPENDIX 6
Normality, outliers, linearity, homoscedasticity and independence of residuals
According to Pallant (2010), one of the ways to check these assumptions is by examining the Normal Probability Plot of the Regression Standardized Residual and the Scatterplot that are outputted in the SPSS regression results. The Normal Probability Plot points should ” lie in a reasonably straight diagonal line from bottom left to top right” (Pallant, 2010, p. 158). As for the Scatterplot, the points should be roughly distributed in a rectangular shape, with some centralization of the points in the middle (Pallant, 2010). Figure 6 and 7 illustrate that the plots fairly follow these requirements; therefore the above assumptions have been fulfilled in order to carry on the analysis. Figure : Normal Probability PlotSource: APPENDIX 6Figure : ScatterplotSource: APPENDIX 6
Multiple regression analysis -Results
The regression analysis is aimed at testing the conceptual model and the hypotheses developed by the study. To test H1, H2, H3 and H4, the constructs (need for uniqueness, value consciousness, susceptibility to normative influence and experiential/hedonic needs) were used as independent variables and purchase intention was used as dependent variable. Table 6 represents the results of the regression analysis. It indicates that 6. 3% of the variance in purchase intention is explained by the independent variables. The test is statistically significant since the Sig. value is below 0. 05 (Pallant, 2010). Table : Regression analysis summaryModelR SquareSig. NFU, VC, SNI, EN0. 0630. 016* Statistical significance is determined at the 0. 05 levelSource: APPENDIX 7Furthermore, as shown in table 7, the analysis revealed that need for uniqueness (β= 0. 186, p= 0. 01) and experiential need (β= 0. 152, p= 0. 045) are significantly related to purchase intention which supports H1 and H4. However, value consciousness and susceptibility to normative influence are not significantly related to purchase intention (p > 0. 05) meaning that H2 and H3 are rejected. Need for uniqueness makes the strongest unique contribution[7]to purchase intention followed by experiential need. This can be seen through the Part correlation coefficients[8], the coefficients values indicate that need for uniqueness explains 3. 35% of the total variance (R square) in consumers’ purchase intention while experiential need has a contribution of 2. 05%. Table : Unique ContributionsModelBetaSig. Part Correlation CoefficientPart Correlation Coefficient SquaredNeed for UniquenessValue ConsciousnessSusceptibility to Normative InfluenceExperiential Need0. 186-0. 026-0. 0220. 1520. 0100. 7130. 7710. 0450. 183-0. 026-0. 0210. 1420. 0334890. 0006760. 0004410. 020164Source: APPENDIX 7In summary, H1 (need for uniqueness has a positive impact on purchase intention) and H4 (experiential/hedonic need has a positive impact on purchase intention) were accepted. However, H2 (susceptibility to normative influence has a positive impact on purchase intention) and H3 (value consciousness has a positive impact on purchase intention) were rejected.
Comparing high and low income groups’ motivations for purchasing luxury goods
In order to test H5, a T-test was performed in order to investigate whether there are differences in high and low income groups’ motivations for purchasing luxury products. The respondents were divided into two groups, a group earning a yearly household income of £52, 000 or less (low income), and a group with a household income more than £52, 000 (high income) as this is the average household income in the UK in 2012 (figure 2). Table 3 indicates that there is a statistically significant difference between high and low income groups’ purchase intention (p <0. 05). The mean score of the high income group is significantly higher than the mean score of the low income group indicating that high income group has a higher intention for purchasing luxury goods. However, there are no statistically significant differences in the purchase motivations of high and low income groups (p > 0. 05). Table : T-test for low and high income groups’ motivations to purchase luxuryConstruct itemsMeanT-test Sig. Low incomeHigh incomeNeed for UniquenessValue ConsciousnessSusceptibility to Normative InfluenceExperiential NeedPurchase Intention3. 123. 702. 584. 153. 823. 153. 572. 744. 224. 280. 8660. 3500. 3520. 6670. 002Source: Therefore, H5 is rejected since there are no statistically significant differences between the mean scores of consumers’ motivations (need for uniqueness, value consciousness, susceptibility to normative influence and experiential need) for low and high income groups.
Summary
Chapter 6
Limitations and recommendations for future research
First of all, the research built upon previous studies in the literature in order to identify consumers’ motivations and construct a conceptual model. There was enough evidence to suggest that these identified motivations (need for uniqueness, value consciousness, susceptibility to normative influence, and experiential/hedonic needs) are good dimensions to predict consumers’ purchase intention. However, other motivations for luxury consumption could exist. Therefore, future research can adopt a different approach to better identify the underlying motivations for luxury consumption. An inductive research approach that does not rely on existing theory could allow for revealing further consumer motivations through qualitative research. Through this approach, a new theory can be built that can then be tested by further research. Moreover, recalling the self-selection sampling technique adopted by the study, a question of generalizability raises. The sample drawn using this technique may not be representative of the whole population and therefore the results cannot be generalized on statistical grounds. However, the results can be generalized to the theory rather than to the population (Saunders, Lewis and Thornhill, 2009). Actually, the limitations of time and resources as well as the lack of a known sample frame dictated this non-probability sampling technique. Also, these limitations did not allow for other regions of the UK to be included which limits the findings to the city of London as consumers’ in other region may or may not have different motivations. Further research can use a probability sampling technique using an identified sample frame which would result in more generalizable findings. Furthermore, further research can focus on different segments of society in order to investigate whether other factors (other than level of income) influence consumers’ motivations to purchase luxury goods.